You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by John Gant <jo...@gmail.com> on 2005/08/13 18:01:22 UTC
commons math
Hello,
I am currently a graduate student in Computer Science and Computer
Engineering at the University of Louisville, Kentucky. First let me
congratulate the group of developers who commit and architect for
apache commons. I have used many of the libraries and they are all of
excellent quality (but I guess you already know that :)). I am
interested in contributing to open source software and have interests
that are in the domain of statistics with a focus in data mining.
After writing many algorithms for classes, and asking an apache
contributor if any of this would be needed elsewhere, he told me to
purpose something to the dev list. So here it goes, I would like to
help start a data mining section of commons math and advance the
existing statistical libraries. I plan on developing the algorithms
for personal use anyway, and would like to see some of my work be used
by others. If anyone is interested we can continue this thread and I
will email my code, and purpose my new algorithms.
John Gant
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by Lukas Vlcek <lu...@gmail.com>.
Hi John,
As I said I am not an active commited to [math] and I probably can't
speak on behalf of this community but to me this seems interesting. I
will look at it in deeper detail.
Thanks!
Lukas
On 8/16/05, John Gant <jo...@gmail.com> wrote:
> Sorry for the late link to code, and paper. Please note that this code
> is not in a form that I would even think about commiting. I plan to
> restructure all of the code and add unit tests, including all
> algorithms. The best way to describe the current code is 'student
> form', meaning there are file loading methods and other utility
> methods throughout the files. The code was written in haste, and will
> be heavily refactored regardless.
>
> Read at your own risk :)
> Thanks Hen for the space!
> http://people.apache.org/~bayard/jgant/
>
> Thanks,
> John
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by John Gant <jo...@gmail.com>.
Sorry for the late link to code, and paper. Please note that this code
is not in a form that I would even think about commiting. I plan to
restructure all of the code and add unit tests, including all
algorithms. The best way to describe the current code is 'student
form', meaning there are file loading methods and other utility
methods throughout the files. The code was written in haste, and will
be heavily refactored regardless.
Read at your own risk :)
Thanks Hen for the space!
http://people.apache.org/~bayard/jgant/
Thanks,
John
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by Kim van der Linde <ki...@kimvdlinde.com>.
In the past, I have made matrix based distance and special Matrices
classes that can be adapted to do just these things. It contains SSCP,
covariance and correlation matrices and rowbased Euclidian and Mahalanobis
distances. An idea?
Cheers,
Kim
John Gant said:
> What exactly does "column-wise" mean. This just looks like Pearson's
>> R, which is already available in the SimpleRegression class. Do you
>> mean generation of correlation matrices?
>
> Sorry, I should have been more specific. This will allow someone to
> calculate the pearson r coefficient between column vectors. This
> results in a correlation matrix with dimensions (c * c), where c is
> the number of columns in the raw data matrix.
>
>> > Distance measures, are basically a numeric way of classifying a
>> > relationship between two numerical or categorical datasets. Usually
>> > distance measures are used in conjunction with k-means, or
>> > hierarchical clustering (or some type of clustering algorithm).
>
>> Are these essentially metrics on R^n (the "numerical" case) or
>> homogeneity measures (e.g. chi-square, for the categorical case)?
>
> The numerical distance measures can either be something as simple as
> euclidean distance, or correlation cofficient. The categorical
> measures are more logical (less numerical), and something like hamming
> distance could be used. Does this answer your question?
>
>> If a clustering algorithm can use mutlitple different distance
>> measures, then it does make sense to encapsulate the distance measure.
>> Defining a distance measure or metric interface and then defining
>> implementation classes that implement that interface and having the
>> clustering algorithms have instances of these as members is a
>> reasonable way to do this, IMHO.
>
>
> A clustering algorithm is usually independent of the distance measure,
> but relies on this measure to identify clusters. All clustering
> algorithms (that I have experience with) use distance measures, and I
> plan on setting up the implementation so that it is similar to the
> contract of Collections.sort(). I have generated an interface,
> DistanceMeasure, which has only a method calculateDistance(). This
> interface, currently, is implemented in the EulcideanDistance class. I
> have not posted this code, and need to finish the unit tests.
>
> Thanks,
> John
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
--
http://www.kimvdlinde.com
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by John Gant <jo...@gmail.com>.
What exactly does "column-wise" mean. This just looks like Pearson's
> R, which is already available in the SimpleRegression class. Do you
> mean generation of correlation matrices?
Sorry, I should have been more specific. This will allow someone to
calculate the pearson r coefficient between column vectors. This
results in a correlation matrix with dimensions (c * c), where c is
the number of columns in the raw data matrix.
> > Distance measures, are basically a numeric way of classifying a
> > relationship between two numerical or categorical datasets. Usually
> > distance measures are used in conjunction with k-means, or
> > hierarchical clustering (or some type of clustering algorithm).
> Are these essentially metrics on R^n (the "numerical" case) or
> homogeneity measures (e.g. chi-square, for the categorical case)?
The numerical distance measures can either be something as simple as
euclidean distance, or correlation cofficient. The categorical
measures are more logical (less numerical), and something like hamming
distance could be used. Does this answer your question?
