[GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Devs,

I'm Dileepa Jayakody a research student keen on distributed computing
domain, particularly interested in social networking, social identity,
semantic web concepts. I have started studying for a M.Sc by Research at
University of Moratuwa and recently joined LK Domain Registry as a research
assistant. Currently I'm assigned to the 'Digital Identity' research group,
and now doing a literature survey on Digital Identity and identity
verification techniques.  My objective is to develop a methodology to
verify the identity of a person/organization by analyzing profiles, social
activity over different social networking streams, and other verified
data-streams. I'm looking at technologies like RDFa, JSON-LD, FOAF, nosql
databases and other semantic tools which could be helpful to build a
suitable solution. While researching through the web, I came across Apache
Stanbol project and I'm highly interested in learning about Stanbol for my
research project. I'm also fascinated by the GSOC idea on using Stanbol for
federated social web [2]. It would  be great to get ideas on how to
integrate Stanbol to implement a system to verify digital identity of a
person/organization over social networking streams.

To tell a bit more about myself, prior to joining LK Domain Registry as a
research assistant, I was a software engineer for 1.5 years at WSO2 (Pvt)
Ltd [1], the opensource SOA company where I was a member of the carbon team
which is the kernel of the WSO2 Product stack which are OSGi based
products. I have a good development background on java, web services, OSGi
and maven. I also have a good understanding about the opensource
development process and culture as WSO2 follows the apache model for
software development. I have also participated in OSGi Community Event 2012
co-located with EclipseCon 2012 by giving a tech.talk on "Building A
Modular Server Platform with OSGi" [3].

Currently I'm going through Stanbol documentation at [4] and building
Stanbol from trunk source.
Looking forward to your feedback.

Thanks and regards,
Dileepa

[1] http://www.wso2.com
[2] https://issues.apache.org/jira/browse/STANBOL-1009
[3]
http://www.eclipsecon.org/europe2012/sessions/building-server-platform-osgi
[4] http://stanbol.apache.org/docs/trunk/

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi All,

While I continued to browse through the Stanbol JIRA I came across
https://issues.apache.org/jira/browse/STANBOL-1037
in which the usage of entity disambiguation is discussed.  I think
verifying digital identity on web also depends on name ambiguity and the
concept of entity disambiguation is highly applicable in my research idea.
For example there are in-fact 2 'Dileepa Jayakody's in Sri Lanka. One is in
the IT industry and the other in the film industry. So it's essential to
use entity disambiguation techniques to identify the intended Dileepa based
on multiple contexts.

WDYT?

Thanks,
Dileepa


On Mon, Apr 22, 2013 at 1:37 AM, Dileepa Jayakody <dileepajayakody@gmail.com
> wrote:

> Hi Andreas and all,
>
> Thanks a lot for your feedback, I'm definitely open to other's
> ideas/suggestions as well. Please feel free to forward the mail to other
> lists and let me know (I would love to suscribe to other related lists..)
> My idea about the research project on *'verified digital identity'* is as
> below.
>
> We need a verified identity of people we interact with on the internet.
> There maybe so many false profiles, impersonating real people/organization
> and we need to be sure that we are interacting with the real person. For
> example I want to be sure that the person who has  sent me a friend request
> on FB, or an email is the real physical person I meet f2f.
> IMO there are several methods we can use to verify the identity of a
> person. We can use a verified identity provider or we can adapt a
> trust-model based on social networks/social activity of the person to
> verify his/her digital identity based on social proof concepts.
> To build a trust-model based on online social interactions, we can use
> semantic web techniques and I think FOAF, RDFa can be very useful in
> building such a framework. By analyzing the social activity streams, social
> graph and network of the person we can verify the identity.
> As you have mentioned the privacy implications in such a system are
> important. We might need to implement an access control mechanism over the
> data exchanging via the system to protect privacy. Also as you have
> suggested the trust-model can comprise of a reputation calculation module
> based on the feedback a person gets regarding his identity over the web.
>
> So above is my general idea of the research project I'm looking at. I
> think Stanbol, with it's features like Enhancer, Ontology Manager and
> Reasoner is one of the best choices to implement such a framework to verify
> digital identity over the web. Such a system to analyse social data can be
> further improved to implement a perception analysis engine for people. :)
>
> On a related note, I'm currently going through the documentation of
> Stanbol. While groing through I came across a typo mistake and a invailid
> link, wanted to point it out :).
> Created a jira for this at
> https://issues.apache.org/jira/browse/STANBOL-1050
>
> Thanks,
> Dileepa
>
>
> On Sun, Apr 21, 2013 at 10:33 PM, Andreas Kuckartz <A....@ping.de>wrote:
>
>> Hi Dileepa,
>>
>> thanks a lot for your mail.
>>
>> At the moment I only want to provide a quick but brief feedback.
>>
>> What do others think?
>>
>> Your project idea definitely fits the general project idea formulated in
>> STANBOL-1009.
>>
>> Can you tel us a bit more about the use case? Who would use such a
>> feature?
>>
>> Which implications would there be regarding privacy? Would it be
>> possible to use such a feature to improve privacy (for example by
>> providing feedback to a user regaring his or her identifiability)?
>>
>> I would like to forward your mail to two other (public) mailing lists to
>> ask for suggestions. Would that be ok with you?
>>
>> Cheers,
>> Andreas
>> ---
>>
>> Dileepa Jayakody:
>> > Hi Devs,
>> >
>> > I'm Dileepa Jayakody a research student keen on distributed computing
>> > domain, particularly interested in social networking, social identity,
>> > semantic web concepts. I have started studying for a M.Sc by Research at
>> > University of Moratuwa and recently joined LK Domain Registry as a
>> research
>> > assistant. Currently I'm assigned to the 'Digital Identity' research
>> group,
>> > and now doing a literature survey on Digital Identity and identity
>> > verification techniques.  My objective is to develop a methodology to
>> > verify the identity of a person/organization by analyzing profiles,
>> social
>> > activity over different social networking streams, and other verified
>> > data-streams. I'm looking at technologies like RDFa, JSON-LD, FOAF,
>> nosql
>> > databases and other semantic tools which could be helpful to build a
>> > suitable solution. While researching through the web, I came across
>> Apache
>> > Stanbol project and I'm highly interested in learning about Stanbol for
>> my
>> > research project. I'm also fascinated by the GSOC idea on using Stanbol
>> for
>> > federated social web [2]. It would  be great to get ideas on how to
>> > integrate Stanbol to implement a system to verify digital identity of a
>> > person/organization over social networking streams.
>> >
>> > To tell a bit more about myself, prior to joining LK Domain Registry as
>> a
>> > research assistant, I was a software engineer for 1.5 years at WSO2
>> (Pvt)
>> > Ltd [1], the opensource SOA company where I was a member of the carbon
>> team
>> > which is the kernel of the WSO2 Product stack which are OSGi based
>> > products. I have a good development background on java, web services,
>> OSGi
>> > and maven. I also have a good understanding about the opensource
>> > development process and culture as WSO2 follows the apache model for
>> > software development. I have also participated in OSGi Community Event
>> 2012
>> > co-located with EclipseCon 2012 by giving a tech.talk on "Building A
>> > Modular Server Platform with OSGi" [3].
>> >
>> > Currently I'm going through Stanbol documentation at [4] and building
>> > Stanbol from trunk source.
>> > Looking forward to your feedback.
>> >
>> > Thanks and regards,
>> > Dileepa
>> >
>> > [1] http://www.wso2.com
>> > [2] https://issues.apache.org/jira/browse/STANBOL-1009
>> > [3]
>> >
>> http://www.eclipsecon.org/europe2012/sessions/building-server-platform-osgi
>> > [4] http://stanbol.apache.org/docs/trunk/
>> >
>>
>
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@apache.org>]

Hi Dileepa

I think the webID integration code that Reto refers to is
> at [2]?

Yes, with some the code to create browser certificates in
https://github.com/apache/clerezza/tree/trunk/platform.accountcontrolpanel


> Please correct me if I'm wrong. How about we make it working again?
>
That would be great.


> Regarding the ssl http problems maybe we can ask Eclipse Communication
> Framework [3](ECF) guys to collaborate.
>
I was in contact with the jetty guys but then other things came in between.
We ended up using a jetty based OSGi http service here:

https://bitbucket.org/reto/jetty-osgi-httpservice

Basically it's only about making some jetty config option available in the
OSGi evironment.

Cheers,
Reto


>
> On Thu, Apr 25, 2013 at 2:05 PM, Rupert Westenthaler <
> rupert.westenthaler@gmail.com> wrote:
>
> > > @Rupert: With regard to your opinion on using EntityHub as an index
> > rather
> > > than the primary storage for the knowledge base, can you please
> elaborate
> > > more on why do you think so? Maybe Wiki-Link or Freebase [3] is a good
> > > option for such a knowledge base?
> >
> > The Entityhub is designed to play the role of an index & local cache.
> > It was not designed to manage knowledge. Because of that I would
> > recommend to use the knowledge base that does best fit your data
> > (Triple Store, Graph Data Base, SQL, NoSQL ...) and then use the CRUD
> > interface of the Entityhub to create ManagedSites with those data
> > needed for EntiyLinking and Similarity queries.
> >
> > Note that Enhancement Engines can also directly interact with the
> > knowledge base (e.g. if the want to modify the knowledge base such as
> > adding a mention example or creating an unknown Entity). However the
> > Entityhub ensures really fast lookups as required for EntityLinking.
> >
> > best
> > Rupert
> >
> > --
> > | Rupert Westenthaler             rupert.westenthaler@gmail.com
> > | Bodenlehenstraße 11                             ++43-699-11108907
> > | A-5500 Bischofshofen
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

On Thu, May 2, 2013 at 11:35 AM, Reto Bachmann-Gmür <re...@apache.org> wrote:

> On Thu, May 2, 2013 at 7:02 AM, Dileepa Jayakody
> <di...@gmail.com>wrote:
>
> > Hi Reto,
> >
> >
> > On Thu, May 2, 2013 at 10:20 AM, Reto Bachmann-Gmür <reto@wymiwyg.com
> > >wrote:
> >
> > > Hi Dileepa,
> > >
> > > I think this makes sense. Similar to self registered users (previous
> > mail)
> > > a user that comes with WebId or OpenId would get an account and be
> > assigned
> > > a Role (roaming-user role). Additionally to that there could be a
> > mechanism
> > > assigning permissions based on social-network distance.
> > >
> > > A first step might be to add an optional URI field in the user manager,
> > > this URI would be the WebID/OpenID the user can use as an optional
> log-in
> > > mechanism.
> > >
> > > +1, this will be a good improvement in the user management module.
> > Can you please elaborate more regarding the 'mechanism to assign
> > permissions based on social-network distance'?
> > I think we can calculate the social-network distance via WebID, but not
> > sure if it can be done with OpenID. Would like to know more about your
> > idea.It
>
>
> It's sure easier with WebID. But as foaf contains a property to point to
> the OpenID (http://xmlns.com/foaf/spec/#term_openid) it should also be
> possible for OpenID. It would require crawling the social network of the
> isntance owner to a certain distance and see if the person loging in is
> part of it or not.
>
> Great, thanks for the pointer Reto.

> Cheers,
> reto
>
> Thanks,
> > Dileepa
> >
> >
> > > Cheers,
> > > Reto
> > >
> > >
> > > On Wed, May 1, 2013 at 12:13 AM, Dileepa Jayakody <
> > > dileepajayakody@gmail.com
> > > > wrote:
> > >
> > > > Hi Reto et al,
> > > >
> > > > I also, took a look at the user-management feature on the felix
> > console.
> > > > Looks cool.
> > > > I think it can be further improved by integrating OpenID/WebID so
> that
> > > even
> > > > users who are not registered in the Stanbol system can benefit from
> it
> > > > (this is in addition to the WebID integration benefits discussed
> > > previously
> > > > :) ).
> > > > WDYT?
> > > >
> > > > Thanks,
> > > > Dileepa
> > > >
> > > >
> > > > On Wed, May 1, 2013 at 2:09 AM, Michael Luggen <
> > michael.luggen@gmail.com
> > > > >wrote:
> > > >
> > > > > I did check out the usermanagement, looks quite nice,
> > > > >
> > > > > Is there also an web service interface to let self-register new
> users
> > > and
> > > > > let them choose a password from a web application standpoint view?
> > > > >
> > > > > Thanks for inputs,
> > > > > Michael
> > > > >
> > > > >
> > > > > On 26.04.2013, at 09:25, Reto Bachmann-Gmür <re...@wymiwyg.com>
> > wrote:
> > > > >
> > > > > > On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <
> > > dileepajayakody@gmail.com
> > > > >
> > > > > > wrote:
> > > > > >
> > > > > >> 2. WebID based authentication and userstore :Why doesn't Stanbol
> > > have
> > > > a
> > > > > >> user-management module at the moment ? I think a user-store is a
> > > valid
> > > > > >> requirement for any CMS.
> > > > > >
> > > > > > You find some initial Stanbol usermanagement on the felix Web
> > > console.
> > > > > >
> > > > > > Cheers,
> > > > > > Reto
> > > > >
> > > > >
> > > >
> > >
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@apache.org>]

On Thu, May 2, 2013 at 7:02 AM, Dileepa Jayakody
<di...@gmail.com>wrote:

> Hi Reto,
>
>
> On Thu, May 2, 2013 at 10:20 AM, Reto Bachmann-Gmür <reto@wymiwyg.com
> >wrote:
>
> > Hi Dileepa,
> >
> > I think this makes sense. Similar to self registered users (previous
> mail)
> > a user that comes with WebId or OpenId would get an account and be
> assigned
> > a Role (roaming-user role). Additionally to that there could be a
> mechanism
> > assigning permissions based on social-network distance.
> >
> > A first step might be to add an optional URI field in the user manager,
> > this URI would be the WebID/OpenID the user can use as an optional log-in
> > mechanism.
> >
> > +1, this will be a good improvement in the user management module.
> Can you please elaborate more regarding the 'mechanism to assign
> permissions based on social-network distance'?
> I think we can calculate the social-network distance via WebID, but not
> sure if it can be done with OpenID. Would like to know more about your
> idea.It


It's sure easier with WebID. But as foaf contains a property to point to
the OpenID (http://xmlns.com/foaf/spec/#term_openid) it should also be
possible for OpenID. It would require crawling the social network of the
isntance owner to a certain distance and see if the person loging in is
part of it or not.

