Beppler, F. D.; Fonseca, F. T.; Pacheco, R. C. S. Ontology-Driven Information Retrieval: A Hermeneutical Approach to Information Seeking Behavior.

Abstract: Information Retrieval (IR) research has made a great progress in the last few years. However, extracting valuable information from IR systems is still very time consuming. Common problems affecting users, besides uncertainty and anxiety, are deficient query definition, information overload, and poor system interactivity. In our approach we addressed this uncertainty and anxiety which are part of the search process. If the user does not know exactly what s/he wants, how can a framework help in the information seeking process? We resorted to philosophy and the hermeneutic circle of Heidegger and Gadamer to provide the principles of such a framework. In our implementation of the hermeneutic circle in an IR system users develop their ideas while browsing the information and the concepts that represent the information. We chose ontologies to implement this hermeneutic approach. Ontologies improve IR systems regarding its retrieval and presentation of information, which make the task of finding information more effective, efficient, and interactive. We called our framework Hermeneus, which in Greek means the interpreter or translator. Hermeneus works as an intermediary that facilitates the user to move from the initial state of information need to the goal state of resolution. Our system intends to be the bridge between the user's question and the answer to be found while s/he navigates in the ontology concepts and the instances of these concepts in a back and forth way.

Beppler, F. D.; Fonseca, F. T.; Pacheco, R. C. S. Hermeneus: An Architecture for an Ontology-Enabled Information Retrieval.

Abstract: Ontologies improve IR systems regarding its retrieval and presentation of information, which make the task of finding information more effective, efficient, and interactive. In this paper we argue that ontologies also greatly improve the engineering of such systems. We created a framework that uses ontology to drive the process of engineering an IR system. We developed a prototype that shows how a domain specialist without knowledge in the IR field can build an IR system with interactive components. The resulting system provides support for users not only to find their information needs but also to extend their state of knowledge. This way, our approach to ontology-enabled information retrieval addresses both the engineering aspect described here and also the usability aspect described elsewhere.