Emotion and Cause: Linguistic Theory and Computational Implementation

Description

This work argues that cause events, being the most tangible component of emotion, provide a rich dimension of how emotions should be classified. While it is often claimed that emotional concepts cannot be defined, this work views emotion as a response triggered by actual or perceived events, specifically focusing on the interaction between five primary emotions (Happiness, Sadness, Fear, Anger, and Surprise) and cause events. Cause events are examined in terms of two dimensions, namely transitivity and epistemicity. By incorporating the semantic and syntactic information of emotion cause events, this representation of emotion not only provides deep linguistic criteria of emotion cause events, but also offers an event-based approach to emotion classification. A text-driven, rule-based system for detecting the causes of emotion is then developed to establish the validity of the proposed linguistic model for emotion detection and classification. The system shows promising results. Front Matter ....Pages i-xii Towards a Linguistic Theory of Emotion and Expression of Emotion (Sophia Yat Mei Lee)....Pages 1-16 The Linguistic Expression of Emotion and Cause in the Chinese Language (Sophia Yat Mei Lee)....Pages 17-36 Linguistic Resources for Study of Emotion (Sophia Yat Mei Lee)....Pages 37-45 Linguistic Expression of Cause Event I: Transitivity (Sophia Yat Mei Lee)....Pages 47-68 Linguistic Expression of Cause Event II: Epistemicity (Sophia Yat Mei Lee)....Pages 69-85 A Linguistic Model for Emotion Expression (Sophia Yat Mei Lee)....Pages 87-100 Implementation and Verification: Automatic Detection of Emotion Causes (Sophia Yat Mei Lee)....Pages 101-142 Back Matter ....Pages 143-151