Mailing-List: contact java-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: java-user@lucene.apache.org
Received-SPF: pass (nike.apache.org: domain of sxk1969@hotmail.com designates
 65.54.190.207 as permitted sender)
Message-ID: <BAY166-W34C0B8D32E18680A224877A8760@phx.gbl>
Content-Type: multipart/alternative;
	boundary="_7b0a2b94-8fb2-49ed-b734-099208ac3901_"
From: Saikat Kanjilal <sxk1969@hotmail.com>
To: "solr-user@lucene.apache.org" <solr-user@lucene.apache.org>
CC: "java-user@lucene.apache.org" <java-user@lucene.apache.org>
Subject: RE: Content based recommender using lucene/solr
Date: Fri, 28 Jun 2013 11:07:00 -0700
Importance: Normal
In-Reply-To: 
 <CACp51msi4u2zmmuaPHvvOAmqvkKRpXgkhEriMp+U69tTAb8wrQ@mail.gmail.com>
References: 
 <CACp51ms-oXiVUyobVgBwbNUgzRoat4uznOXSJkode2MPrDCtqA@mail.gmail.com>,<BAY166-W22C63D3ADBEB8CD712827A8760@phx.gbl>,<CACp51msi4u2zmmuaPHvvOAmqvkKRpXgkhEriMp+U69tTAb8wrQ@mail.gmail.com>
MIME-Version: 1.0

--_7b0a2b94-8fb2-49ed-b734-099208ac3901_
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

You could build a custom recommender in mahout to accomplish this=2C also j=
ust out of curiosity why the content based approach as opposed to building =
a recommender based on co-occurence.  One other thing=2C what is your data =
size=2C are you looking at scale where you need something like hadoop?

> From: lcguerrerocovo@gmail.com
> Date: Fri=2C 28 Jun 2013 13:02:00 -0500
> Subject: Re: Content based recommender using lucene/solr
> To: solr-user@lucene.apache.org
> CC: java-user@lucene.apache.org
>=20
> Hey saikat=2C thanks for your suggestion. I've looked into mahout and oth=
er
> alternatives for computing k nearest neighbors. I would have to run a job
> and computer the k nearest neighbors and track them in the index for
> retrieval. I wanted to see if this was something I could do with lucene
> using lucene's scoring function and solr's morelikethis component. The jo=
b
> you specifically mention is for Item based recommendation which would
> require me to track the different items users have viewed. I'm looking fo=
r
> a content based approach where I would use a distance measure to establis=
h
> how near items are (how similar) and have some kind of training phase to
> adjust weights.
>=20
>=20
> On Fri=2C Jun 28=2C 2013 at 12:42 PM=2C Saikat Kanjilal <sxk1969@hotmail.=
com>wrote:
>=20
> > Why not just use mahout to do this=2C there is an item similarity algor=
ithm
> > in mahout that does exactly this :)
> >
> >
> > https://builds.apache.org/job/Mahout-Quality/javadoc/org/apache/mahout/=
cf/taste/hadoop/similarity/item/ItemSimilarityJob.html
> >
> > You can use mahout in distributed and non-distributed mode as well.
> >
> > > From: lcguerrerocovo@gmail.com
> > > Date: Fri=2C 28 Jun 2013 12:16:57 -0500
> > > Subject: Content based recommender using lucene/solr
> > > To: solr-user@lucene.apache.org=3B java-user@lucene.apache.org
> > >
> > > Hi=2C
> > >
> > > I'm using lucene and solr right now in a production environment with =
an
> > > index of about a million docs. I'm working on a recommender that
> > basically
> > > would list the n most similar items to the user based on the current =
item
> > > he is viewing.
> > >
> > > I've been thinking of using solr/lucene since I already have all docs
> > > available and I want a quick version that can be deployed while we wo=
rk
> > on
> > > a more robust recommender. How about overriding the default similarit=
y so
> > > that it scores documents based on the euclidean distance of normalize=
d
> > item
> > > attributes and then using a morelikethis component to pass in the
> > > attributes of the item for which I want to generate recommendations? =
I
> > know
> > > it has its issues like recomputing scores/normalization/weight
> > application
> > > at query time which could make this idea unfeasible/impractical. I'm =
at a
> > > very preliminary stage right now with this and would love some
> > suggestions
> > > from experienced users.
> > >
> > > thank you=2C
> > >
> > > Luis Guerrero
> >
> >
>=20
>=20
>=20
> --=20
> Luis Carlos Guerrero Covo
> M.S. Computer Engineering
> (57) 3183542047
 		 	   		  =

--_7b0a2b94-8fb2-49ed-b734-099208ac3901_--