> If a clustering algorithm can use mutlitple different distance
> measures, then it does make sense to encapsulate the distance measure.
> Defining a distance measure or metric interface and then defining
> implementation classes that implement that interface and having the
> clustering algorithms have instances of these as members is a
> reasonable way to do this, IMHO.
A clustering algorithm is usually independent of the distance measure,
but relies on this measure to identify clusters. All clustering
algorithms (that I have experience with) use distance measures, and I
plan on setting up the implementation so that it is similar to the
contract of Collections.sort(). I have generated an interface,
DistanceMeasure, which has only a method calculateDistance(). This
interface, currently, is implemented in the EulcideanDistance class. I
have not posted this code, and need to finish the unit tests.
Thanks,
John
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by Phil Steitz <ph...@gmail.com>.
Sorry for the response latency... See comments interspersed.
On 8/15/05, John Gant <jo...@gmail.com> wrote:
> IP stuff:
> I will send out a link to the pdf that describes KMotif,
Thanks.
and the cross
> correlation comes from
> http://mathworld.wolfram.com/CorrelationCoefficient.html with an
> implementation that correlates column-wise.
What exactly does "column-wise" mean. This just looks like Pearson's
R, which is already available in the SimpleRegression class. Do you
mean generation of correlation matrices?
Saw you just posted some code. Thanks! I will have a look :-)
> Both euclidean and
> city-block distance measures come from basic data mining textbooks (my
> textbook is Data Mining by Mehmed Kantardzic) or
> http://www.statsoft.com/textbook/stcluan.html. Please let me know if
> this is sufficient, or if I need more references.
This is sufficient. Online references are good if you can find stable
links (like above).
>
> Distance measures, are basically a numeric way of classifying a
> relationship between two numerical or categorical datasets. Usually
> distance measures are used in conjunction with k-means, or
> hierarchical clustering (or some type of clustering algorithm).
Are these essentially metrics on R^n (the "numerical" case) or
homogeneity measures (e.g. chi-square, for the categorical case)?
>
> I think the architecture question applies to K-means and
> difference/similarity algorithms. I am not sure of the best
> architecture for these algorithms. Should each distance/similarity
> measure be its own class, allowing these to be passed into an engine
> that is the clustering algorithm?
If the algorithms can make use of "pluggable' distance measures, then
yes, this would make sense.
> For instance have a k-means class
> who has a private variable of type ClusertingMeasurementAlgorithm,
> where:
>
> EuclideanDistance which implements,
> DistanceMeasure which implements,
> ClusteringMeasurementAlgorithm
>
> Does this sound somewhat logical? If we had an engine that took an
> instance of ClusteringMeasurementAlgorithm as a constructor parameter,
> it could handle all operations on the data using the specific
> measurement algorithm.
I am confused about what is being abstracted here. If it is the
distance measure, the interface should be called something that ends
in "Measure" or "Metric"
> The reason I am trying to abstract the
> clustering algorithm more than a difference measure is due to the fact
> that clustering may be done on similiarity and difference measures.
> Please tell me if this sounds outrageous, because I do not have alot
> of architecture experience.
If a clustering algorithm can use mutlitple different distance
measures, then it does make sense to encapsulate the distance measure.
Defining a distance measure or metric interface and then defining
implementation classes that implement that interface and having the
clustering algorithms have instances of these as members is a
reasonable way to do this, IMHO.
Phil
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by John Gant <jo...@gmail.com>.
IP stuff:
I will send out a link to the pdf that describes KMotif, and the cross
correlation comes from
http://mathworld.wolfram.com/CorrelationCoefficient.html with an
implementation that correlates column-wise. Both euclidean and
city-block distance measures come from basic data mining textbooks (my
textbook is Data Mining by Mehmed Kantardzic) or
http://www.statsoft.com/textbook/stcluan.html. Please let me know if
this is sufficient, or if I need more references.
Distance measures, are basically a numeric way of classifying a
relationship between two numerical or categorical datasets. Usually
distance measures are used in conjunction with k-means, or
hierarchical clustering (or some type of clustering algorithm).
I think the architecture question applies to K-means and
difference/similarity algorithms. I am not sure of the best
architecture for these algorithms. Should each distance/similarity
measure be its own class, allowing these to be passed into an engine
that is the clustering algorithm? For instance have a k-means class
who has a private variable of type ClusertingMeasurementAlgorithm,
where:
EuclideanDistance which implements,
DistanceMeasure which implements,
ClusteringMeasurementAlgorithm
Does this sound somewhat logical? If we had an engine that took an
instance of ClusteringMeasurementAlgorithm as a constructor parameter,
it could handle all operations on the data using the specific
measurement algorithm. The reason I am trying to abstract the
clustering algorithm more than a difference measure is due to the fact
that clustering may be done on similiarity and difference measures.
Please tell me if this sounds outrageous, because I do not have alot
of architecture experience.
Thanks,
John
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by Phil Steitz <ph...@gmail.com>.