Cheers,
reto

Thanks,
> Dileepa
>
>
> > Cheers,
> > Reto
> >
> >
> > On Wed, May 1, 2013 at 12:13 AM, Dileepa Jayakody <
> > dileepajayakody@gmail.com
> > > wrote:
> >
> > > Hi Reto et al,
> > >
> > > I also, took a look at the user-management feature on the felix
> console.
> > > Looks cool.
> > > I think it can be further improved by integrating OpenID/WebID so that
> > even
> > > users who are not registered in the Stanbol system can benefit from it
> > > (this is in addition to the WebID integration benefits discussed
> > previously
> > > :) ).
> > > WDYT?
> > >
> > > Thanks,
> > > Dileepa
> > >
> > >
> > > On Wed, May 1, 2013 at 2:09 AM, Michael Luggen <
> michael.luggen@gmail.com
> > > >wrote:
> > >
> > > > I did check out the usermanagement, looks quite nice,
> > > >
> > > > Is there also an web service interface to let self-register new users
> > and
> > > > let them choose a password from a web application standpoint view?
> > > >
> > > > Thanks for inputs,
> > > > Michael
> > > >
> > > >
> > > > On 26.04.2013, at 09:25, Reto Bachmann-Gmür <re...@wymiwyg.com>
> wrote:
> > > >
> > > > > On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <
> > dileepajayakody@gmail.com
> > > >
> > > > > wrote:
> > > > >
> > > > >> 2. WebID based authentication and userstore :Why doesn't Stanbol
> > have
> > > a
> > > > >> user-management module at the moment ? I think a user-store is a
> > valid
> > > > >> requirement for any CMS.
> > > > >
> > > > > You find some initial Stanbol usermanagement on the felix Web
> > console.
> > > > >
> > > > > Cheers,
> > > > > Reto
> > > >
> > > >
> > >
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Reto,


On Thu, May 2, 2013 at 10:20 AM, Reto Bachmann-Gmür <re...@wymiwyg.com>wrote:

> Hi Dileepa,
>
> I think this makes sense. Similar to self registered users (previous mail)
> a user that comes with WebId or OpenId would get an account and be assigned
> a Role (roaming-user role). Additionally to that there could be a mechanism
> assigning permissions based on social-network distance.
>
> A first step might be to add an optional URI field in the user manager,
> this URI would be the WebID/OpenID the user can use as an optional log-in
> mechanism.
>
> +1, this will be a good improvement in the user management module.
Can you please elaborate more regarding the 'mechanism to assign
permissions based on social-network distance'?
I think we can calculate the social-network distance via WebID, but not
sure if it can be done with OpenID. Would like to know more about your idea.

Thanks,
Dileepa


> Cheers,
> Reto
>
>
> On Wed, May 1, 2013 at 12:13 AM, Dileepa Jayakody <
> dileepajayakody@gmail.com
> > wrote:
>
> > Hi Reto et al,
> >
> > I also, took a look at the user-management feature on the felix console.
> > Looks cool.
> > I think it can be further improved by integrating OpenID/WebID so that
> even
> > users who are not registered in the Stanbol system can benefit from it
> > (this is in addition to the WebID integration benefits discussed
> previously
> > :) ).
> > WDYT?
> >
> > Thanks,
> > Dileepa
> >
> >
> > On Wed, May 1, 2013 at 2:09 AM, Michael Luggen <michael.luggen@gmail.com
> > >wrote:
> >
> > > I did check out the usermanagement, looks quite nice,
> > >
> > > Is there also an web service interface to let self-register new users
> and
> > > let them choose a password from a web application standpoint view?
> > >
> > > Thanks for inputs,
> > > Michael
> > >
> > >
> > > On 26.04.2013, at 09:25, Reto Bachmann-Gmür <re...@wymiwyg.com> wrote:
> > >
> > > > On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <
> dileepajayakody@gmail.com
> > >
> > > > wrote:
> > > >
> > > >> 2. WebID based authentication and userstore :Why doesn't Stanbol
> have
> > a
> > > >> user-management module at the moment ? I think a user-store is a
> valid
> > > >> requirement for any CMS.
> > > >
> > > > You find some initial Stanbol usermanagement on the felix Web
> console.
> > > >
> > > > Cheers,
> > > > Reto
> > >
> > >
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@wymiwyg.com>]

Hi Dileepa,

I think this makes sense. Similar to self registered users (previous mail)
a user that comes with WebId or OpenId would get an account and be assigned
a Role (roaming-user role). Additionally to that there could be a mechanism
assigning permissions based on social-network distance.

A first step might be to add an optional URI field in the user manager,
this URI would be the WebID/OpenID the user can use as an optional log-in
mechanism.

Cheers,
Reto


On Wed, May 1, 2013 at 12:13 AM, Dileepa Jayakody <dileepajayakody@gmail.com
> wrote:

> Hi Reto et al,
>
> I also, took a look at the user-management feature on the felix console.
> Looks cool.
> I think it can be further improved by integrating OpenID/WebID so that even
> users who are not registered in the Stanbol system can benefit from it
> (this is in addition to the WebID integration benefits discussed previously
> :) ).
> WDYT?
>
> Thanks,
> Dileepa
>
>
> On Wed, May 1, 2013 at 2:09 AM, Michael Luggen <michael.luggen@gmail.com
> >wrote:
>
> > I did check out the usermanagement, looks quite nice,
> >
> > Is there also an web service interface to let self-register new users and
> > let them choose a password from a web application standpoint view?
> >
> > Thanks for inputs,
> > Michael
> >
> >
> > On 26.04.2013, at 09:25, Reto Bachmann-Gmür <re...@wymiwyg.com> wrote:
> >
> > > On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <dileepajayakody@gmail.com
> >
> > > wrote:
> > >
> > >> 2. WebID based authentication and userstore :Why doesn't Stanbol have
> a
> > >> user-management module at the moment ? I think a user-store is a valid
> > >> requirement for any CMS.
> > >
> > > You find some initial Stanbol usermanagement on the felix Web console.
> > >
> > > Cheers,
> > > Reto
> >
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Reto et al,

I also, took a look at the user-management feature on the felix console.
Looks cool.
I think it can be further improved by integrating OpenID/WebID so that even
users who are not registered in the Stanbol system can benefit from it
(this is in addition to the WebID integration benefits discussed previously
:) ).
WDYT?

Thanks,
Dileepa


On Wed, May 1, 2013 at 2:09 AM, Michael Luggen <mi...@gmail.com>wrote:

> I did check out the usermanagement, looks quite nice,
>
> Is there also an web service interface to let self-register new users and
> let them choose a password from a web application standpoint view?
>
> Thanks for inputs,
> Michael
>
>
> On 26.04.2013, at 09:25, Reto Bachmann-Gmür <re...@wymiwyg.com> wrote:
>
> > On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <di...@gmail.com>
> > wrote:
> >
> >> 2. WebID based authentication and userstore :Why doesn't Stanbol have a
> >> user-management module at the moment ? I think a user-store is a valid
> >> requirement for any CMS.
> >
> > You find some initial Stanbol usermanagement on the felix Web console.
> >
> > Cheers,
> > Reto
>
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@wymiwyg.com>]

Hi Michael

Self-Registration sounds like a useful feature. I think all user that
self-register should automatically get a role (self-registerer role)
assigned so that the admin can assign some particular right to this group.

How should self-registration look like? The most primitive version (submit
username and password and you have your account) could be enhanced with
capchca or with email verification. What is the minimum to have a useful
first version? If you have a suggestion you're welcome to create an issue.

Cheers,
Reto


On Tue, Apr 30, 2013 at 10:39 PM, Michael Luggen
<mi...@gmail.com>wrote:

> I did check out the usermanagement, looks quite nice,
>
> Is there also an web service interface to let self-register new users and
> let them choose a password from a web application standpoint view?
>
> Thanks for inputs,
> Michael
>
>
> On 26.04.2013, at 09:25, Reto Bachmann-Gmür <re...@wymiwyg.com> wrote:
>
> > On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <di...@gmail.com>
> > wrote:
> >
> >> 2. WebID based authentication and userstore :Why doesn't Stanbol have a
> >> user-management module at the moment ? I think a user-store is a valid
> >> requirement for any CMS.
> >
> > You find some initial Stanbol usermanagement on the felix Web console.
> >
> > Cheers,
> > Reto
>
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Michael Luggen <mi...@gmail.com>]

I did check out the usermanagement, looks quite nice,

Is there also an web service interface to let self-register new users and let them choose a password from a web application standpoint view?

Thanks for inputs,
Michael


On 26.04.2013, at 09:25, Reto Bachmann-Gmür <re...@wymiwyg.com> wrote:

> On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <di...@gmail.com>
> wrote:
> 
>> 2. WebID based authentication and userstore :Why doesn't Stanbol have a
>> user-management module at the moment ? I think a user-store is a valid
>> requirement for any CMS.
> 
> You find some initial Stanbol usermanagement on the felix Web console.
> 
> Cheers,
> Reto

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@wymiwyg.com>]

On Apr 26, 2013 4:04 AM, "Dileepa Jayakody" <di...@gmail.com>
wrote:

> 2. WebID based authentication and userstore :Why doesn't Stanbol have a
> user-management module at the moment ? I think a user-store is a valid
> requirement for any CMS.

You find some initial Stanbol usermanagement on the felix Web console.

Cheers,
Reto

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Andreas et al,

Let's not just fix the WebID feature, but also use it for greater purposes
in Stanbol :)
Several potential use-cases of WebID integration in Stanbol were formulated
in this mail-thread with the help of Rupert, Fabian etc.

Some interesting usecases discussed in this mail thread are as below. I
think one or more items can be done in addition to fixing the WebID feature
in Clerezza as the GSoC project.
Would like to get suggestions from everyone as much as possible and prepare
the proposal including your ideas.

1. WebID of the user can be used by the enhancement engines to provide
more relevant and contextual enhancement results to the user.

2. WebID based authentication and userstore :Why doesn't Stanbol have a
user-management module at the moment ? I think a user-store is a valid
requirement for any CMS.

3. Entity Disambiguation for Persons using WebID : A WebID has a lot of
linked-data which provides contextual information about the person which
could highly useful for Entity disambiguation in the enhancement phase.

4. Knowledge population.
Related to #3.  WebID can be used for entity-disambiguation and also to
introduce new entities to the knowledge base formulated from the WebID

5. Social Profile : WebID can be used as a portable Profile across multiple
SNSs and the data collected can be used to store personalized content in
the CMS.

6. A Federated to Sign-On using WebID : not directly relevant to Stanbol ,
but a valid use case in general

Thoughts are welcome!

