Authors:
Hilal Tarakci
and
Nihan Kesim Cicekli
Affiliation:
Middle East Technical University, Turkey
Keyword(s):
User Modeling, User Profile, Hypergraph-Based User Model, Knowledge Representation, Recommendation System
Related
Ontology
Subjects/Areas/Topics:
Applications and Case-studies
;
Artificial Intelligence
;
Collaboration and e-Services
;
Data Engineering
;
e-Business
;
Enterprise Information Systems
;
Information Integration
;
Integration/Interoperability
;
Knowledge Engineering and Ontology Development
;
Knowledge Representation
;
Knowledge-Based Systems
;
Ontologies and the Semantic Web
;
Process Knowledge and Semantic Services
;
Semantic Web
;
Soft Computing
;
Symbolic Systems
Abstract:
We propose a hypergraph-based user profile which facilitates aggregating partial profiles of the individual and obtain a complete, multi-domain user model. The aggregation involves a semantic enhancement procedure which results in enriched user profiles. The proposed user model is capable of extracting general and domainbased user profiles and answering several connected data queries such as recommendation, in reasonable time. In this paper, we present a recommendation case study which uses the proposed user model and illustrate the traversal algorithms for a variety of connected data problems.