Emotions are found to be frequently expressed in implicit ways. However, most existing approaches solely deal with the emotions explicitly described using emotion-bearing words. This work aims to examine an important yet underdeveloped branch of emotion analysis, i.e. implicit emotion. By constructing and analyzing the Chinese emotion annotated corpus, we recognize the importance of implicit emotion analysis. We propose a list of linguistic cues, such as adjectives and adverbs, for implicit emotion detection in Chinese. It shows that adverbs, in particular, are of great value in detecting implicit emotions when no other cues are found in text, as some adverbs have obvious semantic orientations.
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||16th Workshop on Chinese Lexical Semantics Workshop, CLSW 2015|
|Period||9/05/15 → 11/05/15|
- Theoretical Computer Science
- Computer Science(all)