CoDE Publications CoDE Publications
IRIDIA Publications IRIDIA Publications
SMG Publications
WIT Publications
WIT Publications
SMG Publications
Home People Research Activities Publications Teaching Resources
By Class By Topic By Year Technical Reports
By Class By Topic By Year Technical Reports
login
J.-P. Norguet and E. Zimányi. Topic-Based Audience Metrics for Internet Marketing by Combining Ontologies and Output Page Mining. In Proceedings of the International Conference on Intelligent Agents, Web Technology and Internet Commerce, IAWTIC. IEEE Computer Society Press, Austria, Vienna, November 2005.
© IEEE Computer Society Press 2005

Abstract

In Internet marketing, Web audience analysis is essential to understanding the visitors' needs. However, the existing analysis tools fail to deliver summarized and conceptual metrics needed by organization managers and Web site editors. The reason is that HTTP transaction metadata mined by these tools do not include the text content sent to the browsers. In this paper, we first describe the various methods that we conceived to mine the Web pages output by Web servers. These methods include content journaling, script parsing, server monitoring, network monitoring, and client-side mining. Then, for a given ontology, we count the occurrences of ontology entries in the mined pages, and we compare the results to the term weights in the online pages. By aggregating the metrics in the ontology, we obtain audience metrics which should represent the Web site topics. Finally, we validate our approach with experiments on real data using SQL Server OLAP and our prototype WASA.


Updated: 2017-03-27