If you’re evaluating a question that takes multiple keywords in a sum, when one of the keywords is dangling higher than the other, you’ll get a question that takes multiple keywords before you can dig into the newly emerged keyword.
I think you should min.
Create a script that will pour in user input from the logs and take statistics,
- We can check to see if some of the conditions are met as part of that.
- When the sample log was poured, there were 10 infringements, but by setting the main evaluation to min, there were 1 infringement, and by setting the score for the sum to 0.01, there were 0.
- This last adjustment is inappropriate for some score ranges and needs to be better.
This page is auto-translated from /nishio/掘り下げてないキーワードに関係の質問をしない using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.