ICACT20230036 Question.1
Questioner: ida@niad.ac.jp 2023-02-24 오후 8:11:23 |
ICACT20230036 Answer.1
Answer by Auhor ycc@nttu.edu.tw 2023-02-24 오후 8:11:23
Chrome  Click!! |
| Could you explain more about the comparison with related researches? |
In our paper,the non-weighted k-mean cluster and weighted k-mean cluster has almost same cluster point.However,There is a different point between two pictures in slide 11,there is a point where the right hand has been circle.It is because non-weighted has a point ,but it is out of bound,and the weighted one is approaching the elders who has a high level to take care |
ICACT20230036 Question.2
Questioner: namnh@vnu.edu.vn 2023-02-24 오전 11:54:42 |
ICACT20230036 Answer.2
Answer by Auhor ycc@nttu.edu.tw 2023-02-24 오전 11:54:42
Chrome Click!! |
| Why did you choose the K-mean instead of other clustering methods? |
We have to think about another method k-medoid to solve the problem in the project, but the k-medoid theory takes an actual sample point. In my project, there are many different types of elderly people who need to be taken care of, we can't take an elderly to be a cluster point. Otherwise, the deployment of the police force will lose the meaning of caring for seniors with high weighted |
ICACT20230036 Question.3
Questioner: namnh@vnu.edu.vn 2023-02-24 오후 8:27:34 |
ICACT20230036 Answer.3
Answer by Auhor ycc@nttu.edu.tw 2023-02-24 오후 8:27:34
Chrome Click!! |
| did you tried with other non linear distance instead of weighted K-mean clustering method? what will be happen at that case? |
Thank you for your question. To date, we have not explored other non-linear distance methods. Our current method uses non-linear weights for each care level, with a weight of 81 assigned to the highest level, 16 to the middle level, and 1 to the lowest level. As a result, our method achieves a similar distance effect as non-linear distance methods. |
