Ensure cross_fields always uses valid term statistics #90278
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jtibshirani
added
>regression
:Search Relevance/Ranking
|
Pinging @elastic/es-search (Team:Search) |
|
Hi @jtibshirani, I've created a changelog YAML for you. |
|
I marked this as "non-issue" since it is unreleased on 8.5 and 7.17. However it's a regression in 8.4.2. When I backport I'll make sure to mark it "regression" and add a changes entry. |
|
@elasticmachine run elasticsearch-ci/part-1 |
jtibshirani
added a commit
that referenced
this pull request
Sep 23, 2022
In #89016 we adjusted the `cross_fields` scoring formula to prevent negative scores. This fix accidentally dropped another important fix that was added in #41938. Specifically, we need to make sure to take the minimum between the document frequency (`actualDf`) and the minimum total term frequency (`minTTF`). Otherwise, we can produce invalid term statistics where the total term frequency is less than the document frequency. Fixes #90275
jtibshirani
added a commit
to jtibshirani/elasticsearch
that referenced
this pull request
Sep 23, 2022
In elastic#89016 we adjusted the `cross_fields` scoring formula to prevent negative scores. This fix accidentally dropped another important fix that was added in elastic#41938. Specifically, we need to make sure to take the minimum between the document frequency (`actualDf`) and the minimum total term frequency (`minTTF`). Otherwise, we can produce invalid term statistics where the total term frequency is less than the document frequency. Fixes elastic#90275
jtibshirani
added a commit
that referenced
this pull request
Sep 23, 2022
In #89016 we adjusted the `cross_fields` scoring formula to prevent negative scores. This fix accidentally dropped another important fix that was added in #41938. Specifically, we need to make sure to take the minimum between the document frequency (`actualDf`) and the minimum total term frequency (`minTTF`). Otherwise, we can produce invalid term statistics where the total term frequency is less than the document frequency. Fixes #90275
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In #89016 we adjusted the
cross_fieldsscoring formula to prevent negativescores. This fix accidentally dropped another important fix that was added in
#41938. Specifically, we need to make sure to take the minimum between the
document frequency (
actualDf) and the minimum total term frequency(
minTTF). Otherwise, we can produce invalid term statistics where the totalterm frequency is less than the document frequency.
This PR adds two safeguards to prevent this sort of regression in the future:
BlendedTermQueryTests#testRandomFieldsthat adds a fairlylarge number of documents with random fields and values. I tried running this
test 100 times, and checked it would have caught this issue.
Fixes #90275