Search

In this paper, we present a fast and scalable Bayesian model for improving weakly annotated data — which is typically generated by a (semi) automated information extraction (IE) system from Web documents. Weakly annotated data suffers from incorrect ontological role assignments. Our experimental evaluations with the TAP and a collection of 20,000 home pages from university, shopping and sports Web sites, indicate that the model described here can improve the accuracy of role assignments from 40% to 85% for template driven sites, from 68% to 87% for non-template driven sites. [ABSTRACT FROM AUTHOR]

The increasing number of e-commerce Web sites on the Web introduces numerous challenges in organizing and searching the product information across multiple Web sites. This problem is further exacerbated by various presentation templates that different Web sites use in presenting their product information, and different ways of product information they store in their catalogs. This paper describes the DataRover system, which can automatically crawl and extract all products from online catalogs. DataRover is based on pattern mining algorithms and domain specific heuristics which utilize the navigational and presentation regularities to identify taxonomy, list-of-product and single-product segments within an online catalog. Next, it uses the inferred patterns to extract data from all such data segments and to automatically transform an online catalog into a database of categorized products. We also provide experimental results to demonstrate the efficacy of the DataRover. [ABSTRACT FROM AUTHOR]

In this paper we describe the semantic partitioner algorithm, that uses the structural and presentation regularities of the Web pages to automatically transform them into hierarchical content structures. These content structures enable us to automatically annotate labels in the Web pages with their semantic roles, thus yielding meta-data and instance information for the Web pages. Experimental results with the TAP knowledge base and computer science department Web sites, comprising 16,861 Web pages indicate that our algorithm is able gather meta-data accurately from various types of Web pages. The algorithm is able to achieve this performance without any domain specific engineering requirement. [ABSTRACT FROM AUTHOR]

The Web has established itself as the largest public data repository ever available. Even though the vast majority of information on the Web is formatted to be easily readable by the human eye, "meaningful information" is still largely inaccessible for the computer applications. In this paper, we present automated algorithms to gather meta-data and instance information by utilizing global regularities on the Web and incorporating the contextual information. Our system is distinguished since it does not require domain specific engineering. Experimental evaluations were successfully performed on the TAP knowledge base and the faculty-course home pages of computer science departments containing 16,861 Web pages. Keywords: Semi-structured data, spreading activation, semantic partitioning. [ABSTRACT FROM AUTHOR]