Authors:
Vasileios Lapatas
and
Michalis Stefanidakis
Affiliation:
Ionian University, Greece
Keyword(s):
Web personalization, Personalized web sites and services, Web site classification, Learning user profiles.
Related
Ontology
Subjects/Areas/Topics:
Data Engineering
;
Ontologies and the Semantic Web
;
Personalized Web Sites and Services
;
Web Information Systems and Technologies
;
Web Interfaces and Applications
;
Web Personalization
Abstract:
An interesting promise of Web 3.0 is the seamless integration of desktop and web spaces. Private data, locked until recently inside a user’s computer, can lead to the intelligent generation of web content. This paper presents an idea of how desktop and web data can be integrated in creative ways in the World Wide Web. As a proof of concept, an application which can profile a user based on his bookmarks is being demonstrated. An existing web service is used to classify bookmarks, enabling thus platforms with limited processing power to perform the profiling process. Results gathered from the classification process indicate that even a generic untrainable and not fine-tuneable classifier can produce results with high accuracy. With accurate user profiles web content can be created in an intelligent way, enabling better Web 3.0 applications.