Uma abordagem para recuperação de informações sensível ao contexto usando retroalimentação implícita de relevância
Abstract
This dissertation reports an approach to enhance information retrieval systems which are accessed through mobile devices. This sort of devices can impose constraints on user-computer interaction, mainly concerning expression of contextualized queries and navigation of long lists of results. Our approach integrates user work context in implicit relevance feedback, which is developed over a Case Based Reasoning methodology, with the purpose of providing personalized information retrieval. To tackle these issues, it was developed an architecture to manage transformation and processing of context information, as well as selection of evidences to expand queries through implicit relevance feedback. The project was evaluated on two case studies and the obtained results show that our approach enhances the quality of information retrieval, even under varitions on the size of the document collection, on the diversity of users and on context situations.