Layout Group Extraction from Web Content for Effective Adaptation

These days, people access the Web by using various devices and methods, such as PDAs, cellular phones, and voice-based browsers. However, most Web content is designed for desktop computers. Therefore, alreadyexisting Web content should be transcoded to be suitable for each access device and method. For this purpose, some annotation-based transcoding systems have been developed. An annotation is additional information of Web content, and effective adaptation can be achieved by using it. One of the most difficult problems of annotation is the cost of annotating Web content. Many popular sites, such as news sites, have a large number of Web pages and add new content continually. Hence, it is almost impossible to annotate all of the content in these sites. To solve this problem, we introduce a method to extract common layouts from Web pages. We focus on the structure and characteristics of particular HTML tags that affect the layout of Web pages. Our method calculates the distance between Web pages using this method. When the distance is below the threshold, these pages can be considered as the same layout pages. By using this method, a certain annotation can be applied to any Web pages that have the same layout. Therefore, the cost of adaptation will be reduced. key words: HTML, layout group, transcoding, adaptation, annotation