Learning Web Page Block Functions using Roles of Images

Abstract

Making use ofblock information in Web IR and Data Mining tasks calls for a good understanding of the function of each block. Existing works on classifying block functions andjudging block importance have not made full use of the image factor, and only simple image features were considered. We regard image as a strong indicator of Web page blocks with various functions and propose to learn block functions using roles of images as part of block features. Blocks are generated from Web page segmentation and roles of images are automatically decided by image classification. We experiment on 140 Web pages and demonstrate that utilizing roles of images can significantly improve the classification quality of learning Web page block functions. We also measure the usefulness of diferent roles of images and evaluate the effect oftwo page segmentation methods.

Turn the Page: Automated Traversal of Paginated Websites

“…The authors list site navigation elements among their classification labels, but the F 1 value they achieve for this block (82%) is significantly lower than our result for pagination links (99%) and the definition includes navigation menus and similar blocks. The same is true for [17] and [9]. They also consider navigation blocks, but with a wider definition and significantly lower accuracy (87% and 53%).…”