Keyword Extraction of Document Based on Weighted Complex Network

Abstract

This document explains and demonstrates how to extract keyword from Chinese document based on weighted complex network. The characteristic and disadvantages of several common automatic keyword extraction methods are introduced firstly. Then based on the ideas of complex network, we proposed an improved automatic keyword extraction method. Using complex network, a Chinese document is first represented as a network: the node represents the term, and the edge represents the Co-Occurrence of terms. Then we calculate the integrate value of each term, the keywords are top k terms with greatest value. The experiment results show that the method is more effective and accurate in comparison with the traditional method TFIDF keyword extraction from the same document.