Model-Driven Web Page Segmentation for Non Visual Access

Abstract

Web page segmentation aims to break a large page into smaller blocks, in which contents with coherent semantics are kept together. Within this context, a great deal of approaches have been proposed without any specific end task in mind. In this paper, we study different segmentation strategies for the task of non visual skimming. For that purpose, we propose to segment web pages into visually coherent zones so that each zone can be represented by a set of relevant keywords that can be further synthesized into concurrent speech. As a consequence, we consider web page segmentation as a clustering problem of visual elements, where (1) a fixed number of clusters must be discovered, (2) the elements of a cluster should be visually connected and (3) all visual elements must be clustered. Therefore, we study variations of three existing algorithms, that comply to these constraints: K-means, F-K-means, and Guided Expansion. In particular, we evaluate different reading strategies for the positioning of the initial K seeds as well as a pre-clustering methodology for the Guided Expansion algorithm, which goal is to (1) fasten the clustering process and (2) reduce unbalance between clusters. The performed evaluation shows that the Guided Expansion algorithm evidences statistically increased results over the two other algorithms with the variations of the reading strategies. Nevertheless, improvements still need to be proposed to increase separateness.

Extracting the Main Content of Web Pages Using the First Impression Area

“…The content can be divided horizontally with <HR> tag. These features were used to distinguish areas on the web page rather than extracting the main content directly [3], [18], [37]. Considering some tags give a implicit information about the content, recent methods use these features to preprocess the web page.…”

Multi-Purpose Dataset of Webpages and Its Content Blocks: Design and Structure Validation

“…That is why often visual representation is used for block identification. One of the most popular algorithms for this is VIPS [18][19][20]. Akpinar et al took the VIPS algorithm as a base and further improved it by adding HTML5 tags under original tag sets, improved handling of invisible nodes, and ported the algorithm to Java [21].…”

Web Page Segmentation Revisited

“…Inspired by their approach, we integrate our version with Mechanical Turk to enable manaul segmentation at scale via crowdsourcing. 3 Evaluation of Web Page Segmentation. Previous attempts to evaluate web page segmentations fall short in some respects or others.…”