Combining Lexico-semantic Features for Emotion Classification in Suicide Notes.

摘要: This paper describes a system for automatic emotion classification, developed the 2011 i2b2 Natural Language Processing Challenge, Track 2. The objective of shared task was to label suicide notes with 15 relevant emotions on sentence level. Our uses SVM models (one each emotion) using combination features that found perform best given emotion. Features included lemmas and trigram bag words, information from semantic resources such as WordNet, SentiWordNet subjectivity clues. best-performing labeled 7 achieved an F-score 53.31% test data.