OPSHNN: Ontology Based Personalized Searching Using Hierarchical Neural Networks Evidence Combination

In this paper we propose a novel ontology based personalized searching method called OPSHNN which uses hierarchical neural networks to classify documents into concepts in the reference ontology. The user profile is modeled as a weighted concept hierarchy. We use weighing methods based on the user¿s surfing pattern to weigh the concepts in the reference ontology. The system adapts itself to the changing interests of the user by means of aging. To overcome the problem of training in cases where insufficient documents are available for a particular concept and to increase the scalability we propose to use two different hierarchical neural network classifiers, each using a different learning function. Their beliefs are combined using Dempster-Shafer theory to eliminate any weaknesses in classification into concepts in the ontology. Results show that our system has superior classification accuracy, convergence and a precision of 16%.