On 8/13/05, John Gant <jo...@gmail.com> wrote:
> Algorithms:
>
> - Feature reduction
> a. Basic cross correlation, including both spearman and pearson cross
> correlation algorithms.
> b. Principal Component Analysis.
> c. Entropy Based reduction.
All good, IMHO, esp. a. and b.
>
> I currently have a, and b finished but need to brush up on my junit skills :)
>
> -Difference Measures
> I had in mind a difference engine, basically an engine that handles
> all difference operations. This difference engine could, in the
> constructor or using set methods, take an instance of one of the
> following difference methods.
>
> a. euclidean distance
> b. city-block distance
Can you describe a little more exactly what you mean here and how it
would be used?
>
> -Pattern Discovery
> a. KMotif Discovery Algorithm.
I am not familiar with this algorithm or its uses. Can you provide
some more info and references?
>
> Again I have this algorithm completed, just need to boundary test everything.
>
> -Clustering Algorithms
> a. K-means Algorithm.
+1
>
> I'd like to discuss the architecture of the k-means, I have a few
> ideas and would like a little feedback. I know this is just a small
> subset of the available algorithms, but this seems to be a good start.
Yes. Ask away...
>
> Thanks,
> John
>
> On 8/13/05, Phil Steitz <ph...@gmail.com> wrote:
> > John,
> >
> > Sounds great! Extending the stat package to include some data mining
> > capabilities would be a good and useful addition to commons-math,
> > IMHO. To get started, the first thing to do is to read the
> > developer's guide
> > (http://jakarta.apache.org/commons/math/developers.html), which will
> > tell point you to the general apache references and go over some IP
> > stuff that we have to worry about in [math].
> >
> > Then either here or on the Wiki (see the guide for a link), post a
> > brief description of the kinds of mining algorithms that you are
> > interested in developing and we can get this going. On this list, pls
> > begin the subject line of all [math] messages with [math].
> >
> > Thanks in advance for your contributions!
> >
> > Phil
> >
> > On 8/13/05, John Gant <jo...@gmail.com> wrote:
> > > Hello,
> > > I am currently a graduate student in Computer Science and Computer
> > > Engineering at the University of Louisville, Kentucky. First let me
> > > congratulate the group of developers who commit and architect for
> > > apache commons. I have used many of the libraries and they are all of
> > > excellent quality (but I guess you already know that :)). I am
> > > interested in contributing to open source software and have interests
> > > that are in the domain of statistics with a focus in data mining.
> > > After writing many algorithms for classes, and asking an apache
> > > contributor if any of this would be needed elsewhere, he told me to
> > > purpose something to the dev list. So here it goes, I would like to
> > > help start a data mining section of commons math and advance the
> > > existing statistical libraries. I plan on developing the algorithms
> > > for personal use anyway, and would like to see some of my work be used
> > > by others. If anyone is interested we can continue this thread and I
> > > will email my code, and purpose my new algorithms.
> > >
> > >
> > > John Gant
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > >
> > >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
> >
>
>
> --
> John Gant
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by Phil Steitz <ph...@gmail.com>.
Great! As long as the code is original and the algorithms are
standard, there should be no problem with IP. Please just include
references to sources describing the algorithms and, of course, unit
tests when you submit code. If you are familiar with R and have
access to it, validating the unit test target values against R is a
nice to have. (There are some examples in src/experimental/R/).
Please also try to include API and inline documentation similar to
what you see elsewhere in [math].
Thanks!
Phil
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by John Gant <jo...@gmail.com>.
I will, later this evening, reply with the resources I used to create
the software. When I mentioned textbook, I meant that I used the
pseudocode as a guideline for development. None of the software I
use/or have used was a direct copy from a book like Numerical Recipes,
due to my need to customize the data structures.
Thanks,
John
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by John Gant <jo...@gmail.com>.
I have no link currently, although I can create a link soon (within
the day). All of my algorithms are textbook algorithms, that I have
used for class or other purposes. Although I am fairly certain these
are not restricted in any way, I am not overly familiar with IP issues
and welcome help on determining the status of these algorithms. Let me
work on putting this code online this evening (I'm on EST), and I'll
send a link out tonight so that my code is available. Please keep in
mind that this does need refactoring to include commons math basic
statistical and numerical classes. By this I mean that I have used my
own mean, sum, max, min methods within these classes. I plan on
refactoring this to include commons math based methods.
Thanks,
John
On 8/15/05, Lukas Vlcek <lu...@gmail.com> wrote:
> Hi,
>
> As Phil noted the question of IP is very important. As far as I know
> association mining concept is not covered by any patent or copyright
> (in fact I would be very surprised if it is).
>
> Also I think (and authors of [math] would confirm) that it is
> important to provide good and clear documentation for every new
> function to [math].
>
> John, do you have any documentation for you functions (I mean links to
> some papers where the function is described)? I am not [math] commiter
> but I am interested in your stuff. Is there any link where I could
> download your code?
>
> Regards,
> Lukas
>
>
> On 8/15/05, John Gant <jo...@gmail.com> wrote:
> > Excellent! I also found some older stuff that I had written yesterday,
> > and it included a tree clustering algorithm. Within this code, I also
> > found implementations of distibution-specific random number
> > generators, i.e. triangular, uniform, exponential, etc. I am not sure
> > if these would be useful in the statistical section, but I could
> > refactor them to use commons math classes.
