Host: The Japanese Society for Artificial Intelligence
Name : The 33rd Annual Conference of the Japanese Society for Artificial Intelligence, 2019
Number : 33
Location : [in Japanese]
Date : June 04, 2019 - June 07, 2019
Existing emotion estimation system always estimates some emotion from sentences. However, some sentences are difficult to estimate by humans. In order for the dialogue system to estimate emotion like humans, we think that it should also give results that "emotion estimation is difficult" as humans do. Therefore, we aim to construct a system to calculate the estimation difficulty level when estimating the author 's emotions in Japanese sentences. Emotion estimation difficulty was judged by combination of classification by classifier, whether there is negative expression, emotional expression or not. F1 value exceeded 80%.