Thanks,
Dileepa


On Thu, Apr 25, 2013 at 5:23 PM, Andreas Kuckartz <A....@ping.de>wrote:

> Hi Dileepa and others,
>
> in general I do think that this can be part of a GSoC project.
>
> A project involving both, the Apache Stanbol and Apache Clerezza
> communities, should not be an unsolvable problem as those are
> overlapping. I also would not be against someone from Apache Clerezza
> mentoring this alone if that is necessary for any reasons (especially
> formal ones).
>
> But I fear that "only" fixing a feature which already worked in the past
> would not be enough to convince the evaluators of the proposal -
> independent of the amount of work involved (and I have no idea how much
> that would be).
>
> What can be done realistically in addition to fixing this feature in
> Clerezza?
>
> Can you describe what WebID integration in Stanbol would look like so
> that proposal evaluators can easily understand it? (I might have missed
> that in the other mails)
>
> Cheers,
> Andreas
> ---
>
> Dileepa Jayakody:
> > Thanks Rupert, Reto for your ideas..
> >
> > I went through Apache Clerazza project by checking out the source from
> the
> > git repo at [1].I think the webID integration code that Reto refers to is
> > at [2]? Please correct me if I'm wrong. How about we make it working
> again?
> > Regarding the ssl http problems maybe we can ask Eclipse Communication
> > Framework [3](ECF) guys to collaborate.
> >
> > As you have listed down revoking the WebID in Clerazza and in-effect
> > Stanbol can bring so many benefits to it as a semantic access-control and
> > authentication framework. WebID integration can be very useful in the
> > entity disambiguation as well.
> >
> > I would love to know if this can become a possible GSoC project.
> >
> > Thanks,
> > Dileepa
> >
> > [1] https://github.com/apache/clerezza
> > [2]
> https://github.com/apache/clerezza/tree/trunk/platform.security.foafssl
> > [3] http://wiki.eclipse.org/ECF
> >
> >
> >
> > On Thu, Apr 25, 2013 at 2:05 PM, Rupert Westenthaler <
> > rupert.westenthaler@gmail.com> wrote:
> >
> >>> @Rupert: With regard to your opinion on using EntityHub as an index
> >> rather
> >>> than the primary storage for the knowledge base, can you please
> elaborate
> >>> more on why do you think so? Maybe Wiki-Link or Freebase [3] is a good
> >>> option for such a knowledge base?
> >>
> >> The Entityhub is designed to play the role of an index & local cache.
> >> It was not designed to manage knowledge. Because of that I would
> >> recommend to use the knowledge base that does best fit your data
> >> (Triple Store, Graph Data Base, SQL, NoSQL ...) and then use the CRUD
> >> interface of the Entityhub to create ManagedSites with those data
> >> needed for EntiyLinking and Similarity queries.
> >>
> >> Note that Enhancement Engines can also directly interact with the
> >> knowledge base (e.g. if the want to modify the knowledge base such as
> >> adding a mention example or creating an unknown Entity). However the
> >> Entityhub ensures really fast lookups as required for EntityLinking.
> >>
> >> best
> >> Rupert
> >>
> >> --
> >> | Rupert Westenthaler             rupert.westenthaler@gmail.com
> >> | Bodenlehenstraße 11                             ++43-699-11108907
> >> | A-5500 Bischofshofen
> >>
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Andreas Kuckartz <A....@ping.de>]

Hi Dileepa and others,

in general I do think that this can be part of a GSoC project.

A project involving both, the Apache Stanbol and Apache Clerezza
communities, should not be an unsolvable problem as those are
overlapping. I also would not be against someone from Apache Clerezza
mentoring this alone if that is necessary for any reasons (especially
formal ones).

But I fear that "only" fixing a feature which already worked in the past
would not be enough to convince the evaluators of the proposal -
independent of the amount of work involved (and I have no idea how much
that would be).

What can be done realistically in addition to fixing this feature in
Clerezza?

Can you describe what WebID integration in Stanbol would look like so
that proposal evaluators can easily understand it? (I might have missed
that in the other mails)

Cheers,
Andreas
---

Dileepa Jayakody:
> Thanks Rupert, Reto for your ideas..
> 
> I went through Apache Clerazza project by checking out the source from the
> git repo at [1].I think the webID integration code that Reto refers to is
> at [2]? Please correct me if I'm wrong. How about we make it working again?
> Regarding the ssl http problems maybe we can ask Eclipse Communication
> Framework [3](ECF) guys to collaborate.
> 
> As you have listed down revoking the WebID in Clerazza and in-effect
> Stanbol can bring so many benefits to it as a semantic access-control and
> authentication framework. WebID integration can be very useful in the
> entity disambiguation as well.
> 
> I would love to know if this can become a possible GSoC project.
> 
> Thanks,
> Dileepa
> 
> [1] https://github.com/apache/clerezza
> [2] https://github.com/apache/clerezza/tree/trunk/platform.security.foafssl
> [3] http://wiki.eclipse.org/ECF
> 
> 
> 
> On Thu, Apr 25, 2013 at 2:05 PM, Rupert Westenthaler <
> rupert.westenthaler@gmail.com> wrote:
> 
>>> @Rupert: With regard to your opinion on using EntityHub as an index
>> rather
>>> than the primary storage for the knowledge base, can you please elaborate
>>> more on why do you think so? Maybe Wiki-Link or Freebase [3] is a good
>>> option for such a knowledge base?
>>
>> The Entityhub is designed to play the role of an index & local cache.
>> It was not designed to manage knowledge. Because of that I would
>> recommend to use the knowledge base that does best fit your data
>> (Triple Store, Graph Data Base, SQL, NoSQL ...) and then use the CRUD
>> interface of the Entityhub to create ManagedSites with those data
>> needed for EntiyLinking and Similarity queries.
>>
>> Note that Enhancement Engines can also directly interact with the
>> knowledge base (e.g. if the want to modify the knowledge base such as
>> adding a mention example or creating an unknown Entity). However the
>> Entityhub ensures really fast lookups as required for EntityLinking.
>>
>> best
>> Rupert
>>
>> --
>> | Rupert Westenthaler             rupert.westenthaler@gmail.com
>> | Bodenlehenstraße 11                             ++43-699-11108907
>> | A-5500 Bischofshofen
>>
> 

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@wymiwyg.com>]

On Thu, Apr 25, 2013 at 11:40 AM, Dileepa Jayakody <
dileepajayakody@gmail.com> wrote:

> Thanks Rupert, Reto for your ideas..
>
> I went through Apache Clerazza project by checking out the source from the
> git repo at [1].I think the webID integration code that Reto refers to is
> at [2]? Please correct me if I'm wrong. How about we make it working again?
> Regarding the ssl http problems maybe we can ask Eclipse Communication
> Framework [3](ECF) guys to collaborate.
>
> As you have listed down revoking the WebID in Clerazza and in-effect
> Stanbol can bring so many benefits to it as a semantic access-control and
> authentication framework. WebID integration can be very useful in the
> entity disambiguation as well.
>
> I would love to know if this can become a possible GSoC project.
>
> Thanks,
> Dileepa
>
> [1] https://github.com/apache/clerezza
> [2]
> https://github.com/apache/clerezza/tree/trunk/platform.security.foafssl
> [3] http://wiki.eclipse.org/ECF
>
>
>
> On Thu, Apr 25, 2013 at 2:05 PM, Rupert Westenthaler <
> rupert.westenthaler@gmail.com> wrote:
>
> > > @Rupert: With regard to your opinion on using EntityHub as an index
> > rather
> > > than the primary storage for the knowledge base, can you please
> elaborate
> > > more on why do you think so? Maybe Wiki-Link or Freebase [3] is a good
> > > option for such a knowledge base?
> >
> > The Entityhub is designed to play the role of an index & local cache.
> > It was not designed to manage knowledge. Because of that I would
> > recommend to use the knowledge base that does best fit your data
> > (Triple Store, Graph Data Base, SQL, NoSQL ...) and then use the CRUD
> > interface of the Entityhub to create ManagedSites with those data
> > needed for EntiyLinking and Similarity queries.
> >
> > Note that Enhancement Engines can also directly interact with the
> > knowledge base (e.g. if the want to modify the knowledge base such as
> > adding a mention example or creating an unknown Entity). However the
> > Entityhub ensures really fast lookups as required for EntityLinking.
> >
> > best
> > Rupert
> >
> > --
> > | Rupert Westenthaler             rupert.westenthaler@gmail.com
> > | Bodenlehenstraße 11                             ++43-699-11108907
> > | A-5500 Bischofshofen
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Thanks Rupert, Reto for your ideas..

I went through Apache Clerazza project by checking out the source from the
git repo at [1].I think the webID integration code that Reto refers to is
at [2]? Please correct me if I'm wrong. How about we make it working again?
Regarding the ssl http problems maybe we can ask Eclipse Communication
Framework [3](ECF) guys to collaborate.

As you have listed down revoking the WebID in Clerazza and in-effect
Stanbol can bring so many benefits to it as a semantic access-control and
authentication framework. WebID integration can be very useful in the
entity disambiguation as well.

I would love to know if this can become a possible GSoC project.

Thanks,
Dileepa

[1] https://github.com/apache/clerezza
[2] https://github.com/apache/clerezza/tree/trunk/platform.security.foafssl
[3] http://wiki.eclipse.org/ECF



On Thu, Apr 25, 2013 at 2:05 PM, Rupert Westenthaler <
rupert.westenthaler@gmail.com> wrote:

> > @Rupert: With regard to your opinion on using EntityHub as an index
> rather
> > than the primary storage for the knowledge base, can you please elaborate
> > more on why do you think so? Maybe Wiki-Link or Freebase [3] is a good
> > option for such a knowledge base?
>
> The Entityhub is designed to play the role of an index & local cache.
> It was not designed to manage knowledge. Because of that I would
> recommend to use the knowledge base that does best fit your data
> (Triple Store, Graph Data Base, SQL, NoSQL ...) and then use the CRUD
> interface of the Entityhub to create ManagedSites with those data
> needed for EntiyLinking and Similarity queries.
>
> Note that Enhancement Engines can also directly interact with the
> knowledge base (e.g. if the want to modify the knowledge base such as
> adding a mention example or creating an unknown Entity). However the
> Entityhub ensures really fast lookups as required for EntityLinking.
>
> best
> Rupert
>
> --
> | Rupert Westenthaler             rupert.westenthaler@gmail.com
> | Bodenlehenstraße 11                             ++43-699-11108907
> | A-5500 Bischofshofen
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Rupert Westenthaler <ru...@gmail.com>]

> @Rupert: With regard to your opinion on using EntityHub as an index rather
> than the primary storage for the knowledge base, can you please elaborate
> more on why do you think so? Maybe Wiki-Link or Freebase [3] is a good
> option for such a knowledge base?

The Entityhub is designed to play the role of an index & local cache.
It was not designed to manage knowledge. Because of that I would
recommend to use the knowledge base that does best fit your data
(Triple Store, Graph Data Base, SQL, NoSQL ...) and then use the CRUD
interface of the Entityhub to create ManagedSites with those data
needed for EntiyLinking and Similarity queries.

Note that Enhancement Engines can also directly interact with the
knowledge base (e.g. if the want to modify the knowledge base such as
adding a mention example or creating an unknown Entity). However the
Entityhub ensures really fast lookups as required for EntityLinking.

best
Rupert

--
| Rupert Westenthaler             rupert.westenthaler@gmail.com
| Bodenlehenstraße 11                             ++43-699-11108907
| A-5500 Bischofshofen

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@apache.org>]

> If we go forward with integrating WebID in Stanbol
> as part of the federated social network integration or disambiguation tasks
> we can implement a WebID connector to such systems or implement a WebID
> provider on Stanbol side. WDYT?
>

This feature was present in Clerezza (which in terms of authentication is
identical to Stanbol now) till shortly. It's currently disabled because it
needs to be ported to the new servlet/jaxrs architecture and adapted to
recent changed to the WebID spec.

Basically it provided following features:
- Associate a WebID to an account
- Accept WebID for login
- Create WebIDs that are accepted by other systems
- Assign some permissions to unknown WebID such as by computing the
distance in the social graph

I guess the currently disabled code might give a starting point for the
WebID support. A difficulty is that webID supports needs interfering with
the ssl stack provided by the https server so we needed a special osgi http
service.

Cheers,
Reto

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Rupert et al,

Thanks a lot for your valuable insights regarding use cases of digital
identity with Stanbol and in semantic web in general. I think verified
digital identity and semantic-web goes hand in hand, merging the concepts
is beneficial for efficient and accurate disambiguation of entities and
accurate semantic enhancements.

As you have suggested WebID can be used as a federated social profile
across multiple SNSs, connecting to the same resources (contact
info,photos, friends, interests) identified uniquely via WebID. The
linked-data described by the WebID can be used to extract verified
knowledge about the person. As Antonio has described WebID is a unique URI
representing an entity and security is implemented using PKI. It can be
used as an authentication mechanism across the web with WebID supported
platforms. Essentially WebID is written in FOAF + some SSL which can be
used to create a trusted graph of entities. For more use cases please refer
[1], [2].

WebID is an emerging protocol and it's yet to be widely adopted. I too
don't have prior experience with WebID integrations, but I'm very much
interested to take a deep dive into the subject domain based on my research
interests. There are several WebID providers such as OpenLink Data Spaces
[4] and My-Profile [5]. If we go forward with integrating WebID in Stanbol
as part of the federated social network integration or disambiguation tasks
we can implement a WebID connector to such systems or implement a WebID
provider on Stanbol side. WDYT?

@Rupert: With regard to your opinion on using EntityHub as an index rather
than the primary storage for the knowledge base, can you please elaborate
more on why do you think so? Maybe Wiki-Link or Freebase [3] is a good
option for such a knowledge base?