> >
> > Thanks,
> > John
> >
> > On 8/15/05, Lukas Vlcek <lu...@gmail.com> wrote:
> > > Hi,
> > > this sounds great to me as well!
> > >
> > > I have been using [math] for some time and I found it very handy
> > > (congratulations to math team!). As for data mining functions I didn't
> > > have a chance to look what is implemented elsewhere yet (e.g: WEKA)
> > > but if there is anything what could be started in commons-math domena
> > > then I could offer my implementation of association mining package. It
> > > is not finished yet and still needs more tuning but I hope to finish
> > > it soon and see if it is helpful to other people.
> > >
> > > Regards,
> > > Lukas
> > >
> > >
> > > On 8/14/05, John Gant <jo...@gmail.com> wrote:
> > > > Algorithms:
> > > >
> > > > - Feature reduction
> > > > a. Basic cross correlation, including both spearman and pearson cross
> > > > correlation algorithms.
> > > > b. Principal Component Analysis.
> > > > c. Entropy Based reduction.
> > > >
> > > > I currently have a, and b finished but need to brush up on my junit skills :)
> > > >
> > > > -Difference Measures
> > > > I had in mind a difference engine, basically an engine that handles
> > > > all difference operations. This difference engine could, in the
> > > > constructor or using set methods, take an instance of one of the
> > > > following difference methods.
> > > >
> > > > a. euclidean distance
> > > > b. city-block distance
> > > >
> > > > -Pattern Discovery
> > > > a. KMotif Discovery Algorithm.
> > > >
> > > > Again I have this algorithm completed, just need to boundary test everything.
> > > >
> > > > -Clustering Algorithms
> > > > a. K-means Algorithm.
> > > >
> > > > I'd like to discuss the architecture of the k-means, I have a few
> > > > ideas and would like a little feedback. I know this is just a small
> > > > subset of the available algorithms, but this seems to be a good start.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > > On 8/13/05, Phil Steitz <ph...@gmail.com> wrote:
> > > > > John,
> > > > >
> > > > > Sounds great! Extending the stat package to include some data mining
> > > > > capabilities would be a good and useful addition to commons-math,
> > > > > IMHO. To get started, the first thing to do is to read the
> > > > > developer's guide
> > > > > (http://jakarta.apache.org/commons/math/developers.html), which will
> > > > > tell point you to the general apache references and go over some IP
> > > > > stuff that we have to worry about in [math].
> > > > >
> > > > > Then either here or on the Wiki (see the guide for a link), post a
> > > > > brief description of the kinds of mining algorithms that you are
> > > > > interested in developing and we can get this going. On this list, pls
> > > > > begin the subject line of all [math] messages with [math].
> > > > >
> > > > > Thanks in advance for your contributions!
> > > > >
> > > > > Phil
> > > > >
> > > > > On 8/13/05, John Gant <jo...@gmail.com> wrote:
> > > > > > Hello,
> > > > > > I am currently a graduate student in Computer Science and Computer
> > > > > > Engineering at the University of Louisville, Kentucky. First let me
> > > > > > congratulate the group of developers who commit and architect for
> > > > > > apache commons. I have used many of the libraries and they are all of
> > > > > > excellent quality (but I guess you already know that :)). I am
> > > > > > interested in contributing to open source software and have interests
> > > > > > that are in the domain of statistics with a focus in data mining.
> > > > > > After writing many algorithms for classes, and asking an apache
> > > > > > contributor if any of this would be needed elsewhere, he told me to
> > > > > > purpose something to the dev list. So here it goes, I would like to
> > > > > > help start a data mining section of commons math and advance the
> > > > > > existing statistical libraries. I plan on developing the algorithms
> > > > > > for personal use anyway, and would like to see some of my work be used
> > > > > > by others. If anyone is interested we can continue this thread and I
> > > > > > will email my code, and purpose my new algorithms.
> > > > > >
> > > > > >
> > > > > > John Gant
> > > > > >
> > > > > > ---------------------------------------------------------------------
> > > > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > > > >
> > > > > >
> > > > >
> > > > > ---------------------------------------------------------------------
> > > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > John Gant
> > > >
> > > > ---------------------------------------------------------------------
> > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > >
> > > >
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > >
> > >
> >
> >
> > --
> > John Gant
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
--
John Gant
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
[math] Re: commons math
Posted by Lukas Vlcek <lu...@gmail.com>.
Hi,
As Phil noted the question of IP is very important. As far as I know
association mining concept is not covered by any patent or copyright
(in fact I would be very surprised if it is).
Also I think (and authors of [math] would confirm) that it is
important to provide good and clear documentation for every new
function to [math].
John, do you have any documentation for you functions (I mean links to
some papers where the function is described)? I am not [math] commiter
but I am interested in your stuff. Is there any link where I could
download your code?
