Skip to content
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

Boost Scoring/Weighting #5068

Open
1 task done
msj121 opened this issue Jun 2, 2024 · 0 comments
Open
1 task done

Boost Scoring/Weighting #5068

msj121 opened this issue Jun 2, 2024 · 0 comments

Comments

@msj121
Copy link

msj121 commented Jun 2, 2024

Describe your feature request

When searching to allow other factors, like recency or popularity to affect the weighting.

Re-ranking is not suitable - since it might only re-rank the page of 10, not "all" results. To prevent the database from querying every single element to adjust, perhaps a cosine/weight cut-off or a limit of results to be considered. Or a max on the adjustment could cut off what effect this weight adjustment would have.

However, the database would after this weight cut-off (or recognizing no adjustment could become a result), re rank with a boost from other properties, perhaps like so:

// Goal: views will boost the result, similar to a filter perhaps?
additionalFilter: {path[“views”],adjust:0.1,max:0.01};
// max(0.1,1000 views*0.1)+weight
or
additionalFilter: {path[“views”],adjust:0.1,max:0.01, method: “multiply”};
// max(0.01,1000 views*0.1) * weight
or
additionalFilter: {path[“views”],adjust:0.1, max:0.01,percentMax:10000};
// adjust here would be meaningless.
// max(0.01,1000 views/10000)*weight

I think multiple options should be made available to boost - but I am not sure all the possibilities, but even a few would be extremely useful.

Code of Conduct

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant