• I think the choice of keywords and questions is an attention mechanism.

    • Some time after I wrote this, I came to the conclusion that [Keyword and question selection is an internal volume caution.
  • As for the lone keyword/question pair, that’s exactly what it is.

    • Now we’re multiplying the “keyword score” by the “score for the degree to which the keyword is likely to apply to the question.
  • Questions that do not take a keyword or two keywords exist.

    • The mediation there is a mystery.
    • If you have promising keywords, you want to drill down on those keywords
    • If multiple high scoring keywords come up, I’d like to hear the relationship between those keywords.
    • It would be nice to have a recap of the initial objective so we don’t go off topic.
      • = Attention to conversation in the early stages is maintained.
  • What is “promising” and “high scoring?” image


This page is auto-translated from /nishio/キーワードと質問の選択は注意機構 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.