Key Technologies Of Web Media Resources Intelligent Discovery

With the rapid development of information technology, especially the extensive use of Web 2.0, the user participation significantly improved. Online users can not only enjoy the web videos but also be a resource provider, which brings about the rapid increase of web videos. The request of search and monitor become an urgent problem.This pager focuses on how to mining the videos resource from the Internet quickly, accurately and comprehensively. In depth study of the working principle of crawler, an evaluation model is proposed. In order to improve the web video resource features library, we design an online incremental learning algorithm. After that, optimization on the whole system is ongoing taken place. Thus the efficiency of the system is effectively enhanced. And it achieves the performance of the actual operation in the end. The main outcome of this article can be summarized as follows:1. According to the features and distribution of the web video resources on the Web, we build a knowledge-based system of media resources. The features library contains static and dynamic video resources knowledge, which can be used for evaluating the similarity of a web page. The construction of web video features library is the base of algorithm and system model.2. Based on the web video resource features library, combined with the idea of re-crawl policy, we take example by focus crawler, and put forward a web video mining model. After analysis of the static mining system experiment result, we can reach this conclusion that the system base on static features library can meet the demand well.3. After a series of analysis and experiment, a new crawler strategy Best First was proposed. In order to improve the evaluate capabilities of video discovering, we import the incremental learning algorithm for the web video resource keyword features, which can adjust and replacement.4. Based on the above results of theoretical studies, this paper describes design and development of a high-performance web video resource mining system.