Emotion Prediction System for Japanese Language Considering Compound Sentences, Double Negatives and Adverbs
Rafał Rzepka, Mitsuru Takizawa, Kenji Araki · Institutional Repositories DataBase (IRDB) · 2016
In this paper we introduce an algorithm that is ca- pable of recognizing emotions of user’s statements in order to achieve more effective and smoother human-machine conversation. Many studies of the emotion recognition have been actively conducted in order to quantify affect, but it is rather difficult to recognize it from more complicated sentences, often having double negatives. We describe our enhancements of emotion recognizer by combin- ing emotive expressions lexicon, web-mining tech- niques, processing compound sentences, and ad- verb weighting for emotiveness degree modifica- tion. The effectiveness of the proposed algorithm for recognizing more complicated sentences was confirmed through evaluation experiments which results are also introduced.