Context Analysis for Computer-Assisted Near-Synonym Learning

Despite their similar meanings, near-synonyms may have different usages in different contexts. For second-language learners, such differences are not easily grasped in practical use. This chapter introduces several context analysis techniques such as pointwise mutual information (PMI), n-gram language model, latent semantic analysis (LSA), and independent component analysis (ICA) to verify whether near-synonyms do match the given contexts. Applications can benefit from such techniques to provide useful contextual information for learners, making it easier for them to understand different usages of various near-synonyms. Based on these context analysis techniques, we build a prototype computer-assisted near-synonym learning system. In experiments, we evaluate the context analysis methods on both Chinese and English sentences, and compared its performance to several previously proposed supervised and unsupervised methods. Experimental results show that training on the independent components that contain useful contextual features with minimized term dependence can improve the classifiers’ ability to discriminate among near-synonyms, thus yielding better performance.