Regards,
Lukas
On 8/15/05, John Gant <jo...@gmail.com> wrote:
> Excellent! I also found some older stuff that I had written yesterday,
> and it included a tree clustering algorithm. Within this code, I also
> found implementations of distibution-specific random number
> generators, i.e. triangular, uniform, exponential, etc. I am not sure
> if these would be useful in the statistical section, but I could
> refactor them to use commons math classes.
>
> Thanks,
> John
>
> On 8/15/05, Lukas Vlcek <lu...@gmail.com> wrote:
> > Hi,
> > this sounds great to me as well!
> >
> > I have been using [math] for some time and I found it very handy
> > (congratulations to math team!). As for data mining functions I didn't
> > have a chance to look what is implemented elsewhere yet (e.g: WEKA)
> > but if there is anything what could be started in commons-math domena
> > then I could offer my implementation of association mining package. It
> > is not finished yet and still needs more tuning but I hope to finish
> > it soon and see if it is helpful to other people.
> >
> > Regards,
> > Lukas
> >
> >
> > On 8/14/05, John Gant <jo...@gmail.com> wrote:
> > > Algorithms:
> > >
> > > - Feature reduction
> > > a. Basic cross correlation, including both spearman and pearson cross
> > > correlation algorithms.
> > > b. Principal Component Analysis.
> > > c. Entropy Based reduction.
> > >
> > > I currently have a, and b finished but need to brush up on my junit skills :)
> > >
> > > -Difference Measures
> > > I had in mind a difference engine, basically an engine that handles
> > > all difference operations. This difference engine could, in the
> > > constructor or using set methods, take an instance of one of the
> > > following difference methods.
> > >
> > > a. euclidean distance
> > > b. city-block distance
> > >
> > > -Pattern Discovery
> > > a. KMotif Discovery Algorithm.
> > >
> > > Again I have this algorithm completed, just need to boundary test everything.
> > >
> > > -Clustering Algorithms
> > > a. K-means Algorithm.
> > >
> > > I'd like to discuss the architecture of the k-means, I have a few
> > > ideas and would like a little feedback. I know this is just a small
> > > subset of the available algorithms, but this seems to be a good start.
> > >
> > > Thanks,
> > > John
> > >
> > > On 8/13/05, Phil Steitz <ph...@gmail.com> wrote:
> > > > John,
> > > >
> > > > Sounds great! Extending the stat package to include some data mining
> > > > capabilities would be a good and useful addition to commons-math,
> > > > IMHO. To get started, the first thing to do is to read the
> > > > developer's guide
> > > > (http://jakarta.apache.org/commons/math/developers.html), which will
> > > > tell point you to the general apache references and go over some IP
> > > > stuff that we have to worry about in [math].
> > > >
> > > > Then either here or on the Wiki (see the guide for a link), post a
> > > > brief description of the kinds of mining algorithms that you are
> > > > interested in developing and we can get this going. On this list, pls
> > > > begin the subject line of all [math] messages with [math].
> > > >
> > > > Thanks in advance for your contributions!
> > > >
> > > > Phil
> > > >
> > > > On 8/13/05, John Gant <jo...@gmail.com> wrote:
> > > > > Hello,
> > > > > I am currently a graduate student in Computer Science and Computer
> > > > > Engineering at the University of Louisville, Kentucky. First let me
> > > > > congratulate the group of developers who commit and architect for
> > > > > apache commons. I have used many of the libraries and they are all of
> > > > > excellent quality (but I guess you already know that :)). I am
> > > > > interested in contributing to open source software and have interests
> > > > > that are in the domain of statistics with a focus in data mining.
> > > > > After writing many algorithms for classes, and asking an apache
> > > > > contributor if any of this would be needed elsewhere, he told me to
> > > > > purpose something to the dev list. So here it goes, I would like to
> > > > > help start a data mining section of commons math and advance the
> > > > > existing statistical libraries. I plan on developing the algorithms
> > > > > for personal use anyway, and would like to see some of my work be used
> > > > > by others. If anyone is interested we can continue this thread and I
> > > > > will email my code, and purpose my new algorithms.
> > > > >
> > > > >
> > > > > John Gant
> > > > >
> > > > > ---------------------------------------------------------------------
> > > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > > >
> > > > >
> > > >
> > > > ---------------------------------------------------------------------
> > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > >
> > > >
> > >
> > >
> > > --
> > > John Gant
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > >
> > >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
> >
>
>
> --
> John Gant
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by Phil Steitz <ph...@gmail.com>.
On 8/15/05, John Gant <jo...@gmail.com> wrote:
> Excellent! I also found some older stuff that I had written yesterday,
> and it included a tree clustering algorithm. Within this code, I also
> found implementations of distibution-specific random number
> generators, i.e. triangular, uniform, exponential, etc.
Have a look at the o.a.c.math.random package. Uniform and exponential
are covered there, bt triangular is not. If you have others, we can
talk about adding these.
> I am not sure
> if these would be useful in the statistical section, but I could
> refactor them to use commons math classes.
>
> Thanks,
> John
>
> On 8/15/05, Lukas Vlcek <lu...@gmail.com> wrote:
> > Hi,
> > this sounds great to me as well!
