Homepage Live: Automatic Block Tracing for Web Personalization
Jie Han (Shanghai Jiaotong University)
Dingyi Han (Shanghai Jiaotong University)
Chenxi Lin (Microsoft Research Asia)
Hua-Jun Zeng (Microsoft Research Asia)
Zheng Chen (Microsoft Research Asia)
Yong Yu (Shanghai Jiaotong University)
The emergence of personalized homepage providers, e.g. Google Homepage and Microsoft Windows Live, has enabled Web users to select interesting Web contents and to aggregate them in a single Web page. The interesting web contents are usually some specific blocks of Web pages rather than a whole Web page. To satisfy the users' requirements, personalized homepage providers predefine a lot of candidate content blocks for users to composite their own homepages. However, it requires tremendous manual efforts to define the content blocks and the coverage is still very limited. In this paper, we propose a novel personalized homepage system, called "Homepage Live", to allow users to use drag-and-drop action to collect their favorite Web content blocks and organize them in a single page. Moreover, Homepage Live can also automatically trace the changes of blocks with the evolvement of the containing pages by measuring the tree edit distance of the selected blocks. Besides, by exploiting the immutable elements of Web pages, the tracing algorithm performance is significant improved. Experimental results demonstrate the effectiveness and efficiency of our proposed algorithm.
New Brunswick, Wednesday, May 9, 2007, 1:30pm to 3:00pm.