Thanks,
Dileepa

[1]http://www.w3.org/wiki/WebID
[2] http://www.w3.org/wiki/Foaf%2Bssl/RelyingParties
[3] http://wiki.freebase.com/wiki/Main_Page
[4] http://ods.openlinksw.com/wiki/ODS/VirtODSSecurityWebID
[5] https://my-profile.eu



On Wed, Apr 24, 2013 at 3:42 PM, Antonio Perez <ap...@zaizi.com> wrote:

> Hi Rupert, all
>
> First of all, let me introduce myself. I'm Antonio and recently I have
> started to work with Stanbol at Zaizi. I would like to prepare a proposal
> for GSoC related to the disambiguation problem.
>
> Regarding your last question Rupert, the objective of WebID is precisely
> that, to represent an entity by an URI and also give security (trust) using
> client certificates.
> These certificates (representing entities for example) contains the URI of
> the resource and you can access it to obtain more information.
> Take into account that Webid is basically a fancy name for SSL client side
> certificates with a profile url in it. So you can trust that the resource
> is a valid resource (using the certificate public key against the public
> key contained in the URI of the resource) and you can extract more
> information about it using the data given in the content of that URI.
>
> Regards
>
>
>
> On Wed, Apr 24, 2013 at 10:43 AM, Rupert Westenthaler <
> rupert.westenthaler@gmail.com> wrote:
>
> > Hi Dileepa, all
> >
> > Interesting discussion with a lot of very relevant topics. In this
> > mail I select (1..4) points of the ongoing discussion and add my
> > thoughts to it. In addition I will add (5) Knowledge base population
> > as an additional usage case scenario.
> >
> > (1) Digital Identity in social Networks: I am not a export for the
> > ontonet, rules and reasoning component, so I will not comment on how
> > those could be used to process structured information of profiles.
> > However the Stanbol Enhancer would be well suited to deal with
> > unstructured information provided by such profiles.
> >
> > (2) Social Profile of a Person/Organization: Building up a knowledge
> > base by using information collected from Social Network activities. I
> > would not recommend to use the Stanbol Entityhub as primary storage
> > for such a knowledge base. However the Entityhub could be used to
> > create/store specialized indexes derived from such a knowledge base.
> > Those indexes would then be used (instead of the knowledge base) by
> > Enhancement Engines to make lookups for words of the text, Prefix
> > based suggestion as used by user interfaces ...
> >
> > (3) Disambiguation #1: With (1) and (2) you will have multiple Social
> > Profiles using the same name from different information sources (e.g.
> > Twitter and Facebook). Deciding if those profiles are in fact the same
> > Entity will need some kind of disambiguation. The result of those
> > could mean to merge/split profiles in the knowledge base.
> >
> > (4) Disambiguation #2: Profiles in the Knowledge might be referenced
> > in texts by the same word/phrase (e.g. similar person names, nick
> > names, acronyms or organizations ...). Profile information can be used
> > to ensure that the correct Person/Organization is linked.
> >
> > (5) Knowledge base population: This would allow (a) Users or (b) an
> > Enhancement Engine to create new Persons/Organizations in the
> > knowledge base. An Example for (b) could be detected Named Entities
> > (NER) that where not linked to an known Entity. In addition an (c)
> > Entity Disambiguation Engine that comes to the conclusion that none of
> > the existing Entities is a fit for the mention in the text might also
> > create an new Entity. (1) and (2) could then be used to populate the
> > knowledge base with further information originating from social
> > networks.
> >
> > Finally I would like to ask how widely used is WebId? Or in other
> > words: Would it usage limit the list of possible information sources?
> > Could it be used as unique identifier for data collected in the
> > knowledge base holding the Social Profile Information of a
> > Person/Organizations?
> >
> > best
> > Rupert
> >
> > On Tue, Apr 23, 2013 at 10:32 PM, Dileepa Jayakody
> > <di...@gmail.com> wrote:
> > > Hi Rafa,
> > >
> > >
> > > Thanks for your ideas.
> > >
> > > WebID [1] is a URI based protocol to identify an entity (person,
> > > organization) uniquely over the web. Since WebID is based on
> linked-data
> > > (FOAF [3] syntax) we can extract information such as person's
> interests,
> > > employment history, cv, publications, and friends by analyzing the
> linked
> > > resources described in the WebID. When a WebID is associated with a
> > person
> > > it can be used as a single-sign-on protocol like OpenID [2] as well.
> > >
> > > WebID can also be instrumental in social-web to have a single-profile
> > over
> > > multiple social networks with linked resources like photos,
> friend-lists
> > > etc. Also security layers can be implemented on top WebID to apply
> > > fine-grained access control to such linked-resources described by the
> > > WebID. l
> > >
> > > As you have pointed out WebID can be used to implement a disambiguation
> > > algorithm for people entities in Stanbol. I also think it can be highly
> > > useful in semantic lifting by extracting information like, employment
> > data,
> > > friend's profiles, publications etc described by the WebID. IMO
> sentiment
> > > analysis can also benefit from WebID.
> > >
> > >
> > > Thanks,
> > >
> > > Dileepa
> > >
> > > [1] http://www.w3.org/wiki/WebID
> > >
> > > [2] http://en.wikipedia.org/wiki/OpenID
> > > [3] http://xmlns.com/foaf/spec/
> > >
> > >
> > > On Tue, Apr 23, 2013 at 8:37 PM, Rafa Haro <rh...@zaizi.com> wrote:
> > >
> > >> Hi Dileepa,
> > >>
> > >> El 23/04/13 13:45, Dileepa Jayakody escribió:
> > >>
> > >>  Hi Fabian et al,
> > >>>
> > >>> Thanks a lot for your valuable ideas.
> > >>> Yes it's really interesting to implement a 'person | organization'
> > >>> disambiguation module using WebID protocol as part of Stanbol
> > Enhancement
> > >>> Engine. I went through the documentation of Stanbol and I have gained
> > an
> > >>> overall idea about the architecture of Stanbol.
> > >>>
> > >> +1. That's a great idea. I don't know very much about WebID protocol,
> > but
> > >> as far as you could use some profile data as disambiguation contexts,
> it
> > >> should be feasible to implement a disambiguation algorithm. Could you
> > >> please give us a concrete example where WebID is used? I suppose that
> > the
> > >> general use case is to link name mentions in web pages with digital
> > >> identities. What kind of information is it possible to gather from
> WebID
> > >> identities?
> > >>
> > >>
> > >>> It would be great to get more ideas, suggestions about how to use
> > Stanbol
> > >>> for people, organization disambiguation and to discuss the objectives
> > and
> > >>> milestones in the GSOC project idea at [1].
> > >>> I also think one of the main factor for disambiguation is the
> > >>> data-set/knowledge base used for the process. What is the data-set
> > Stanbol
> > >>> uses to verify data? Is Google Wiki-links released recently [1] a
> > >>> candidate
> > >>> for Stanbol data-set?
> > >>>
> > >> Initially, you can use any knowledge base in Stanbol. I always
> identify
> > >> EntityHub component as a "Knowledge Base" management system, although
> > maybe
> > >> formally the EntityHub is not exactly that. Anyway, Google Wiki-links
> > could
> > >> be a good resource for disambiguation when the knowledge base is
> > Wikipedia
> > >> or DBpedia. In fact, Wiki-links contains 40 millions of mentions and
> its
> > >> contexts retrieved from web pages. This information can be eventually
> > added
> > >> to a Wikipedia or DBpedia knowledge base as disambiguation contexts
> for
> > the
> > >> entities covered in the dataset. Another interesting resource, as the
> > new
> > >> in techcrunch points, is the dictionary of Wikipedia concepts released
> > last
> > >> year [1]. This resource can be used to include more labels for each
> > entity
> > >> (possible names), improving then the candidate selection step. As
> > always,
> > >> we face a recall/precision problem with such dictionary.
> > >>
> > >> [1]  - http://googleresearch.**blogspot.com.es/2012/05/from-**
> > >> words-to-concepts-and-back.**html<
> >
> http://googleresearch.blogspot.com.es/2012/05/from-words-to-concepts-and-back.html
> > >
> > >>
> > >> Regards!
> > >>
> > >>
> > >>> Thanks,
> > >>> Dileepa
> > >>>
> > >>> [1]
> > >>> http://techcrunch.com/2013/03/**08/google-research-releases-**
> > >>> wikilinks-corpus-with-40m-**mentions-and-3m-entities/<
> >
> http://techcrunch.com/2013/03/08/google-research-releases-wikilinks-corpus-with-40m-mentions-and-3m-entities/
> > >
> > >>> On Mon, Apr 22, 2013 at 7:32 PM, Fabian Christ <
> > >>> christ.fabian@googlemail.com
> > >>>
> > >>>> wrote:
> > >>>> Hi,
> > >>>>
> > >>>> 2013/4/22 Dileepa Jayakody <di...@gmail.com>:
> > >>>>
> > >>>>> Could it be a valid use-case to integrate WebID protocol in Stanbol
> > to
> > >>>>> create social graphs and related ontologies?
> > >>>>>
> > >>>> the already mentioned entity disambiguation for persons might be
> such
> > >>>> a use case.
> > >>>>
> > >>>> Another idea could be that the enhancement process uses some
> > >>>> information from the personal profile of the user who sends the
> > >>>> request. I do not have any concrete example at the moment but
> engines
> > >>>> might be interested in knowing who is sending an enhancement
> request.
> > >>>> This may also be a relevant information for the disambiguation task.
> > >>>>
> > >>>> Best,
> > >>>>    - Fabian
> > >>>>
> > >>>>
> > >>>> --
> > >>>> Fabian
> > >>>> http://twitter.com/fctwitt
> > >>>>
> > >>>>
> > >>
> > >> --
> > >>
> > >> ------------------------------
> > >> This message should be regarded as confidential. If you have received
> > this
> > >> email in error please notify the sender and destroy it immediately.
> > >> Statements of intent shall only become binding when confirmed in hard
> > copy
> > >> by an authorised signatory.
> > >>
> > >> Zaizi Ltd is registered in England and Wales with the registration
> > number
> > >> 6440931. The Registered Office is 222 Westbourne Studios, 242 Acklam
> > Road,
> > >> London W10 5JJ, UK.
> > >>
> >
> >
> >
> > --
> > | Rupert Westenthaler             rupert.westenthaler@gmail.com
> > | Bodenlehenstraße 11                             ++43-699-11108907
> > | A-5500 Bischofshofen
> >
>
> --
>
> ------------------------------
> This message should be regarded as confidential. If you have received this
> email in error please notify the sender and destroy it immediately.
> Statements of intent shall only become binding when confirmed in hard copy
> by an authorised signatory.
>
> Zaizi Ltd is registered in England and Wales with the registration number
> 6440931. The Registered Office is 222 Westbourne Studios, 242 Acklam Road,
> London W10 5JJ, UK.
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Antonio Perez <ap...@zaizi.com>]

Hi Rupert, all

First of all, let me introduce myself. I'm Antonio and recently I have
started to work with Stanbol at Zaizi. I would like to prepare a proposal
for GSoC related to the disambiguation problem.

Regarding your last question Rupert, the objective of WebID is precisely
that, to represent an entity by an URI and also give security (trust) using
client certificates.
These certificates (representing entities for example) contains the URI of
the resource and you can access it to obtain more information.
Take into account that Webid is basically a fancy name for SSL client side
certificates with a profile url in it. So you can trust that the resource
is a valid resource (using the certificate public key against the public
key contained in the URI of the resource) and you can extract more
information about it using the data given in the content of that URI.