> >
> > I have been using [math] for some time and I found it very handy
> > (congratulations to math team!). As for data mining functions I didn't
> > have a chance to look what is implemented elsewhere yet (e.g: WEKA)
> > but if there is anything what could be started in commons-math domena
> > then I could offer my implementation of association mining package. It
> > is not finished yet and still needs more tuning but I hope to finish
> > it soon and see if it is helpful to other people.
> >
> > Regards,
> > Lukas
> >
> >
> > On 8/14/05, John Gant <jo...@gmail.com> wrote:
> > > Algorithms:
> > >
> > > - Feature reduction
> > > a. Basic cross correlation, including both spearman and pearson cross
> > > correlation algorithms.
> > > b. Principal Component Analysis.
> > > c. Entropy Based reduction.
> > >
> > > I currently have a, and b finished but need to brush up on my junit skills :)
> > >
> > > -Difference Measures
> > > I had in mind a difference engine, basically an engine that handles
> > > all difference operations. This difference engine could, in the
> > > constructor or using set methods, take an instance of one of the
> > > following difference methods.
> > >
> > > a. euclidean distance
> > > b. city-block distance
> > >
> > > -Pattern Discovery
> > > a. KMotif Discovery Algorithm.
> > >
> > > Again I have this algorithm completed, just need to boundary test everything.
> > >
> > > -Clustering Algorithms
> > > a. K-means Algorithm.
> > >
> > > I'd like to discuss the architecture of the k-means, I have a few
> > > ideas and would like a little feedback. I know this is just a small
> > > subset of the available algorithms, but this seems to be a good start.
> > >
> > > Thanks,
> > > John
> > >
> > > On 8/13/05, Phil Steitz <ph...@gmail.com> wrote:
> > > > John,
> > > >
> > > > Sounds great! Extending the stat package to include some data mining
> > > > capabilities would be a good and useful addition to commons-math,
> > > > IMHO. To get started, the first thing to do is to read the
> > > > developer's guide
> > > > (http://jakarta.apache.org/commons/math/developers.html), which will
> > > > tell point you to the general apache references and go over some IP
> > > > stuff that we have to worry about in [math].
> > > >
> > > > Then either here or on the Wiki (see the guide for a link), post a
> > > > brief description of the kinds of mining algorithms that you are
> > > > interested in developing and we can get this going. On this list, pls
> > > > begin the subject line of all [math] messages with [math].
> > > >
> > > > Thanks in advance for your contributions!
> > > >
> > > > Phil
> > > >
> > > > On 8/13/05, John Gant <jo...@gmail.com> wrote:
> > > > > Hello,
> > > > > I am currently a graduate student in Computer Science and Computer
> > > > > Engineering at the University of Louisville, Kentucky. First let me
> > > > > congratulate the group of developers who commit and architect for
> > > > > apache commons. I have used many of the libraries and they are all of
> > > > > excellent quality (but I guess you already know that :)). I am
> > > > > interested in contributing to open source software and have interests
> > > > > that are in the domain of statistics with a focus in data mining.
> > > > > After writing many algorithms for classes, and asking an apache
> > > > > contributor if any of this would be needed elsewhere, he told me to
> > > > > purpose something to the dev list. So here it goes, I would like to
> > > > > help start a data mining section of commons math and advance the
> > > > > existing statistical libraries. I plan on developing the algorithms
> > > > > for personal use anyway, and would like to see some of my work be used
> > > > > by others. If anyone is interested we can continue this thread and I
> > > > > will email my code, and purpose my new algorithms.
> > > > >
> > > > >
> > > > > John Gant
> > > > >
> > > > > ---------------------------------------------------------------------
> > > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > > >
> > > > >
> > > >
> > > > ---------------------------------------------------------------------
> > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > >
> > > >
> > >
> > >
> > > --
> > > John Gant
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > >
> > >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
> >
>
>
> --
> John Gant
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by John Gant <jo...@gmail.com>.
Excellent! I also found some older stuff that I had written yesterday,
and it included a tree clustering algorithm. Within this code, I also
found implementations of distibution-specific random number
generators, i.e. triangular, uniform, exponential, etc. I am not sure
if these would be useful in the statistical section, but I could
refactor them to use commons math classes.
Thanks,
John
On 8/15/05, Lukas Vlcek <lu...@gmail.com> wrote:
> Hi,
> this sounds great to me as well!
>
> I have been using [math] for some time and I found it very handy
> (congratulations to math team!). As for data mining functions I didn't
> have a chance to look what is implemented elsewhere yet (e.g: WEKA)
> but if there is anything what could be started in commons-math domena
> then I could offer my implementation of association mining package. It
> is not finished yet and still needs more tuning but I hope to finish
> it soon and see if it is helpful to other people.
>
> Regards,
> Lukas
>
>
> On 8/14/05, John Gant <jo...@gmail.com> wrote:
> > Algorithms:
> >
> > - Feature reduction
> > a. Basic cross correlation, including both spearman and pearson cross
> > correlation algorithms.
> > b. Principal Component Analysis.
> > c. Entropy Based reduction.
