Research on Pattern Representation Method in Semi-supervised Semantic Relation Extraction Based on Bootstrapping

Abstract

Semantic relation extraction is an important part of information extraction, it has application value in the automatic question answering system, retrieval system, ontology learning, semantic web annotation, and many other areas. Pattern representation method is context pattern in previous semi-Supervised semantic relation extraction based on bootstrapping, but it did not consider the role of the keywords in the semantic relation. This paper presents an improved context pattern, which has a stronger semantic expressiveness, which is used to extract semantic relations and makes the semantic relation extraction more accurate. First of all, the sentence context pattern is obtained by lexical analysis. Then, the syntax tree model is obtained by syntactic analysis, calculate words weight using the syntax tree pattern. Finally, extract semantic relations using semi-Supervised machine learning method based on bootstrapping. The experimental results show that this method can effectively extract the semantic relations.