Regards



On Wed, Apr 24, 2013 at 10:43 AM, Rupert Westenthaler <
rupert.westenthaler@gmail.com> wrote:

> Hi Dileepa, all
>
> Interesting discussion with a lot of very relevant topics. In this
> mail I select (1..4) points of the ongoing discussion and add my
> thoughts to it. In addition I will add (5) Knowledge base population
> as an additional usage case scenario.
>
> (1) Digital Identity in social Networks: I am not a export for the
> ontonet, rules and reasoning component, so I will not comment on how
> those could be used to process structured information of profiles.
> However the Stanbol Enhancer would be well suited to deal with
> unstructured information provided by such profiles.
>
> (2) Social Profile of a Person/Organization: Building up a knowledge
> base by using information collected from Social Network activities. I
> would not recommend to use the Stanbol Entityhub as primary storage
> for such a knowledge base. However the Entityhub could be used to
> create/store specialized indexes derived from such a knowledge base.
> Those indexes would then be used (instead of the knowledge base) by
> Enhancement Engines to make lookups for words of the text, Prefix
> based suggestion as used by user interfaces ...
>
> (3) Disambiguation #1: With (1) and (2) you will have multiple Social
> Profiles using the same name from different information sources (e.g.
> Twitter and Facebook). Deciding if those profiles are in fact the same
> Entity will need some kind of disambiguation. The result of those
> could mean to merge/split profiles in the knowledge base.
>
> (4) Disambiguation #2: Profiles in the Knowledge might be referenced
> in texts by the same word/phrase (e.g. similar person names, nick
> names, acronyms or organizations ...). Profile information can be used
> to ensure that the correct Person/Organization is linked.
>
> (5) Knowledge base population: This would allow (a) Users or (b) an
> Enhancement Engine to create new Persons/Organizations in the
> knowledge base. An Example for (b) could be detected Named Entities
> (NER) that where not linked to an known Entity. In addition an (c)
> Entity Disambiguation Engine that comes to the conclusion that none of
> the existing Entities is a fit for the mention in the text might also
> create an new Entity. (1) and (2) could then be used to populate the
> knowledge base with further information originating from social
> networks.
>
> Finally I would like to ask how widely used is WebId? Or in other
> words: Would it usage limit the list of possible information sources?
> Could it be used as unique identifier for data collected in the
> knowledge base holding the Social Profile Information of a
> Person/Organizations?
>
> best
> Rupert
>
> On Tue, Apr 23, 2013 at 10:32 PM, Dileepa Jayakody
> <di...@gmail.com> wrote:
> > Hi Rafa,
> >
> >
> > Thanks for your ideas.
> >
> > WebID [1] is a URI based protocol to identify an entity (person,
> > organization) uniquely over the web. Since WebID is based on linked-data
> > (FOAF [3] syntax) we can extract information such as person's interests,
> > employment history, cv, publications, and friends by analyzing the linked
> > resources described in the WebID. When a WebID is associated with a
> person
> > it can be used as a single-sign-on protocol like OpenID [2] as well.
> >
> > WebID can also be instrumental in social-web to have a single-profile
> over
> > multiple social networks with linked resources like photos, friend-lists
> > etc. Also security layers can be implemented on top WebID to apply
> > fine-grained access control to such linked-resources described by the
> > WebID. l
> >
> > As you have pointed out WebID can be used to implement a disambiguation
> > algorithm for people entities in Stanbol. I also think it can be highly
> > useful in semantic lifting by extracting information like, employment
> data,
> > friend's profiles, publications etc described by the WebID. IMO sentiment
> > analysis can also benefit from WebID.
> >
> >
> > Thanks,
> >
> > Dileepa
> >
> > [1] http://www.w3.org/wiki/WebID
> >
> > [2] http://en.wikipedia.org/wiki/OpenID
> > [3] http://xmlns.com/foaf/spec/
> >
> >
> > On Tue, Apr 23, 2013 at 8:37 PM, Rafa Haro <rh...@zaizi.com> wrote:
> >
> >> Hi Dileepa,
> >>
> >> El 23/04/13 13:45, Dileepa Jayakody escribió:
> >>
> >>  Hi Fabian et al,
> >>>
> >>> Thanks a lot for your valuable ideas.
> >>> Yes it's really interesting to implement a 'person | organization'
> >>> disambiguation module using WebID protocol as part of Stanbol
> Enhancement
> >>> Engine. I went through the documentation of Stanbol and I have gained
> an
> >>> overall idea about the architecture of Stanbol.
> >>>
> >> +1. That's a great idea. I don't know very much about WebID protocol,
> but
> >> as far as you could use some profile data as disambiguation contexts, it
> >> should be feasible to implement a disambiguation algorithm. Could you
> >> please give us a concrete example where WebID is used? I suppose that
> the
> >> general use case is to link name mentions in web pages with digital
> >> identities. What kind of information is it possible to gather from WebID
> >> identities?
> >>
> >>
> >>> It would be great to get more ideas, suggestions about how to use
> Stanbol
> >>> for people, organization disambiguation and to discuss the objectives
> and
> >>> milestones in the GSOC project idea at [1].
> >>> I also think one of the main factor for disambiguation is the
> >>> data-set/knowledge base used for the process. What is the data-set
> Stanbol
> >>> uses to verify data? Is Google Wiki-links released recently [1] a
> >>> candidate
> >>> for Stanbol data-set?
> >>>
> >> Initially, you can use any knowledge base in Stanbol. I always identify
> >> EntityHub component as a "Knowledge Base" management system, although
> maybe
> >> formally the EntityHub is not exactly that. Anyway, Google Wiki-links
> could
> >> be a good resource for disambiguation when the knowledge base is
> Wikipedia
> >> or DBpedia. In fact, Wiki-links contains 40 millions of mentions and its
> >> contexts retrieved from web pages. This information can be eventually
> added
> >> to a Wikipedia or DBpedia knowledge base as disambiguation contexts for
> the
> >> entities covered in the dataset. Another interesting resource, as the
> new
> >> in techcrunch points, is the dictionary of Wikipedia concepts released
> last
> >> year [1]. This resource can be used to include more labels for each
> entity
> >> (possible names), improving then the candidate selection step. As
> always,
> >> we face a recall/precision problem with such dictionary.
> >>
> >> [1]  - http://googleresearch.**blogspot.com.es/2012/05/from-**
> >> words-to-concepts-and-back.**html<
> http://googleresearch.blogspot.com.es/2012/05/from-words-to-concepts-and-back.html
> >
> >>
> >> Regards!
> >>
> >>
> >>> Thanks,
> >>> Dileepa
> >>>
> >>> [1]
> >>> http://techcrunch.com/2013/03/**08/google-research-releases-**
> >>> wikilinks-corpus-with-40m-**mentions-and-3m-entities/<
> http://techcrunch.com/2013/03/08/google-research-releases-wikilinks-corpus-with-40m-mentions-and-3m-entities/
> >
> >>> On Mon, Apr 22, 2013 at 7:32 PM, Fabian Christ <
> >>> christ.fabian@googlemail.com
> >>>
> >>>> wrote:
> >>>> Hi,
> >>>>
> >>>> 2013/4/22 Dileepa Jayakody <di...@gmail.com>:
> >>>>
> >>>>> Could it be a valid use-case to integrate WebID protocol in Stanbol
> to
> >>>>> create social graphs and related ontologies?
> >>>>>
> >>>> the already mentioned entity disambiguation for persons might be such
> >>>> a use case.
> >>>>
> >>>> Another idea could be that the enhancement process uses some
> >>>> information from the personal profile of the user who sends the
> >>>> request. I do not have any concrete example at the moment but engines
> >>>> might be interested in knowing who is sending an enhancement request.
> >>>> This may also be a relevant information for the disambiguation task.
> >>>>
> >>>> Best,
> >>>>    - Fabian
> >>>>
> >>>>
> >>>> --
> >>>> Fabian
> >>>> http://twitter.com/fctwitt
> >>>>
> >>>>
> >>
> >> --
> >>
> >> ------------------------------
> >> This message should be regarded as confidential. If you have received
> this
> >> email in error please notify the sender and destroy it immediately.
> >> Statements of intent shall only become binding when confirmed in hard
> copy
> >> by an authorised signatory.
> >>
> >> Zaizi Ltd is registered in England and Wales with the registration
> number
> >> 6440931. The Registered Office is 222 Westbourne Studios, 242 Acklam
> Road,
> >> London W10 5JJ, UK.
> >>
>
>
>
> --
> | Rupert Westenthaler             rupert.westenthaler@gmail.com
> | Bodenlehenstraße 11                             ++43-699-11108907
> | A-5500 Bischofshofen
>

-- 

------------------------------
This message should be regarded as confidential. If you have received this 
email in error please notify the sender and destroy it immediately. 
Statements of intent shall only become binding when confirmed in hard copy 
by an authorised signatory.

Zaizi Ltd is registered in England and Wales with the registration number 
6440931. The Registered Office is 222 Westbourne Studios, 242 Acklam Road, 
London W10 5JJ, UK.

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Rupert Westenthaler <ru...@gmail.com>]

Hi Dileepa, all

Interesting discussion with a lot of very relevant topics. In this
mail I select (1..4) points of the ongoing discussion and add my
thoughts to it. In addition I will add (5) Knowledge base population
as an additional usage case scenario.

(1) Digital Identity in social Networks: I am not a export for the
ontonet, rules and reasoning component, so I will not comment on how
those could be used to process structured information of profiles.
However the Stanbol Enhancer would be well suited to deal with
unstructured information provided by such profiles.

(2) Social Profile of a Person/Organization: Building up a knowledge
base by using information collected from Social Network activities. I
would not recommend to use the Stanbol Entityhub as primary storage
for such a knowledge base. However the Entityhub could be used to
create/store specialized indexes derived from such a knowledge base.
Those indexes would then be used (instead of the knowledge base) by
Enhancement Engines to make lookups for words of the text, Prefix
based suggestion as used by user interfaces ...

(3) Disambiguation #1: With (1) and (2) you will have multiple Social
Profiles using the same name from different information sources (e.g.
Twitter and Facebook). Deciding if those profiles are in fact the same
Entity will need some kind of disambiguation. The result of those
could mean to merge/split profiles in the knowledge base.

(4) Disambiguation #2: Profiles in the Knowledge might be referenced
in texts by the same word/phrase (e.g. similar person names, nick
names, acronyms or organizations ...). Profile information can be used
to ensure that the correct Person/Organization is linked.

(5) Knowledge base population: This would allow (a) Users or (b) an
Enhancement Engine to create new Persons/Organizations in the
knowledge base. An Example for (b) could be detected Named Entities
(NER) that where not linked to an known Entity. In addition an (c)
Entity Disambiguation Engine that comes to the conclusion that none of
the existing Entities is a fit for the mention in the text might also
create an new Entity. (1) and (2) could then be used to populate the
knowledge base with further information originating from social
networks.

Finally I would like to ask how widely used is WebId? Or in other
words: Would it usage limit the list of possible information sources?
Could it be used as unique identifier for data collected in the
knowledge base holding the Social Profile Information of a
Person/Organizations?

best
Rupert

On Tue, Apr 23, 2013 at 10:32 PM, Dileepa Jayakody
<di...@gmail.com> wrote:
> Hi Rafa,
>
>
> Thanks for your ideas.
>
> WebID [1] is a URI based protocol to identify an entity (person,
> organization) uniquely over the web. Since WebID is based on linked-data
> (FOAF [3] syntax) we can extract information such as person's interests,
> employment history, cv, publications, and friends by analyzing the linked
> resources described in the WebID. When a WebID is associated with a person
> it can be used as a single-sign-on protocol like OpenID [2] as well.
>
> WebID can also be instrumental in social-web to have a single-profile over
> multiple social networks with linked resources like photos, friend-lists
> etc. Also security layers can be implemented on top WebID to apply
> fine-grained access control to such linked-resources described by the
> WebID. l
>
> As you have pointed out WebID can be used to implement a disambiguation
> algorithm for people entities in Stanbol. I also think it can be highly
> useful in semantic lifting by extracting information like, employment data,
> friend's profiles, publications etc described by the WebID. IMO sentiment
> analysis can also benefit from WebID.
>
>
> Thanks,
>
> Dileepa
>
> [1] http://www.w3.org/wiki/WebID
>
> [2] http://en.wikipedia.org/wiki/OpenID
> [3] http://xmlns.com/foaf/spec/
>
>
> On Tue, Apr 23, 2013 at 8:37 PM, Rafa Haro <rh...@zaizi.com> wrote:
>
>> Hi Dileepa,
>>
>> El 23/04/13 13:45, Dileepa Jayakody escribió:
>>
>>  Hi Fabian et al,
>>>
>>> Thanks a lot for your valuable ideas.
>>> Yes it's really interesting to implement a 'person | organization'
>>> disambiguation module using WebID protocol as part of Stanbol Enhancement
>>> Engine. I went through the documentation of Stanbol and I have gained an
>>> overall idea about the architecture of Stanbol.
>>>
>> +1. That's a great idea. I don't know very much about WebID protocol, but
>> as far as you could use some profile data as disambiguation contexts, it
>> should be feasible to implement a disambiguation algorithm. Could you
>> please give us a concrete example where WebID is used? I suppose that the
>> general use case is to link name mentions in web pages with digital
>> identities. What kind of information is it possible to gather from WebID
>> identities?
>>
>>
>>> It would be great to get more ideas, suggestions about how to use Stanbol
>>> for people, organization disambiguation and to discuss the objectives and
>>> milestones in the GSOC project idea at [1].
>>> I also think one of the main factor for disambiguation is the
>>> data-set/knowledge base used for the process. What is the data-set Stanbol
>>> uses to verify data? Is Google Wiki-links released recently [1] a
>>> candidate
>>> for Stanbol data-set?
>>>
>> Initially, you can use any knowledge base in Stanbol. I always identify
>> EntityHub component as a "Knowledge Base" management system, although maybe
>> formally the EntityHub is not exactly that. Anyway, Google Wiki-links could
>> be a good resource for disambiguation when the knowledge base is Wikipedia
>> or DBpedia. In fact, Wiki-links contains 40 millions of mentions and its
>> contexts retrieved from web pages. This information can be eventually added
>> to a Wikipedia or DBpedia knowledge base as disambiguation contexts for the
>> entities covered in the dataset. Another interesting resource, as the new
>> in techcrunch points, is the dictionary of Wikipedia concepts released last
>> year [1]. This resource can be used to include more labels for each entity
>> (possible names), improving then the candidate selection step. As always,
>> we face a recall/precision problem with such dictionary.
>>
>> [1]  - http://googleresearch.**blogspot.com.es/2012/05/from-**
>> words-to-concepts-and-back.**html<http://googleresearch.blogspot.com.es/2012/05/from-words-to-concepts-and-back.html>
>>
>> Regards!
>>
>>
>>> Thanks,
>>> Dileepa
>>>
>>> [1]
>>> http://techcrunch.com/2013/03/**08/google-research-releases-**
>>> wikilinks-corpus-with-40m-**mentions-and-3m-entities/<http://techcrunch.com/2013/03/08/google-research-releases-wikilinks-corpus-with-40m-mentions-and-3m-entities/>
>>> On Mon, Apr 22, 2013 at 7:32 PM, Fabian Christ <
>>> christ.fabian@googlemail.com
>>>
>>>> wrote:
>>>> Hi,
>>>>
>>>> 2013/4/22 Dileepa Jayakody <di...@gmail.com>:
>>>>
>>>>> Could it be a valid use-case to integrate WebID protocol in Stanbol to
>>>>> create social graphs and related ontologies?
>>>>>
>>>> the already mentioned entity disambiguation for persons might be such
>>>> a use case.
>>>>
>>>> Another idea could be that the enhancement process uses some
>>>> information from the personal profile of the user who sends the
>>>> request. I do not have any concrete example at the moment but engines
>>>> might be interested in knowing who is sending an enhancement request.
>>>> This may also be a relevant information for the disambiguation task.
>>>>
>>>> Best,
>>>>    - Fabian
>>>>
>>>>
>>>> --
>>>> Fabian
>>>> http://twitter.com/fctwitt
>>>>
>>>>
>>
>> --
>>
>> ------------------------------
>> This message should be regarded as confidential. If you have received this
>> email in error please notify the sender and destroy it immediately.
>> Statements of intent shall only become binding when confirmed in hard copy
>> by an authorised signatory.
>>
>> Zaizi Ltd is registered in England and Wales with the registration number
>> 6440931. The Registered Office is 222 Westbourne Studios, 242 Acklam Road,
>> London W10 5JJ, UK.
>>



