Sentiment Translation through Multi-Edge Graphs

Sentiment analysis systems can benefit from the translation of sentiment information. We present a novel, graph-based approach using SimRank, a well-established graph-theoretic algorithm, to transfer sentiment information from a source language to a target language. We evaluate this method in comparison with semantic orientation using pointwise mutual information (SO-PMI), an established unsupervised method for learning the sentiment of phrases.

• Type: Journal
• Year: 2010
• Where: COLING
• Authors: Christian Scheible, Florian Laws, Lukas Michelbacher, Hinrich Schütze
• Keywords: COLING 2010, Computational Linguistics, Sentiment Analysis Systems, Sentiment Information, Well-established Graph-theoretic Algorithm