Research On The Contents Of Web Education Resources

Great promoting along with the mobile Internet era, a large number of information is passed to the public through a variety of mobile terminal, anytime and anywhere everyone can obtain resources quickly through a variety of mobile devices. Recently, network resources based on PC platform is very abundant and various, however, network educational resources which are suitable for mobile terminal pages is very few. So, from the perspective of sharing resources and reducing repeat development, the goal of mobile education in the modern educational theory must be achieved is making people carry out effective mobile learning.This paper will introduce the rich Web education resources which are suitable for mobile learning to the mobile terminal network page, and put forward a valid and generalization technology of Web educational resources content extraction and slicing. The technology includes the content extraction of Web education resource page and the slicing of extracted content. The major contributions of this paper are:1. Proposed a method which base on improved line-block distribution function and more clues to extract the web page content. Our algorithm first preprocess the web page to get the rough web content according to the special HTML tag, the text features of text semantic information, the visual layout features of page structure information on web page. Then, the algorithm extracts the content of the web page combined with improved line-block distribution function.2. We present the granularity of Web educational resource content slice. The slice granularity is the basic element for slicing the resource text. So clearing the size of slice granularity is must for slicing processing. According to different types of resources, the size of slice granularity can be delimited by considering the layout information and the text information of sentences, paragraphs, blocks.3. The algorithm of text slice with HMM model based on pattern recognition is proposed. The algorithm first determines the boundary of subtopics in the text on the basis of the text semantic features and visual layout features combined with the regular expression match, then uses the HMM model to judge the boundary in order to slice the text. Experiments show that our algorithm is very effective, and the HMM model is ideal for text slice.4. We elaborate on the process of the web page content extraction and slicing with teaching plan and test paper as examples. Because of the much more types of education resources, the processing methods are different. So this paper describes the concrete process of the teaching plan and test paper which are special education resources in detail. Finally, we conduct experiments on the random crawled pages from the Internet to prove the validity and generalization ability of the algorithms for the resources of teaching plan and test paper.