-- 
| Rupert Westenthaler             rupert.westenthaler@gmail.com
| Bodenlehenstraße 11                             ++43-699-11108907
| A-5500 Bischofshofen

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Rafa,


Thanks for your ideas.

WebID [1] is a URI based protocol to identify an entity (person,
organization) uniquely over the web. Since WebID is based on linked-data
(FOAF [3] syntax) we can extract information such as person's interests,
employment history, cv, publications, and friends by analyzing the linked
resources described in the WebID. When a WebID is associated with a person
it can be used as a single-sign-on protocol like OpenID [2] as well.

WebID can also be instrumental in social-web to have a single-profile over
multiple social networks with linked resources like photos, friend-lists
etc. Also security layers can be implemented on top WebID to apply
fine-grained access control to such linked-resources described by the
WebID. l

As you have pointed out WebID can be used to implement a disambiguation
algorithm for people entities in Stanbol. I also think it can be highly
useful in semantic lifting by extracting information like, employment data,
friend's profiles, publications etc described by the WebID. IMO sentiment
analysis can also benefit from WebID.


Thanks,

Dileepa

[1] http://www.w3.org/wiki/WebID

[2] http://en.wikipedia.org/wiki/OpenID
[3] http://xmlns.com/foaf/spec/


On Tue, Apr 23, 2013 at 8:37 PM, Rafa Haro <rh...@zaizi.com> wrote:

> Hi Dileepa,
>
> El 23/04/13 13:45, Dileepa Jayakody escribió:
>
>  Hi Fabian et al,
>>
>> Thanks a lot for your valuable ideas.
>> Yes it's really interesting to implement a 'person | organization'
>> disambiguation module using WebID protocol as part of Stanbol Enhancement
>> Engine. I went through the documentation of Stanbol and I have gained an
>> overall idea about the architecture of Stanbol.
>>
> +1. That's a great idea. I don't know very much about WebID protocol, but
> as far as you could use some profile data as disambiguation contexts, it
> should be feasible to implement a disambiguation algorithm. Could you
> please give us a concrete example where WebID is used? I suppose that the
> general use case is to link name mentions in web pages with digital
> identities. What kind of information is it possible to gather from WebID
> identities?
>
>
>> It would be great to get more ideas, suggestions about how to use Stanbol
>> for people, organization disambiguation and to discuss the objectives and
>> milestones in the GSOC project idea at [1].
>> I also think one of the main factor for disambiguation is the
>> data-set/knowledge base used for the process. What is the data-set Stanbol
>> uses to verify data? Is Google Wiki-links released recently [1] a
>> candidate
>> for Stanbol data-set?
>>
> Initially, you can use any knowledge base in Stanbol. I always identify
> EntityHub component as a "Knowledge Base" management system, although maybe
> formally the EntityHub is not exactly that. Anyway, Google Wiki-links could
> be a good resource for disambiguation when the knowledge base is Wikipedia
> or DBpedia. In fact, Wiki-links contains 40 millions of mentions and its
> contexts retrieved from web pages. This information can be eventually added
> to a Wikipedia or DBpedia knowledge base as disambiguation contexts for the
> entities covered in the dataset. Another interesting resource, as the new
> in techcrunch points, is the dictionary of Wikipedia concepts released last
> year [1]. This resource can be used to include more labels for each entity
> (possible names), improving then the candidate selection step. As always,
> we face a recall/precision problem with such dictionary.
>
> [1]  - http://googleresearch.**blogspot.com.es/2012/05/from-**
> words-to-concepts-and-back.**html<http://googleresearch.blogspot.com.es/2012/05/from-words-to-concepts-and-back.html>
>
> Regards!
>
>
>> Thanks,
>> Dileepa
>>
>> [1]
>> http://techcrunch.com/2013/03/**08/google-research-releases-**
>> wikilinks-corpus-with-40m-**mentions-and-3m-entities/<http://techcrunch.com/2013/03/08/google-research-releases-wikilinks-corpus-with-40m-mentions-and-3m-entities/>
>> On Mon, Apr 22, 2013 at 7:32 PM, Fabian Christ <
>> christ.fabian@googlemail.com
>>
>>> wrote:
>>> Hi,
>>>
>>> 2013/4/22 Dileepa Jayakody <di...@gmail.com>:
>>>
>>>> Could it be a valid use-case to integrate WebID protocol in Stanbol to
>>>> create social graphs and related ontologies?
>>>>
>>> the already mentioned entity disambiguation for persons might be such
>>> a use case.
>>>
>>> Another idea could be that the enhancement process uses some
>>> information from the personal profile of the user who sends the
>>> request. I do not have any concrete example at the moment but engines
>>> might be interested in knowing who is sending an enhancement request.
>>> This may also be a relevant information for the disambiguation task.
>>>
>>> Best,
>>>    - Fabian
>>>
>>>
>>> --
>>> Fabian
>>> http://twitter.com/fctwitt
>>>
>>>
>
> --
>
> ------------------------------
> This message should be regarded as confidential. If you have received this
> email in error please notify the sender and destroy it immediately.
> Statements of intent shall only become binding when confirmed in hard copy
> by an authorised signatory.
>
> Zaizi Ltd is registered in England and Wales with the registration number
> 6440931. The Registered Office is 222 Westbourne Studios, 242 Acklam Road,
> London W10 5JJ, UK.
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Rafa Haro <rh...@zaizi.com>]

Hi Dileepa,

El 23/04/13 13:45, Dileepa Jayakody escribió:
> Hi Fabian et al,
>
> Thanks a lot for your valuable ideas.
> Yes it's really interesting to implement a 'person | organization'
> disambiguation module using WebID protocol as part of Stanbol Enhancement
> Engine. I went through the documentation of Stanbol and I have gained an
> overall idea about the architecture of Stanbol.
+1. That's a great idea. I don't know very much about WebID protocol, 
but as far as you could use some profile data as disambiguation 
contexts, it should be feasible to implement a disambiguation algorithm. 
Could you please give us a concrete example where WebID is used? I 
suppose that the general use case is to link name mentions in web pages 
with digital identities. What kind of information is it possible to 
gather from WebID identities?
>
> It would be great to get more ideas, suggestions about how to use Stanbol
> for people, organization disambiguation and to discuss the objectives and
> milestones in the GSOC project idea at [1].
> I also think one of the main factor for disambiguation is the
> data-set/knowledge base used for the process. What is the data-set Stanbol
> uses to verify data? Is Google Wiki-links released recently [1] a candidate
> for Stanbol data-set?
Initially, you can use any knowledge base in Stanbol. I always identify 
EntityHub component as a "Knowledge Base" management system, although 
maybe formally the EntityHub is not exactly that. Anyway, Google 
Wiki-links could be a good resource for disambiguation when the 
knowledge base is Wikipedia or DBpedia. In fact, Wiki-links contains 40 
millions of mentions and its contexts retrieved from web pages. This 
information can be eventually added to a Wikipedia or DBpedia knowledge 
base as disambiguation contexts for the entities covered in the dataset. 
Another interesting resource, as the new in techcrunch points, is the 
dictionary of Wikipedia concepts released last year [1]. This resource 
can be used to include more labels for each entity (possible names), 
improving then the candidate selection step. As always, we face a 
recall/precision problem with such dictionary.

[1]  - 
http://googleresearch.blogspot.com.es/2012/05/from-words-to-concepts-and-back.html

Regards!
>
> Thanks,
> Dileepa
>
> [1]
> http://techcrunch.com/2013/03/08/google-research-releases-wikilinks-corpus-with-40m-mentions-and-3m-entities/
> On Mon, Apr 22, 2013 at 7:32 PM, Fabian Christ <christ.fabian@googlemail.com
>> wrote:
>> Hi,
>>
>> 2013/4/22 Dileepa Jayakody <di...@gmail.com>:
>>> Could it be a valid use-case to integrate WebID protocol in Stanbol to
>>> create social graphs and related ontologies?
>> the already mentioned entity disambiguation for persons might be such
>> a use case.
>>
>> Another idea could be that the enhancement process uses some
>> information from the personal profile of the user who sends the
>> request. I do not have any concrete example at the moment but engines
>> might be interested in knowing who is sending an enhancement request.
>> This may also be a relevant information for the disambiguation task.
>>
>> Best,
>>    - Fabian
>>
>>
>> --
>> Fabian
>> http://twitter.com/fctwitt
>>


-- 

------------------------------
This message should be regarded as confidential. If you have received this 
email in error please notify the sender and destroy it immediately. 
Statements of intent shall only become binding when confirmed in hard copy 
by an authorised signatory.

Zaizi Ltd is registered in England and Wales with the registration number 
6440931. The Registered Office is 222 Westbourne Studios, 242 Acklam Road, 
London W10 5JJ, UK.

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Fabian et al,

Thanks a lot for your valuable ideas.
Yes it's really interesting to implement a 'person | organization'
disambiguation module using WebID protocol as part of Stanbol Enhancement
Engine. I went through the documentation of Stanbol and I have gained an
overall idea about the architecture of Stanbol.

It would be great to get more ideas, suggestions about how to use Stanbol
for people, organization disambiguation and to discuss the objectives and
milestones in the GSOC project idea at [1].
I also think one of the main factor for disambiguation is the
data-set/knowledge base used for the process. What is the data-set Stanbol
uses to verify data? Is Google Wiki-links released recently [1] a candidate
for Stanbol data-set?

Thanks,
Dileepa

[1]
http://techcrunch.com/2013/03/08/google-research-releases-wikilinks-corpus-with-40m-mentions-and-3m-entities/
On Mon, Apr 22, 2013 at 7:32 PM, Fabian Christ <christ.fabian@googlemail.com
> wrote:

> Hi,
>
> 2013/4/22 Dileepa Jayakody <di...@gmail.com>:
> > Could it be a valid use-case to integrate WebID protocol in Stanbol to
> > create social graphs and related ontologies?
>
> the already mentioned entity disambiguation for persons might be such
> a use case.
>
> Another idea could be that the enhancement process uses some
> information from the personal profile of the user who sends the
> request. I do not have any concrete example at the moment but engines
> might be interested in knowing who is sending an enhancement request.
> This may also be a relevant information for the disambiguation task.
>
> Best,
>   - Fabian
>
>
> --
> Fabian
> http://twitter.com/fctwitt
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Fabian Christ <ch...@googlemail.com>]

Hi,

2013/4/22 Dileepa Jayakody <di...@gmail.com>:
> Could it be a valid use-case to integrate WebID protocol in Stanbol to
> create social graphs and related ontologies?

the already mentioned entity disambiguation for persons might be such
a use case.

Another idea could be that the enhancement process uses some
information from the personal profile of the user who sends the
request. I do not have any concrete example at the moment but engines
might be interested in knowing who is sending an enhancement request.
This may also be a relevant information for the disambiguation task.

