from pKeicho Is the timing of the value question appropriate?
- https://keicho.netlify.app/#talk=alUKPkDIxF4BN18b4Yoe
- Isn’t it too early to ask the value question?
- Observation of what was happening
- Improved the script I created when observing Keyword appearance speed to allow for Observation of keyword score variation during the dialogue by specifying a single log
see_score.py.
- Improved the script I created when observing Keyword appearance speed to allow for Observation of keyword score variation during the dialogue by specifying a single log
- Observation: Higher scores for “do
- [Re-learn Motion Extraction
- Past work Motion Extraction Work Memo.
- This will improve on this case, but we’ll see what happens in general.
What would you like to see happen in this conversation?
I thought it might be a good idea to mention the 3As in the explanation of the good use of this chat system, so I want to organize this before I write the manuscript.
What kind of “chat system” is this “chat system”?
A system that asks questions in response to user statements. He explained that the system helps users to organize their thoughts by answering questions.
What kind of “user” is this “user”?
It is human and basically alone. And the beauty of this system is that even one person can “organize thoughts through dialogue”.
What kind of “system” is this “system”?
It can be said that the system becomes a dialogue partner instead of a human being. So, 3A is a technique to enhance the dialogue between humans.
What kind of “dialogue” is this “dialogue”?
This one is human-to-human. The 3A stands for “answer, add, ask,” which means to add information and ask more questions about it, rather than just answering the question.
:
prev: 100:answer, 90:ask, 83:add, 75:add information and more it, 57:abbreviation, 53:add information and more, 38:human to human, 30:question, 26:3A, 24:do, 17:information
total: 400:dialog, 176:system, 120:human, 106:do, 105:human to human, 100:user, 100:chat system, 100:answer, 99:mention, 99:techniques for boosting dialog, 99:to boost dialog, 99:3A, 90:. ask, 86:question, 85:thinking, 84:to enliven, 83:add, 82:technique, 81:statement, 80:is, 75:adding information to, 75:organizing, 62:asking questions, 62:opponent, 57:abbreviation, 55:manuscript, 53:adding information to, 48:for statements System for asking questions, 44:System for asking questions. D, 34:Explanation, 33:One, 32:Good use, 17:Information, 16:Goodness in, 8:Mentioning 3A, 0:Can
What is it that you value?
I see, this means that verbs are now extracted by Motion Extraction and “do” is also extracted, and although I thought it was okay because of its low score on its own, it appears at high frequency, so the scores are accumulated and it is recognized as “a developed and important word”.
These two.
- 82:Do
- 24: do
This is not good either.
- 80: Yes.
Just tell them that this is not good enough because it’s done by machine learning.
This page is auto-translated from /nishio/✅価値の質問のタイミングが早い→動きの抽出が過剰 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.