[
https://issues.apache.org/jira/browse/LUCENE-5350?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13842461#comment-13842461
]
Areek Zillur commented on LUCENE-5350:
--------------------------------------
Disregard the previous benchmark stats. There was a bug in how the keys were
used for building the suggester (hence showed the unexplainable QPS).
The updated benchmark results is as follows:
{code}
-- Input stats
Input size: 53022, numContexts: 2666, Avg. Input/Context: 20
-- prefixes: 2-4, num: 7, onlyMorePopular: false
ContextAwareAnalyzingSuggester queries: 53022, time[ms]: 2630 [+- 124.14],
~kQPS: 20
AnalyzingSuggester queries: 53022, time[ms]: 2249 [+- 25.16], ~kQPS: 24
-- RAM consumption
AnalyzingSuggester size[B]: 4,767,705
ContextAwareAnalyzingSuggester size[B]: 4,837,187
-- construction time
AnalyzingSuggester input: 53022, time[ms]: 10184 [+- 207.64]
ContextAwareAnalyzingSuggester input: 53022, time[ms]: 1831 [+- 81.89]
-- prefixes: 6-9, num: 7, onlyMorePopular: false
ContextAwareAnalyzingSuggester queries: 53022, time[ms]: 1457 [+- 163.04],
~kQPS: 36
AnalyzingSuggester queries: 53022, time[ms]: 1140 [+- 28.59], ~kQPS: 47
-- prefixes: 100-200, num: 7, onlyMorePopular: false
ContextAwareAnalyzingSuggester queries: 53022, time[ms]: 1276 [+- 58.97],
~kQPS: 42
AnalyzingSuggester queries: 53022, time[ms]: 1004 [+- 81.69], ~kQPS: 53
{code}
>From the above benchmarks, it seems the only improvement for the new suggester
>is in the construction time. The QPS for all three cases seems to be ~20% less
>and the RAM usage is ~3% more.
> Add Context Aware Suggester
> ---------------------------
>
> Key: LUCENE-5350
> URL: https://issues.apache.org/jira/browse/LUCENE-5350
> Project: Lucene - Core
> Issue Type: New Feature
> Components: core/search
> Reporter: Areek Zillur
> Fix For: 5.0, 4.7
>
> Attachments: LUCENE-5350-benchmark.patch,
> LUCENE-5350-benchmark.patch, LUCENE-5350.patch, LUCENE-5350.patch
>
>
> It would be nice to have a Context Aware Suggester (i.e. a suggester that
> could return suggestions depending on some specified context(s)).
> Use-cases:
> - location-based suggestions:
> -- returns suggestions which 'match' the context of a particular area
> --- suggest restaurants names which are in Palo Alto (context ->
> Palo Alto)
> - category-based suggestions:
> -- returns suggestions for items that are only in certain
> categories/genres (contexts)
> --- suggest movies that are of the genre sci-fi and adventure
> (context -> [sci-fi, adventure])
--
This message was sent by Atlassian JIRA
(v6.1#6144)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