Best,
  - Fabian


--
Fabian
http://twitter.com/fctwitt

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Ed, Reto,

Thanks for the inputs..WebID protocol seems to be an excellent fit for what
I'm looking at.
Does Stanbol support WebID protocol? AFAIK it doesn't support WebID yet.
Could it be a valid use-case to integrate WebID protocol in Stanbol to
create social graphs and related ontologies?

Thanks,
Dileepa


On Mon, Apr 22, 2013 at 6:30 AM, Reto Bachmann-Gmür <re...@wymiwyg.com>wrote:

> I also think such a project would be very useful combined with a
> (re)activation of WebId and possibly OpenId authentication.
>
> Cheers,
> Reto
>
>
> On Sun, Apr 21, 2013 at 10:56 PM, Ed - 0x1b, Inc. <st...@0x1b.com>
> wrote:
>
> > Dileepa,
> >
> > You may want to look into the W3C WebID effort - I would recommend
> > this as a good tool for social/semantic identity.
> > www.w3.org/wiki/WebID
> > Ed
> >
> > On Sun, Apr 21, 2013 at 1:07 PM, Dileepa Jayakody
> > <di...@gmail.com> wrote:
> > > Hi Andreas and all,
> > >
> > > Thanks a lot for your feedback, I'm definitely open to other's
> > > ideas/suggestions as well. Please feel free to forward the mail to
> other
> > > lists and let me know (I would love to suscribe to other related
> lists..)
> > > My idea about the research project on *'verified digital identity'* is
> as
> > > below.
> > >
> > > We need a verified identity of people we interact with on the internet.
> > > There maybe so many false profiles, impersonating real
> > people/organization
> > > and we need to be sure that we are interacting with the real person.
> For
> > > example I want to be sure that the person who has  sent me a friend
> > request
> > > on FB, or an email is the real physical person I meet f2f.
> > > IMO there are several methods we can use to verify the identity of a
> > > person. We can use a verified identity provider or we can adapt a
> > > trust-model based on social networks/social activity of the person to
> > > verify his/her digital identity based on social proof concepts.
> > > To build a trust-model based on online social interactions, we can use
> > > semantic web techniques and I think FOAF, RDFa can be very useful in
> > > building such a framework. By analyzing the social activity streams,
> > social
> > > graph and network of the person we can verify the identity.
> > > As you have mentioned the privacy implications in such a system are
> > > important. We might need to implement an access control mechanism over
> > the
> > > data exchanging via the system to protect privacy. Also as you have
> > > suggested the trust-model can comprise of a reputation calculation
> module
> > > based on the feedback a person gets regarding his identity over the
> web.
> > >
> > > So above is my general idea of the research project I'm looking at. I
> > think
> > > Stanbol, with it's features like Enhancer, Ontology Manager and
> Reasoner
> > is
> > > one of the best choices to implement such a framework to verify digital
> > > identity over the web. Such a system to analyse social data can be
> > further
> > > improved to implement a perception analysis engine for people. :)
> > >
> > > On a related note, I'm currently going through the documentation of
> > > Stanbol. While groing through I came across a typo mistake and a
> invailid
> > > link, wanted to point it out :).
> > > Created a jira for this at
> > > https://issues.apache.org/jira/browse/STANBOL-1050
> > >
> > > Thanks,
> > > Dileepa
> > >
> > >
> > > On Sun, Apr 21, 2013 at 10:33 PM, Andreas Kuckartz <A.Kuckartz@ping.de
> > >wrote:
> > >
> > >> Hi Dileepa,
> > >>
> > >> thanks a lot for your mail.
> > >>
> > >> At the moment I only want to provide a quick but brief feedback.
> > >>
> > >> What do others think?
> > >>
> > >> Your project idea definitely fits the general project idea formulated
> in
> > >> STANBOL-1009.
> > >>
> > >> Can you tel us a bit more about the use case? Who would use such a
> > feature?
> > >>
> > >> Which implications would there be regarding privacy? Would it be
> > >> possible to use such a feature to improve privacy (for example by
> > >> providing feedback to a user regaring his or her identifiability)?
> > >>
> > >> I would like to forward your mail to two other (public) mailing lists
> to
> > >> ask for suggestions. Would that be ok with you?
> > >>
> > >> Cheers,
> > >> Andreas
> > >> ---
> > >>
> > >> Dileepa Jayakody:
> > >> > Hi Devs,
> > >> >
> > >> > I'm Dileepa Jayakody a research student keen on distributed
> computing
> > >> > domain, particularly interested in social networking, social
> identity,
> > >> > semantic web concepts. I have started studying for a M.Sc by
> Research
> > at
> > >> > University of Moratuwa and recently joined LK Domain Registry as a
> > >> research
> > >> > assistant. Currently I'm assigned to the 'Digital Identity' research
> > >> group,
> > >> > and now doing a literature survey on Digital Identity and identity
> > >> > verification techniques.  My objective is to develop a methodology
> to
> > >> > verify the identity of a person/organization by analyzing profiles,
> > >> social
> > >> > activity over different social networking streams, and other
> verified
> > >> > data-streams. I'm looking at technologies like RDFa, JSON-LD, FOAF,
> > nosql
> > >> > databases and other semantic tools which could be helpful to build a
> > >> > suitable solution. While researching through the web, I came across
> > >> Apache
> > >> > Stanbol project and I'm highly interested in learning about Stanbol
> > for
> > >> my
> > >> > research project. I'm also fascinated by the GSOC idea on using
> > Stanbol
> > >> for
> > >> > federated social web [2]. It would  be great to get ideas on how to
> > >> > integrate Stanbol to implement a system to verify digital identity
> of
> > a
> > >> > person/organization over social networking streams.
> > >> >
> > >> > To tell a bit more about myself, prior to joining LK Domain Registry
> > as a
> > >> > research assistant, I was a software engineer for 1.5 years at WSO2
> > (Pvt)
> > >> > Ltd [1], the opensource SOA company where I was a member of the
> carbon
> > >> team
> > >> > which is the kernel of the WSO2 Product stack which are OSGi based
> > >> > products. I have a good development background on java, web
> services,
> > >> OSGi
> > >> > and maven. I also have a good understanding about the opensource
> > >> > development process and culture as WSO2 follows the apache model for
> > >> > software development. I have also participated in OSGi Community
> Event
> > >> 2012
> > >> > co-located with EclipseCon 2012 by giving a tech.talk on "Building A
> > >> > Modular Server Platform with OSGi" [3].
> > >> >
> > >> > Currently I'm going through Stanbol documentation at [4] and
> building
> > >> > Stanbol from trunk source.
> > >> > Looking forward to your feedback.
> > >> >
> > >> > Thanks and regards,
> > >> > Dileepa
> > >> >
> > >> > [1] http://www.wso2.com
> > >> > [2] https://issues.apache.org/jira/browse/STANBOL-1009
> > >> > [3]
> > >> >
> > >>
> >
> http://www.eclipsecon.org/europe2012/sessions/building-server-platform-osgi
> > >> > [4] http://stanbol.apache.org/docs/trunk/
> > >> >
> > >>
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Reto Bachmann-Gmür <re...@wymiwyg.com>]

I also think such a project would be very useful combined with a
(re)activation of WebId and possibly OpenId authentication.

Cheers,
Reto


On Sun, Apr 21, 2013 at 10:56 PM, Ed - 0x1b, Inc. <st...@0x1b.com> wrote:

> Dileepa,
>
> You may want to look into the W3C WebID effort - I would recommend
> this as a good tool for social/semantic identity.
> www.w3.org/wiki/WebID
> Ed
>
> On Sun, Apr 21, 2013 at 1:07 PM, Dileepa Jayakody
> <di...@gmail.com> wrote:
> > Hi Andreas and all,
> >
> > Thanks a lot for your feedback, I'm definitely open to other's
> > ideas/suggestions as well. Please feel free to forward the mail to other
> > lists and let me know (I would love to suscribe to other related lists..)
> > My idea about the research project on *'verified digital identity'* is as
> > below.
> >
> > We need a verified identity of people we interact with on the internet.
> > There maybe so many false profiles, impersonating real
> people/organization
> > and we need to be sure that we are interacting with the real person. For
> > example I want to be sure that the person who has  sent me a friend
> request
> > on FB, or an email is the real physical person I meet f2f.
> > IMO there are several methods we can use to verify the identity of a
> > person. We can use a verified identity provider or we can adapt a
> > trust-model based on social networks/social activity of the person to
> > verify his/her digital identity based on social proof concepts.
> > To build a trust-model based on online social interactions, we can use
> > semantic web techniques and I think FOAF, RDFa can be very useful in
> > building such a framework. By analyzing the social activity streams,
> social
> > graph and network of the person we can verify the identity.
> > As you have mentioned the privacy implications in such a system are
> > important. We might need to implement an access control mechanism over
> the
> > data exchanging via the system to protect privacy. Also as you have
> > suggested the trust-model can comprise of a reputation calculation module
> > based on the feedback a person gets regarding his identity over the web.
> >
> > So above is my general idea of the research project I'm looking at. I
> think
> > Stanbol, with it's features like Enhancer, Ontology Manager and Reasoner
> is
> > one of the best choices to implement such a framework to verify digital
> > identity over the web. Such a system to analyse social data can be
> further
> > improved to implement a perception analysis engine for people. :)
> >
> > On a related note, I'm currently going through the documentation of
> > Stanbol. While groing through I came across a typo mistake and a invailid
> > link, wanted to point it out :).
> > Created a jira for this at
> > https://issues.apache.org/jira/browse/STANBOL-1050
> >
> > Thanks,
> > Dileepa
> >
> >
> > On Sun, Apr 21, 2013 at 10:33 PM, Andreas Kuckartz <A.Kuckartz@ping.de
> >wrote:
> >
> >> Hi Dileepa,
> >>
> >> thanks a lot for your mail.
> >>
> >> At the moment I only want to provide a quick but brief feedback.
> >>
> >> What do others think?
> >>
> >> Your project idea definitely fits the general project idea formulated in
> >> STANBOL-1009.
> >>
> >> Can you tel us a bit more about the use case? Who would use such a
> feature?
> >>
> >> Which implications would there be regarding privacy? Would it be
> >> possible to use such a feature to improve privacy (for example by
> >> providing feedback to a user regaring his or her identifiability)?
> >>
> >> I would like to forward your mail to two other (public) mailing lists to
> >> ask for suggestions. Would that be ok with you?
> >>
> >> Cheers,
> >> Andreas
> >> ---
> >>
> >> Dileepa Jayakody:
> >> > Hi Devs,
> >> >
> >> > I'm Dileepa Jayakody a research student keen on distributed computing
> >> > domain, particularly interested in social networking, social identity,
> >> > semantic web concepts. I have started studying for a M.Sc by Research
> at
> >> > University of Moratuwa and recently joined LK Domain Registry as a
> >> research
> >> > assistant. Currently I'm assigned to the 'Digital Identity' research
> >> group,
> >> > and now doing a literature survey on Digital Identity and identity
> >> > verification techniques.  My objective is to develop a methodology to
> >> > verify the identity of a person/organization by analyzing profiles,
> >> social
> >> > activity over different social networking streams, and other verified
> >> > data-streams. I'm looking at technologies like RDFa, JSON-LD, FOAF,
> nosql
> >> > databases and other semantic tools which could be helpful to build a
> >> > suitable solution. While researching through the web, I came across
> >> Apache
> >> > Stanbol project and I'm highly interested in learning about Stanbol
> for
> >> my
> >> > research project. I'm also fascinated by the GSOC idea on using
> Stanbol
> >> for
> >> > federated social web [2]. It would  be great to get ideas on how to
> >> > integrate Stanbol to implement a system to verify digital identity of
> a
> >> > person/organization over social networking streams.
> >> >
> >> > To tell a bit more about myself, prior to joining LK Domain Registry
> as a
> >> > research assistant, I was a software engineer for 1.5 years at WSO2
> (Pvt)
> >> > Ltd [1], the opensource SOA company where I was a member of the carbon
> >> team
> >> > which is the kernel of the WSO2 Product stack which are OSGi based
> >> > products. I have a good development background on java, web services,
> >> OSGi
> >> > and maven. I also have a good understanding about the opensource
> >> > development process and culture as WSO2 follows the apache model for
> >> > software development. I have also participated in OSGi Community Event
> >> 2012
> >> > co-located with EclipseCon 2012 by giving a tech.talk on "Building A
> >> > Modular Server Platform with OSGi" [3].
> >> >
> >> > Currently I'm going through Stanbol documentation at [4] and building
> >> > Stanbol from trunk source.
> >> > Looking forward to your feedback.
> >> >
> >> > Thanks and regards,
> >> > Dileepa
> >> >
> >> > [1] http://www.wso2.com
> >> > [2] https://issues.apache.org/jira/browse/STANBOL-1009
> >> > [3]
> >> >
> >>
> http://www.eclipsecon.org/europe2012/sessions/building-server-platform-osgi
> >> > [4] http://stanbol.apache.org/docs/trunk/
> >> >
> >>
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

