Contextual Sentiment Analysis
Principal Investigator: Hinrich Schütze
Researchers: Christian Scheible
Sentiment analysis is a highly context-dependent task. Many linguistic units can express a completely different sentiment depending on the topic or the situation in which they are used. This project aims at modeling this context-dependency through a generative statistical model which is able to represent relationships between words, to generalize these relationships, and to capture topical relations. Furthermore, interlingual contexts are modeled through graph-based representations.