> >
> > I currently have a, and b finished but need to brush up on my junit skills :)
> >
> > -Difference Measures
> > I had in mind a difference engine, basically an engine that handles
> > all difference operations. This difference engine could, in the
> > constructor or using set methods, take an instance of one of the
> > following difference methods.
> >
> > a. euclidean distance
> > b. city-block distance
> >
> > -Pattern Discovery
> > a. KMotif Discovery Algorithm.
> >
> > Again I have this algorithm completed, just need to boundary test everything.
> >
> > -Clustering Algorithms
> > a. K-means Algorithm.
> >
> > I'd like to discuss the architecture of the k-means, I have a few
> > ideas and would like a little feedback. I know this is just a small
> > subset of the available algorithms, but this seems to be a good start.
> >
> > Thanks,
> > John
> >
> > On 8/13/05, Phil Steitz <ph...@gmail.com> wrote:
> > > John,
> > >
> > > Sounds great! Extending the stat package to include some data mining
> > > capabilities would be a good and useful addition to commons-math,
> > > IMHO. To get started, the first thing to do is to read the
> > > developer's guide
> > > (http://jakarta.apache.org/commons/math/developers.html), which will
> > > tell point you to the general apache references and go over some IP
> > > stuff that we have to worry about in [math].
> > >
> > > Then either here or on the Wiki (see the guide for a link), post a
> > > brief description of the kinds of mining algorithms that you are
> > > interested in developing and we can get this going. On this list, pls
> > > begin the subject line of all [math] messages with [math].
> > >
> > > Thanks in advance for your contributions!
> > >
> > > Phil
> > >
> > > On 8/13/05, John Gant <jo...@gmail.com> wrote:
> > > > Hello,
> > > > I am currently a graduate student in Computer Science and Computer
> > > > Engineering at the University of Louisville, Kentucky. First let me
> > > > congratulate the group of developers who commit and architect for
> > > > apache commons. I have used many of the libraries and they are all of
> > > > excellent quality (but I guess you already know that :)). I am
> > > > interested in contributing to open source software and have interests
> > > > that are in the domain of statistics with a focus in data mining.
> > > > After writing many algorithms for classes, and asking an apache
> > > > contributor if any of this would be needed elsewhere, he told me to
> > > > purpose something to the dev list. So here it goes, I would like to
> > > > help start a data mining section of commons math and advance the
> > > > existing statistical libraries. I plan on developing the algorithms
> > > > for personal use anyway, and would like to see some of my work be used
> > > > by others. If anyone is interested we can continue this thread and I
> > > > will email my code, and purpose my new algorithms.
> > > >
> > > >
> > > > John Gant
> > > >
> > > > ---------------------------------------------------------------------
> > > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > > >
> > > >
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > >
> > >
> >
> >
> > --
> > John Gant
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
--
John Gant
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by Lukas Vlcek <lu...@gmail.com>.
Hi,
this sounds great to me as well!
I have been using [math] for some time and I found it very handy
(congratulations to math team!). As for data mining functions I didn't
have a chance to look what is implemented elsewhere yet (e.g: WEKA)
but if there is anything what could be started in commons-math domena
then I could offer my implementation of association mining package. It
is not finished yet and still needs more tuning but I hope to finish
it soon and see if it is helpful to other people.
Regards,
Lukas
On 8/14/05, John Gant <jo...@gmail.com> wrote:
> Algorithms:
>
> - Feature reduction
> a. Basic cross correlation, including both spearman and pearson cross
> correlation algorithms.
> b. Principal Component Analysis.
> c. Entropy Based reduction.
>
> I currently have a, and b finished but need to brush up on my junit skills :)
>
> -Difference Measures
> I had in mind a difference engine, basically an engine that handles
> all difference operations. This difference engine could, in the
> constructor or using set methods, take an instance of one of the
> following difference methods.
>
> a. euclidean distance
> b. city-block distance
>
> -Pattern Discovery
> a. KMotif Discovery Algorithm.
>
> Again I have this algorithm completed, just need to boundary test everything.
>
> -Clustering Algorithms
> a. K-means Algorithm.
>
> I'd like to discuss the architecture of the k-means, I have a few
> ideas and would like a little feedback. I know this is just a small
> subset of the available algorithms, but this seems to be a good start.
>
> Thanks,
> John
>
> On 8/13/05, Phil Steitz <ph...@gmail.com> wrote:
> > John,
> >
> > Sounds great! Extending the stat package to include some data mining
> > capabilities would be a good and useful addition to commons-math,
> > IMHO. To get started, the first thing to do is to read the
> > developer's guide
> > (http://jakarta.apache.org/commons/math/developers.html), which will
> > tell point you to the general apache references and go over some IP
> > stuff that we have to worry about in [math].
> >
> > Then either here or on the Wiki (see the guide for a link), post a
> > brief description of the kinds of mining algorithms that you are
> > interested in developing and we can get this going. On this list, pls
> > begin the subject line of all [math] messages with [math].
> >
> > Thanks in advance for your contributions!
