There is a wealth of information to be mined from the World Wide Web. Unfortunately, standard natural language processing (NLP) extraction techniques perform poorly on the choppy, semi-structured information fragments, such as sports results, which are popular to be published on the Web pages nowadays. In this paper, we present an information agent: SportsFinder, an agent to extract sports scores from the World Wide Web, as well as the knowledge discovering method to learn new express patterns to improve the agent's performance.
Lecture notes in computer science: proceedings of the 3rd Pacific-Asia Conference on Methodologies for Knowledge Discovery and Data Mining (PAKDD-99), Beijing, China, 26-28 April 1999 / Ning Zhong and Lizhu Zhou (eds.),
Vol. 1574, pp. 469-473