News & Events

Adjectives are quite important to detect the polarity (i.e. positive, negative or neutral) of texts. Some adjectives have definite polarity in different contexts, such as the positive ones including good, excellent, and the negative ones including bad, awful. However, some other adjectives are neutral in polarity out of context, but could show positive or negative meaning within specific contexts. Such adjectives are called sentiment-ambiguous adjectives (SAAs). For instance, "價格高|the price is high" indicates negative meaning, while "質量高|the quality is high" has positive connotation.

Disambiguating sentiment-ambiguous adjectives is an interesting task, which is an interaction between word sense disambiguation (WSD) and sentiment analysis. This presentation introduces our study on disambiguating the polarity of SAAs. To disambiguate SAAs, we compare the machine learning-based and the lexicon-based methods:

1) Maximum entropy is used to train classifiers based on the annotated Chinese data from the NTCIR opinion analysis tasks, and the clause-level and sentence-level classifiers are compared;

2) For the lexicon-based method, we first classify the adjectives into two classes: intensifiers (i.e. adjectives intensifying the strength of context) and diminishers (i.e. adjectives decreasing the strength of context), and then use the polarity of context to get the SAAs??contextual polarity based on a sentiment lexicon.

The results show that the performance of maximum entropy is not quite high due to little training data; on the other hand, the lexicon-based method could improve the precision by considering the context of SAAs.