You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mahout.apache.org by fht <li...@gmail.com> on 2012/05/18 23:00:25 UTC
How to approach this? Classification vs Recommendation
Hi,
I suppose this a combination of a generic machine learning question and a
mahout question.
I have a data set. A user may or may not be part of a funded scheme.
If there are not part of the funded scheme they might be susceptible to
certain problems a, b, c and d.
If there are part of the funded scheme they might incur problems a, b and c
but not d.
I want to process the data set to infer something like people who *are* part
of funded scheme won't encounter problem c and d.
Is this a recommendation or a classification - How do I approach this?
Also can hive inteactt with mahout - I read (correct me if I'm wrong) that
it's probably best to input data to mahout in csv format - I assume this is
possible with hive?
many thanks.
--
View this message in context: http://lucene.472066.n3.nabble.com/How-to-approach-this-Classification-vs-Recommendation-tp3984795.html
Sent from the Mahout User List mailing list archive at Nabble.com.
Re: How to approach this? Classification vs Recommendation
Posted by Ted Dunning <te...@gmail.com>.
Not so trivially, these classifiers can help each other. What you have is
a form of transduction or example based learnng.
On Fri, May 18, 2012 at 5:24 PM, Sean Owen <sr...@gmail.com> wrote:
> Trivially it's four classifiers. You have just one input here, and
> it's binary. That seems like too little info to discriminate on. All
> you can learn -- and it doesn't really need a classifier algorithm --
> is there's an x% chance of encountering problem a if funded, and
> (100-x)% of a if not.
>
> On Fri, May 18, 2012 at 10:00 PM, fht <li...@gmail.com> wrote:
> > Hi,
> >
> > I suppose this a combination of a generic machine learning question and a
> > mahout question.
> >
> > I have a data set. A user may or may not be part of a funded scheme.
> >
> > If there are not part of the funded scheme they might be susceptible to
> > certain problems a, b, c and d.
> > If there are part of the funded scheme they might incur problems a, b
> and c
> > but not d.
> >
> > I want to process the data set to infer something like people who *are*
> part
> > of funded scheme won't encounter problem c and d.
> >
> > Is this a recommendation or a classification - How do I approach this?
> >
> > Also can hive inteactt with mahout - I read (correct me if I'm wrong)
> that
> > it's probably best to input data to mahout in csv format - I assume this
> is
> > possible with hive?
> >
> > many thanks.
> >
> > --
> > View this message in context:
> http://lucene.472066.n3.nabble.com/How-to-approach-this-Classification-vs-Recommendation-tp3984795.html
> > Sent from the Mahout User List mailing list archive at Nabble.com.
>
Re: How to approach this? Classification vs Recommendation
Posted by Sean Owen <sr...@gmail.com>.
Trivially it's four classifiers. You have just one input here, and
it's binary. That seems like too little info to discriminate on. All
you can learn -- and it doesn't really need a classifier algorithm --
is there's an x% chance of encountering problem a if funded, and
(100-x)% of a if not.
On Fri, May 18, 2012 at 10:00 PM, fht <li...@gmail.com> wrote:
> Hi,
>
> I suppose this a combination of a generic machine learning question and a
> mahout question.
>
> I have a data set. A user may or may not be part of a funded scheme.
>
> If there are not part of the funded scheme they might be susceptible to
> certain problems a, b, c and d.
> If there are part of the funded scheme they might incur problems a, b and c
> but not d.
>
> I want to process the data set to infer something like people who *are* part
> of funded scheme won't encounter problem c and d.
>
> Is this a recommendation or a classification - How do I approach this?
>
> Also can hive inteactt with mahout - I read (correct me if I'm wrong) that
> it's probably best to input data to mahout in csv format - I assume this is
> possible with hive?
>
> many thanks.
>
> --
> View this message in context: http://lucene.472066.n3.nabble.com/How-to-approach-this-Classification-vs-Recommendation-tp3984795.html
> Sent from the Mahout User List mailing list archive at Nabble.com.