Keyphrases extraction research based on structure of document

Keyphrase is the foundation of text categorization, automatic summary and information retrieval, so the research of automatic keyphrase extraction has important significance. The current keyphrase extraction methods don´t take full advantage of the structural features of the document, too much emphasis on the importance of term frequency, which result in the low accuracy of keyphrase extraction. According to these, the paper proposed a keyphrase extraction method based on structure features of the document. It combined with term frequency, location and length information to automatically extract keyphrases.