Extracting semantic relations between vietnamese legislative documents

The semantic relations between entities in documents play vital roles in many applications such as information extraction, and information retrieval. The problem of extracting the semantic relations in Vietnamese documents is more difficult because Vietnamese textual analysis tools are in progress. In this paper, we propose an approach based on syntactic dependency feature between provisions in each sentence and the Support Vector Machine classifier to extract the semantic relations between entities in Vietnamese documents. Our results show that syntactic dependency method is more stable and accuracy (F-measure) than the regular expression method especially for long and complex texts of Vietnamese legislative documents.