Search

Distributional similarity has been widely used to capture the semantic relatedness of words in many NLP tasks. However, parameters such as similarity measures must be manually tuned to make distributional similarity work effectively. To address this problem, we propose a novel approach to synonym identification based on supervised learning and distributional features, which correspond to the commonality of individual context types shared by word pairs. This approach also enables the integration with pattern-based features. In our experiment, we have built and compared eight synonym classifiers, and showed a drastic performance increase of over 60% on F-1 measure, compared to the conventional similarity-based classification. Distributional features that we have proposed are better in classifying synonyms than the conventional common features, while the pattern-based features have appeared almost redundant.

Distributional similarity is a widely adopted concept to compute lexical semantic relatedness of words.Whereas the calculation is based on the distributional hypothesis and utilizes contextual clues of words, little attention has been paid to what kind of contextual information is effective for the purpose.As one of the ways to extend contextual information, we pay attention to the use of indirect dependency, where two or more words are related via several contiguous dependency relations.We have investigated the effect of indirect dependency using automatic synonym acquisition task, and shown that the performance can be improved by using indirect dependency in addition to normal direct dependency.We have also verified its effectiveness under various experimental settings including weight functions, similarity measures, and context representations, and shown that context representations which incorporate richer syntactic information are more effective.

Distributional similarity is a widely adopted concept to capture the semantic relatedness of words based on their context in various NLP tasks. While accurate similarity calculation requires a huge number of context types and co-occurrences, the contribution to the similarity calcualtion depends on individual context types, and some of them even act as noise. To select well-performing context and alleviate the high computational cost, we propose and investigate the effectiveness of three context selection schemes: category-based, type-based, and co-occurrence based selection. Categorybased selection is a conventional, simplest selection method which limits the context types based on the syntactic category. Finer-grained, type-based selection assigns importance scores to each context type, which we make possible by proposing a novel formalization of distibutional similarity as a classification problem, and applying feature selection techniques. The finest-grained, co-occurrence based selection assigns importance scores to each co-occurrence of words and context types. We evaluate the effectiveness and the trade-off between co-occurrence data size and synonym acquisition performance. Our experiments show that, on the whole, the finest-grained, co-occurrence based selection achieves better performane, although some of the simple category-based selection show comparable performance/cost trade-off.