["Ed - 0x1b, Inc." <st...@0x1b.com>]

Dileepa,

You may want to look into the W3C WebID effort - I would recommend
this as a good tool for social/semantic identity.
www.w3.org/wiki/WebID
Ed

On Sun, Apr 21, 2013 at 1:07 PM, Dileepa Jayakody
<di...@gmail.com> wrote:
> Hi Andreas and all,
>
> Thanks a lot for your feedback, I'm definitely open to other's
> ideas/suggestions as well. Please feel free to forward the mail to other
> lists and let me know (I would love to suscribe to other related lists..)
> My idea about the research project on *'verified digital identity'* is as
> below.
>
> We need a verified identity of people we interact with on the internet.
> There maybe so many false profiles, impersonating real people/organization
> and we need to be sure that we are interacting with the real person. For
> example I want to be sure that the person who has  sent me a friend request
> on FB, or an email is the real physical person I meet f2f.
> IMO there are several methods we can use to verify the identity of a
> person. We can use a verified identity provider or we can adapt a
> trust-model based on social networks/social activity of the person to
> verify his/her digital identity based on social proof concepts.
> To build a trust-model based on online social interactions, we can use
> semantic web techniques and I think FOAF, RDFa can be very useful in
> building such a framework. By analyzing the social activity streams, social
> graph and network of the person we can verify the identity.
> As you have mentioned the privacy implications in such a system are
> important. We might need to implement an access control mechanism over the
> data exchanging via the system to protect privacy. Also as you have
> suggested the trust-model can comprise of a reputation calculation module
> based on the feedback a person gets regarding his identity over the web.
>
> So above is my general idea of the research project I'm looking at. I think
> Stanbol, with it's features like Enhancer, Ontology Manager and Reasoner is
> one of the best choices to implement such a framework to verify digital
> identity over the web. Such a system to analyse social data can be further
> improved to implement a perception analysis engine for people. :)
>
> On a related note, I'm currently going through the documentation of
> Stanbol. While groing through I came across a typo mistake and a invailid
> link, wanted to point it out :).
> Created a jira for this at
> https://issues.apache.org/jira/browse/STANBOL-1050
>
> Thanks,
> Dileepa
>
>
> On Sun, Apr 21, 2013 at 10:33 PM, Andreas Kuckartz <A....@ping.de>wrote:
>
>> Hi Dileepa,
>>
>> thanks a lot for your mail.
>>
>> At the moment I only want to provide a quick but brief feedback.
>>
>> What do others think?
>>
>> Your project idea definitely fits the general project idea formulated in
>> STANBOL-1009.
>>
>> Can you tel us a bit more about the use case? Who would use such a feature?
>>
>> Which implications would there be regarding privacy? Would it be
>> possible to use such a feature to improve privacy (for example by
>> providing feedback to a user regaring his or her identifiability)?
>>
>> I would like to forward your mail to two other (public) mailing lists to
>> ask for suggestions. Would that be ok with you?
>>
>> Cheers,
>> Andreas
>> ---
>>
>> Dileepa Jayakody:
>> > Hi Devs,
>> >
>> > I'm Dileepa Jayakody a research student keen on distributed computing
>> > domain, particularly interested in social networking, social identity,
>> > semantic web concepts. I have started studying for a M.Sc by Research at
>> > University of Moratuwa and recently joined LK Domain Registry as a
>> research
>> > assistant. Currently I'm assigned to the 'Digital Identity' research
>> group,
>> > and now doing a literature survey on Digital Identity and identity
>> > verification techniques.  My objective is to develop a methodology to
>> > verify the identity of a person/organization by analyzing profiles,
>> social
>> > activity over different social networking streams, and other verified
>> > data-streams. I'm looking at technologies like RDFa, JSON-LD, FOAF, nosql
>> > databases and other semantic tools which could be helpful to build a
>> > suitable solution. While researching through the web, I came across
>> Apache
>> > Stanbol project and I'm highly interested in learning about Stanbol for
>> my
>> > research project. I'm also fascinated by the GSOC idea on using Stanbol
>> for
>> > federated social web [2]. It would  be great to get ideas on how to
>> > integrate Stanbol to implement a system to verify digital identity of a
>> > person/organization over social networking streams.
>> >
>> > To tell a bit more about myself, prior to joining LK Domain Registry as a
>> > research assistant, I was a software engineer for 1.5 years at WSO2 (Pvt)
>> > Ltd [1], the opensource SOA company where I was a member of the carbon
>> team
>> > which is the kernel of the WSO2 Product stack which are OSGi based
>> > products. I have a good development background on java, web services,
>> OSGi
>> > and maven. I also have a good understanding about the opensource
>> > development process and culture as WSO2 follows the apache model for
>> > software development. I have also participated in OSGi Community Event
>> 2012
>> > co-located with EclipseCon 2012 by giving a tech.talk on "Building A
>> > Modular Server Platform with OSGi" [3].
>> >
>> > Currently I'm going through Stanbol documentation at [4] and building
>> > Stanbol from trunk source.
>> > Looking forward to your feedback.
>> >
>> > Thanks and regards,
>> > Dileepa
>> >
>> > [1] http://www.wso2.com
>> > [2] https://issues.apache.org/jira/browse/STANBOL-1009
>> > [3]
>> >
>> http://www.eclipsecon.org/europe2012/sessions/building-server-platform-osgi
>> > [4] http://stanbol.apache.org/docs/trunk/
>> >
>>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Dileepa Jayakody <di...@gmail.com>]

Hi Andreas and all,

Thanks a lot for your feedback, I'm definitely open to other's
ideas/suggestions as well. Please feel free to forward the mail to other
lists and let me know (I would love to suscribe to other related lists..)
My idea about the research project on *'verified digital identity'* is as
below.

We need a verified identity of people we interact with on the internet.
There maybe so many false profiles, impersonating real people/organization
and we need to be sure that we are interacting with the real person. For
example I want to be sure that the person who has  sent me a friend request
on FB, or an email is the real physical person I meet f2f.
IMO there are several methods we can use to verify the identity of a
person. We can use a verified identity provider or we can adapt a
trust-model based on social networks/social activity of the person to
verify his/her digital identity based on social proof concepts.
To build a trust-model based on online social interactions, we can use
semantic web techniques and I think FOAF, RDFa can be very useful in
building such a framework. By analyzing the social activity streams, social
graph and network of the person we can verify the identity.
As you have mentioned the privacy implications in such a system are
important. We might need to implement an access control mechanism over the
data exchanging via the system to protect privacy. Also as you have
suggested the trust-model can comprise of a reputation calculation module
based on the feedback a person gets regarding his identity over the web.

So above is my general idea of the research project I'm looking at. I think
Stanbol, with it's features like Enhancer, Ontology Manager and Reasoner is
one of the best choices to implement such a framework to verify digital
identity over the web. Such a system to analyse social data can be further
improved to implement a perception analysis engine for people. :)

On a related note, I'm currently going through the documentation of
Stanbol. While groing through I came across a typo mistake and a invailid
link, wanted to point it out :).
Created a jira for this at
https://issues.apache.org/jira/browse/STANBOL-1050

Thanks,
Dileepa


On Sun, Apr 21, 2013 at 10:33 PM, Andreas Kuckartz <A....@ping.de>wrote:

> Hi Dileepa,
>
> thanks a lot for your mail.
>
> At the moment I only want to provide a quick but brief feedback.
>
> What do others think?
>
> Your project idea definitely fits the general project idea formulated in
> STANBOL-1009.
>
> Can you tel us a bit more about the use case? Who would use such a feature?
>
> Which implications would there be regarding privacy? Would it be
> possible to use such a feature to improve privacy (for example by
> providing feedback to a user regaring his or her identifiability)?
>
> I would like to forward your mail to two other (public) mailing lists to
> ask for suggestions. Would that be ok with you?
>
> Cheers,
> Andreas
> ---
>
> Dileepa Jayakody:
> > Hi Devs,
> >
> > I'm Dileepa Jayakody a research student keen on distributed computing
> > domain, particularly interested in social networking, social identity,
> > semantic web concepts. I have started studying for a M.Sc by Research at
> > University of Moratuwa and recently joined LK Domain Registry as a
> research
> > assistant. Currently I'm assigned to the 'Digital Identity' research
> group,
> > and now doing a literature survey on Digital Identity and identity
> > verification techniques.  My objective is to develop a methodology to
> > verify the identity of a person/organization by analyzing profiles,
> social
> > activity over different social networking streams, and other verified
> > data-streams. I'm looking at technologies like RDFa, JSON-LD, FOAF, nosql
> > databases and other semantic tools which could be helpful to build a
> > suitable solution. While researching through the web, I came across
> Apache
> > Stanbol project and I'm highly interested in learning about Stanbol for
> my
> > research project. I'm also fascinated by the GSOC idea on using Stanbol
> for
> > federated social web [2]. It would  be great to get ideas on how to
> > integrate Stanbol to implement a system to verify digital identity of a
> > person/organization over social networking streams.
> >
> > To tell a bit more about myself, prior to joining LK Domain Registry as a
> > research assistant, I was a software engineer for 1.5 years at WSO2 (Pvt)
> > Ltd [1], the opensource SOA company where I was a member of the carbon
> team
> > which is the kernel of the WSO2 Product stack which are OSGi based
> > products. I have a good development background on java, web services,
> OSGi
> > and maven. I also have a good understanding about the opensource
> > development process and culture as WSO2 follows the apache model for
> > software development. I have also participated in OSGi Community Event
> 2012
> > co-located with EclipseCon 2012 by giving a tech.talk on "Building A
> > Modular Server Platform with OSGi" [3].
> >
> > Currently I'm going through Stanbol documentation at [4] and building
> > Stanbol from trunk source.
> > Looking forward to your feedback.
> >
> > Thanks and regards,
> > Dileepa
> >
> > [1] http://www.wso2.com
> > [2] https://issues.apache.org/jira/browse/STANBOL-1009
> > [3]
> >
> http://www.eclipsecon.org/europe2012/sessions/building-server-platform-osgi
> > [4] http://stanbol.apache.org/docs/trunk/
> >
>

Re: [GSOC Idea] Apache Stanbol to verify digital identity over social web

[Andreas Kuckartz <A....@ping.de>]

Hi Dileepa,

thanks a lot for your mail.

At the moment I only want to provide a quick but brief feedback.

What do others think?

Your project idea definitely fits the general project idea formulated in
STANBOL-1009.

Can you tel us a bit more about the use case? Who would use such a feature?

Which implications would there be regarding privacy? Would it be
possible to use such a feature to improve privacy (for example by
providing feedback to a user regaring his or her identifiability)?

I would like to forward your mail to two other (public) mailing lists to
ask for suggestions. Would that be ok with you?

Cheers,
Andreas
---

Dileepa Jayakody:
> Hi Devs,
> 
> I'm Dileepa Jayakody a research student keen on distributed computing
> domain, particularly interested in social networking, social identity,
> semantic web concepts. I have started studying for a M.Sc by Research at
> University of Moratuwa and recently joined LK Domain Registry as a research
> assistant. Currently I'm assigned to the 'Digital Identity' research group,
> and now doing a literature survey on Digital Identity and identity
> verification techniques.  My objective is to develop a methodology to
> verify the identity of a person/organization by analyzing profiles, social
> activity over different social networking streams, and other verified
> data-streams. I'm looking at technologies like RDFa, JSON-LD, FOAF, nosql
> databases and other semantic tools which could be helpful to build a
> suitable solution. While researching through the web, I came across Apache
> Stanbol project and I'm highly interested in learning about Stanbol for my
> research project. I'm also fascinated by the GSOC idea on using Stanbol for
> federated social web [2]. It would  be great to get ideas on how to
> integrate Stanbol to implement a system to verify digital identity of a
> person/organization over social networking streams.
> 
> To tell a bit more about myself, prior to joining LK Domain Registry as a
> research assistant, I was a software engineer for 1.5 years at WSO2 (Pvt)
> Ltd [1], the opensource SOA company where I was a member of the carbon team
> which is the kernel of the WSO2 Product stack which are OSGi based
> products. I have a good development background on java, web services, OSGi
> and maven. I also have a good understanding about the opensource
> development process and culture as WSO2 follows the apache model for
> software development. I have also participated in OSGi Community Event 2012
> co-located with EclipseCon 2012 by giving a tech.talk on "Building A
> Modular Server Platform with OSGi" [3].
> 
> Currently I'm going through Stanbol documentation at [4] and building
> Stanbol from trunk source.
> Looking forward to your feedback.
> 
> Thanks and regards,
> Dileepa
> 
> [1] http://www.wso2.com
> [2] https://issues.apache.org/jira/browse/STANBOL-1009
> [3]
> http://www.eclipsecon.org/europe2012/sessions/building-server-platform-osgi
> [4] http://stanbol.apache.org/docs/trunk/
>