互联网视频信息获取技术研究与实现-模式识别与智能系统专业论文.docxVIP

A8STRAC丁ABSTRACT A8STRAC丁 ABSTRACT In recent years，along with the vigorous development of web video service the Internet， the video data integl’ation has become the urgent demand of Intemet application Video information is put at countless pages，whose sh’ucture and content very different， therefore，in order to obtain the Internet video information，the following two questions must be solved I)How to obtain these video information page with loose； 2)How wi ll these video information fiom the web page design Usually these two problems not isolated．Generally speaking。its easier to design relatively effective information extraction method for pages whose stt‘ucture and content have some charactel。istic than not．In OI‘der to design efficient and feasible information extraction method，we also want the pages not completely desultorily，but cel’tain kind of web pages． Based the above two problems to solve．this paper designs cornplete solutions of acquisition ofthe Internet video information．The plan includes： 1)In respect of page acquisition，aimin—P at the characteristics of web video service，this paper locks the pages in search result pages and video playing pages，and studies the corresponding acquisition inethod．Acquisition of search result pages is aimed at large video Sel。vice website．Acquisition of video playing pages is aimed at small video service website．The steps of acquisiteing seal’ch results page include：a)positioning web database entry，b)the analog input seal‘ch words，click search button．obtains the search results page entrance，c)with different search words input，get the COl‘responding seal‘ch results page．The steps of acquisiting playing page include：a)use generic web Cl‘awler to obtain pages’b1 through the judge of three elements，which includes characteristic script URL．HTML text of player node’parent node， templates of playing pages，to identify whethel‘OI‘not page is playing page 2)In respect of web information extraction，combining the structure character an