Research On Design And Optimization Of Web Resource Reptiles Based On Slicing

With mobile phones and other mobile devices become more and more popular now, mobile learning in gradually becomes one of the important methods of studying. The quality and quantity of resource of mobile learning is very important. With the help of rich Internet, we can use the web crawler to extract much resource for mobile learning, then according to the standard of mobile learning resources, we can chop the resource into suitable clips for building mobile learning resource bank.Because of the time of the mobile learning is fragment, we must extract as much as possible images, video, audio, flash and other multimedia resources that also are related to our theme from network. In the existing basic education platform, we can extract basic resource from network. But further improvement is needed to realize the target of grabbing resource for mobile learning. This paper mainly carried out the following improvement:1. Analysis of the resources for the mobile learning identified the form of the resources in HTML source code.2. Extended the DOM tree structure of webpage in the semantic, similar links seem like exist the same link block. So we segment webpage into some link block according the DOM tree structure and the character of link. Then filtering the noise links and extracting the related links.3. The links are divided into two categories in three dimensions (the form in source page, the form in web page visually, the content that the link pointed to):The common link, the multimedia link. Then we analysis the related text of the two class link and design different strategies to extract the related text of link from the extended DOM tree and the relation of the page that the link exist in and the other pages.