Mailing List Archive: Lucene: Java-User

Grouping Based on Multiple Fields Similarity

Index | Next | Previous | View Threaded

java.phi at phi-integration

May 21, 2012, 7:31 AM

Post #1 of 2 (195 views)
Permalink

Grouping Based on Multiple Fields Similarity

Hi Everyone,

I'm quite new to Lucene and would like to ask if my case below is possible
with Lucene solution.

Let's say I have 200,000 rows from a relational table with multiple fields,
and I will have them indexed with Lucene. After indexing, I'd like to have
a grouping / clustering based on similarity between four of five fields.

The end result would be something like this :

- Grouping 1, count : 3
- row id = 1
- row id = 23
- row id = 100
- Grouping 2
- row id = 1
- row id = 23
- ...

I have done some research and MoreLikeThis class can be use on this :

http://lucene.apache.org/core/old_versioned_docs/versions/3_0_0/api/contrib-queries/org/apache/lucene/search/similar/MoreLikeThis.html

I'm still learning the usage of this class. But maybe if anyone can confirm
the approach and maybe some guide, it would be very appreciated.

Many thanks before..

Regards,

Robby

java.phi at phi-integration

May 21, 2012, 9:02 AM

Post #2 of 2 (189 views)
Permalink

Re: Grouping Based on Multiple Fields Similarity [In reply to]

Hi All,

Sorry... I give wrong example, should be like this actually..

On Mon, May 21, 2012 at 9:31 PM, Robby <java.phi [at] phi-integration> wrote:

> - Grouping 1, count : 3
> - row id = 1
> - row id = 23
> - row id = 100
> - Grouping 2
> - row id = 11
> - row id = 133
> - ...
>

Regards,

Robby

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads