Research On Technology Of Retrieving Dynamic Web Advertising Intelligently

With the rapidly development of the Internet and electronic commerce, many companies promote their products in web advertising. Many large Internet companies’ major income source come from web advertising. The study of web advertising is increasingly popular. In order to study the Internet advertising and use Internet resources more scientific, collecting the Internet adverting has become a scientifc work. However, there is little achievement in getting Internet advertising for now. Therefore, this article mainly focus on how to obtain web advertising from the huge amount of Internet resources as much as possible.In order to retrieve web advertising quickly and accurately, the major contributions of this thesis are described as follows:First, this thesis presents an algorithm to analyze the dynamic page based on dom tree construction. This algorithm will execute javascript on the page by using the designed javascript engine in this thesis which is based on the Rhino analytical engine. And it will load dynamic data from the server and finally get the whole page content. As the engine allocates resources rationally in the process of requesting data, it is faster than other methods. In this thesis, we compared its efficiency with other methods. The results are better compared to the previous works.Second, a web advertising resource location algorithm is designed in this thesis based on page block classification algorithm. It transforms a resource location problem to a classification problem which is based on page segmentation. After segmentation, all blocks are divided into advertising and non-advertising.Besides, url is used as the property of classification in web advertising resource location algorithm which is based on page block classification algorithm. It is higher in speed and more accurate in advertising location. After that, comparison between several classification algorithms is provided to show the effect on advertising location. Besides, the paper also gives the effects of advertising location algorithm based on decision classification tree on various type of websites. In the end, this thesis proved that the advertising location algorithm based on decision classification tree is better than others.Finally, this thesis implements a system of retrieving dynamic web advertising, validates the algorithm proposed in this thesis with the actual result.