Web Page Filtering Based On Program Slicing

The rapid development of the Internet has to become the world’s most extensive coverage, the largest and most abundant resources of the information network. the Internet has become the main way of getting information. People can input the content that they want to query in the search engine, search out the information. But as the exploding of information, a variety of unrelated information or advertising sneaks in, and the genuine and fake is difficult to distinguish. Facing the gigantic information resources,how to obtain valuable information on the current become a very important question.Traditional network Information filtering technology contrast through digging into URL and text in web page with blacklists in its own database. This way not only consumes too much time and resources, but also increases the server capacity. This test apply the program slicing technology to network information filtering, by forming HTML statement into tree Diagram, to match key words with leaves in tree diagram, extract the line number of parent nodes of nodes that match successfully, work out the criterion of slice,, to form dependency graphs through the new dependency relationship in web page code. To slice the dependency graph on the basic of slice criterion, and then get Slice set,only keep the set of sentence related to slice criterion. At last, it will be revert to Visualization of web page. The network information filtering technology realized in this test not only can filter needless things, but also extract the message that customers are interested in. What’s more, the degree of filtering can be set in different degree.this way not only is fast, but also have a low requirement for the bearing capacity of the server, also can realize all kinds of personalize filtering capabilities.