generated from IBM/repo-template
-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Register logs of requests to allow us track interaction with recommendations #20
Labels
enhancement
New feature or request
Comments
Hi @santanavagner, I started working on this task |
cassiasamp
added a commit
that referenced
this issue
Feb 5, 2025
Hi @santanavagner commit 0efb13e closes part of this on the server side. This is how the log is currently structured:
|
cassiasamp
added a commit
that referenced
this issue
Feb 6, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Description (Actual Behavior)
The current prototype has no tracking mechanism regarding recommendations that users accepted/rejected. Hence, we should have a way to log requests submitted to the recommend/ endpoint and subsequent turns for the same session. This way, we can perform an approach similar to search-log analysis to track recommendations that were added to the prompt as-is or sentences that were rephrased by users.
PS: This analytics feature will be developed after having this logging mechanisms.
Expected Behavior
Our API will register prompts sent to the recommend/ endpoint in a way to keep the history of how the prompt evolved, which allows identifying recommendations that users accepted, the values users used the most, and sentences that users rephrased before submitting to a GenAI.
Log lines should have, at least, timestamp, prompt, and session ID.
Possible Fix
The following stack overflow can be a starting point to add such log:
https://stackoverflow.com/questions/34632730/flask-how-to-store-logs-and-add-additional-information
Moreover, we can also create an endpoint only to register when users click on the recommendations, so we can also register when users interacted with social values and recommendations. This last aspect could be organized in an additional issue as soon as we have the logging mechanism in place.
Steps to Reproduce
NA
Context
API requests.
The text was updated successfully, but these errors were encountered: