An Integrated Crawling Strategy Based on Hyperlinks and Content Similarity
NI Xian-gui · Microcomputer Information · 2008
This paper introduces the web analysis algorithm: the HITS algorithm then improves the HITS algorithm and proposes a VHITS (VSM-HITS) algorithm based on hyperlinks and content similarity. The VHITS algorithm uses VSM (Vector Space Model) to evaluate the importance of hyperlinks. Thus it avoids the topic drifting away in HITS algorithm, and performs quite well in the fo-cused crawling.