Research and Implementation on Multi-Cues Based Page Segmentation Algorithm

In view of that the exiting segmentation algorithms use only one kind of cue, but while users viewing the Web page, human brain can handle many cues and segment the pages subconsciously. So this paper considers many kinds of cues, simulates the process of user-perception, analyzes the structure of Web pages and proposes the multi-cues based page segmentation algorithm. After the segmentation, semantic information is added to every semantic block.