Personalization on E-Content Retrieval Based on Semantic Web Services

Abstract

In the current educational context there has been a significant increase in learning object repositories (LOR), which are found in large databases available on the hidden web. All these information is described in any metadata labeling standard (LOM, Dublin Core, etc). It is necessary to work and develop solutions that provide efficiency in searching for heterogeneous content and finding distributed context. Distributed information retrieval, or federated search, attempts to respond to the problem of information retrieval in the hidden Web. Multi-agent systems are known for their ability to adapt quickly and effectively to changes in their environment. This study presents a model for the development of digital content retrieval based on the paradigm of virtual organizations of agents using a Service Oriented Architecture. The model allows the development of an open and flexible architecture that supports the services necessary to dynamically search for distributed digital content. A major challenge in searching and retrieving digital content is also to efficiently find the most suitable content for the users. This model proposes a new approach to filtering the educational content retrieved based on Case-Based Reasoning (CBR). It is based on the model AIREH (Architecture for Intelligent Recovery of Educational content in Heterogeneous Environments), a multi-agent architecture that can search and integrate heterogeneous educational content through a recovery model that uses a federated search. The model and the technologies presented in this research exemplify the potential for developing personalized recovery systems for digital content based on the paradigm of virtual organizations of agents. The advantages of the proposed architecture, as outlined in this article, are its flexibility, customization, integrative solution and efficiency.