... Association for Computational Linguistics, pages 123–131,Portland, Oregon, June 19-24, 2011.c2011 Association for Computational LinguisticsAutomatically Extracting Polarity-Bearing Topics for Cross-Domain Sentiment ... allocates 5%of probability mass for mixing.6.2 Supervised Sentiment ClassificationWe performed 5-fold cross validation for the per-formance evaluation of supervised sentiment clas-sification. Results ... S ×V which we use to en-code word prior sentiment information into the JSTmodel. For each word w ∈ {1, , V }, if w is foundin the sentiment lexicon, for each l ∈ {1, , S}, theelement λlwis...