[jira] Updated: (LUCENE-836) Benchmarks Enhancements (precision/recall, TREC, Wikipedia)

["Doron Cohen (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/LUCENE-836?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Doron Cohen updated LUCENE-836:
-------------------------------

    Attachment: lucene-836.benchmark.quality.patch

lucene-836.benchmark.quality.patch adds a new package "quality" under o.a.l.benchmark. 

This is also followup to some of http://www.mail-archive.com/java-dev@lucene.apache.org/msg10851.html

Patch is based on trunk folder. 
Fastest way to test it: "ant test" from contrib/benchmark dir.
To see more output in this run, try "ant test -Dtests.verbose=true".

This is early code, not ready to commit - wanted to show it sooner for feedback, especially the API. 

For a quick view of the API see benchmark.quality at http://people.apache.org/~doronc/api (note that not much javadocs yet - I would wait with that for API closure.)

Code in this patch is:
  - extendable.
  - can run a quality benchmark.
  - report quality results, comparing to given judgements (optional).
  - create a submission log (optional).
  - format of submission log can be modified, by extending a logger class.
  - format of inputs - queries, judgments - can be modified, by extending 
    default readers, or by providing pre-read ones.

There is a general "Judge" interface - answering if a given doc name is valid for a given "QualityQuery". And one implementation of it, based on Trec's QRels. The alternative of TRels, for instance, would mean another implementation of the "Judge" interface. (I would love a better name for it, btw...)

A new TestQualityRun tests this package on the Reuters collection - so that test source is a good place to start, to see how to run a quality test.

> Benchmarks Enhancements (precision/recall, TREC, Wikipedia)
> -----------------------------------------------------------
>
>                 Key: LUCENE-836
>                 URL: https://issues.apache.org/jira/browse/LUCENE-836
>             Project: Lucene - Java
>          Issue Type: New Feature
>          Components: Other
>            Reporter: Grant Ingersoll
>            Priority: Minor
>         Attachments: lucene-836.benchmark.quality.patch
>
>
> Would be great if the benchmark contrib had a way of providing precision/recall benchmark information ala TREC.  I don't know what the copyright issues are for the TREC queries/data (I think the queries are available, but not sure about the data), so not sure if the is even feasible, but I could imagine we could at least incorporate support for it for those who have access to the data.  It has been a long time since I have participated in TREC, so perhaps someone more familiar w/ the latest can fill in the blanks here.
> Another option is to ask for volunteers to create queries and make judgments for the Reuters data, but that is a bit more complex and probably not necessary.  Even so, an Apache licensed set of benchmarks may be useful for the community as a whole.  Hmmm.... 
> Wikipedia might be another option instead of Reuters to setup as a download for benchmarking, as it is quite large and I believe the licensing terms are quite amenable.  Having a larger collection would be good for stressing Lucene more and would give many users a demonstration of how Lucene handles large collections.
> At any rate, this kind of information could be useful for people looking at different indexing schemes, formats, payloads and different query strategies.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


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