Barrier Features for Classification of Semantic Relations.

Abstract

Approaches based on machine learning, such as Support Vector Machines, are often used to classify semantic relations between entities. In such framework, classification accuracy strongly depends on the set of features which are used to represent the input to the classifier. We are proposing here a new type of features, namely the barrier features, which can be used in addition to more usual features, such as n-grams of PoS, word suffixes and prefixes, hypernyms from WordNet etc., and to the parse tree of the whole sentence. Barrier features aim at giving a compact representation of the context of each entity involved in the relation. The effectiveness of the new features is assessed on documents from the TREC data set annotated by Roth and Yih. The obtained results show not only that the performance of the proposed approach are state-of-the-art but also that such improvement is due to the introduction of the barrier features. 1