You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Alessandro Benedetti (JIRA)" <ji...@apache.org> on 2016/03/18 17:18:33 UTC

[jira] [Updated] (LUCENE-7116) Document Classifiers to handle numeric fields

     [ https://issues.apache.org/jira/browse/LUCENE-7116?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Alessandro Benedetti updated LUCENE-7116:
-----------------------------------------
    Description: 
DocumentClassifiers should handle numeric fields.
Knn Document Classifier should be able to handle function queries( like the distance from the numeric value in input)  as boost factors on top of the MLT boolean query.
Investigations will follow with tentative design and patch

e.g.
input document has price:5

We can classify the document based on all the ones in the index, in proximity to 5 as the price value.

  was:
DocumentClassifiers should handle numeric fields.
Knn Document Classifier should be able to handle function queries( like the distance from the numeric value in input)  as boost factors on top of the MLT boolean query.
Investigations will follow with tentative design and patch

e.g.
input document has price:5

We can classify the document based on all the ones in the index, close to 5 as the price value.


> Document Classifiers to handle numeric fields
> ---------------------------------------------
>
>                 Key: LUCENE-7116
>                 URL: https://issues.apache.org/jira/browse/LUCENE-7116
>             Project: Lucene - Core
>          Issue Type: Improvement
>          Components: modules/classification
>    Affects Versions: 5.5
>            Reporter: Alessandro Benedetti
>            Priority: Minor
>              Labels: classification, classifier
>
> DocumentClassifiers should handle numeric fields.
> Knn Document Classifier should be able to handle function queries( like the distance from the numeric value in input)  as boost factors on top of the MLT boolean query.
> Investigations will follow with tentative design and patch
> e.g.
> input document has price:5
> We can classify the document based on all the ones in the index, in proximity to 5 as the price value.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org