confidance threshold to word embedding #225
Closed
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
- Improved word embedding preprocessing - Updated read me for embedding generation
KaitCrawford
previously approved these changes
Jan 15, 2024
home/tests/test_api.py
Outdated
| assert content["count"] == 0 | ||
| # it should not return search term matching pages if they are unpublished | ||
| page1.unpublish() | ||
| response = uclient.get("/api/v2/pages/?s=#(&whatsapp=true") |
There was a problem hiding this comment.
This is the same search term that previously didn't return any results. You probably want to use s=help here for the test to be valid
There was a problem hiding this comment.
Just to add to this, I don't think we should be using the # character in the URL for our tests for this. It's the fragment identifier character, and actually needs to be escaped if you want to use it in URL parameters, otherwise it could result in everything after the # being interpreted as a fragment identifier and not as part of the query string: https://www.rfc-editor.org/rfc/rfc3986.html#section-3.5
updated readme
|
I'm closing this PR since word embeddings are no longer a part of the CMS |
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Outdated suggestions cannot be applied.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.
Purpose
The aaq search was returning results even in the cases when the user has entered meaningless invalid text
https://praekelt.leankit.com/card/31512186572175
Approach
Added a confidence threshold so that we can filter out any returned results that have a similarity score of less than 25%
Open Questions and Pre-Merge TODOs
Learning
From analysing how different search inputs compare to the results, in terms of similarity score, 25% similarity gave a balance between giving back relevant answers and filtering out completely unrelated responses
Blog Posts