> >
> > Phil
> >
> > On 8/13/05, John Gant <jo...@gmail.com> wrote:
> > > Hello,
> > > I am currently a graduate student in Computer Science and Computer
> > > Engineering at the University of Louisville, Kentucky. First let me
> > > congratulate the group of developers who commit and architect for
> > > apache commons. I have used many of the libraries and they are all of
> > > excellent quality (but I guess you already know that :)). I am
> > > interested in contributing to open source software and have interests
> > > that are in the domain of statistics with a focus in data mining.
> > > After writing many algorithms for classes, and asking an apache
> > > contributor if any of this would be needed elsewhere, he told me to
> > > purpose something to the dev list. So here it goes, I would like to
> > > help start a data mining section of commons math and advance the
> > > existing statistical libraries. I plan on developing the algorithms
> > > for personal use anyway, and would like to see some of my work be used
> > > by others. If anyone is interested we can continue this thread and I
> > > will email my code, and purpose my new algorithms.
> > >
> > >
> > > John Gant
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> > >
> > >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
> >
>
>
> --
> John Gant
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [math] Re: commons math
Posted by John Gant <jo...@gmail.com>.
Algorithms:
- Feature reduction
a. Basic cross correlation, including both spearman and pearson cross
correlation algorithms.
b. Principal Component Analysis.
c. Entropy Based reduction.
I currently have a, and b finished but need to brush up on my junit skills :)
-Difference Measures
I had in mind a difference engine, basically an engine that handles
all difference operations. This difference engine could, in the
constructor or using set methods, take an instance of one of the
following difference methods.
a. euclidean distance
b. city-block distance
-Pattern Discovery
a. KMotif Discovery Algorithm.
Again I have this algorithm completed, just need to boundary test everything.
-Clustering Algorithms
a. K-means Algorithm.
I'd like to discuss the architecture of the k-means, I have a few
ideas and would like a little feedback. I know this is just a small
subset of the available algorithms, but this seems to be a good start.
Thanks,
John
On 8/13/05, Phil Steitz <ph...@gmail.com> wrote:
> John,
>
> Sounds great! Extending the stat package to include some data mining
> capabilities would be a good and useful addition to commons-math,
> IMHO. To get started, the first thing to do is to read the
> developer's guide
> (http://jakarta.apache.org/commons/math/developers.html), which will
> tell point you to the general apache references and go over some IP
> stuff that we have to worry about in [math].
>
> Then either here or on the Wiki (see the guide for a link), post a
> brief description of the kinds of mining algorithms that you are
> interested in developing and we can get this going. On this list, pls
> begin the subject line of all [math] messages with [math].
>
> Thanks in advance for your contributions!
>
> Phil
>
> On 8/13/05, John Gant <jo...@gmail.com> wrote:
> > Hello,
> > I am currently a graduate student in Computer Science and Computer
> > Engineering at the University of Louisville, Kentucky. First let me
> > congratulate the group of developers who commit and architect for
> > apache commons. I have used many of the libraries and they are all of
> > excellent quality (but I guess you already know that :)). I am
> > interested in contributing to open source software and have interests
> > that are in the domain of statistics with a focus in data mining.
> > After writing many algorithms for classes, and asking an apache
> > contributor if any of this would be needed elsewhere, he told me to
> > purpose something to the dev list. So here it goes, I would like to
> > help start a data mining section of commons math and advance the
> > existing statistical libraries. I plan on developing the algorithms
> > for personal use anyway, and would like to see some of my work be used
> > by others. If anyone is interested we can continue this thread and I
> > will email my code, and purpose my new algorithms.
> >
> >
> > John Gant
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
--
John Gant
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
[math] Re: commons math
Posted by Phil Steitz <ph...@gmail.com>.
John,
Sounds great! Extending the stat package to include some data mining
capabilities would be a good and useful addition to commons-math,
IMHO. To get started, the first thing to do is to read the
developer's guide
(http://jakarta.apache.org/commons/math/developers.html), which will
tell point you to the general apache references and go over some IP
stuff that we have to worry about in [math].
Then either here or on the Wiki (see the guide for a link), post a
brief description of the kinds of mining algorithms that you are
interested in developing and we can get this going. On this list, pls
begin the subject line of all [math] messages with [math].
Thanks in advance for your contributions!
Phil
On 8/13/05, John Gant <jo...@gmail.com> wrote:
> Hello,
> I am currently a graduate student in Computer Science and Computer
> Engineering at the University of Louisville, Kentucky. First let me
> congratulate the group of developers who commit and architect for
> apache commons. I have used many of the libraries and they are all of
> excellent quality (but I guess you already know that :)). I am
> interested in contributing to open source software and have interests
> that are in the domain of statistics with a focus in data mining.
> After writing many algorithms for classes, and asking an apache
> contributor if any of this would be needed elsewhere, he told me to
> purpose something to the dev list. So here it goes, I would like to
> help start a data mining section of commons math and advance the
> existing statistical libraries. I plan on developing the algorithms
> for personal use anyway, and would like to see some of my work be used
> by others. If anyone is interested we can continue this thread and I
> will email my code, and purpose my new algorithms.
>
>
> John Gant
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org