Uma proposta para a evolução de ontologias a partir de personomias em sistemas baseados em tagging

Abstract

Creating and publishing content in World Wide Web has become much easier for the users. As a result, there has never been so much information available in that mean before. It generates an information overload on the users since the process of identifying the desired information requires a great cognitive effort from them in order to discern which is relevant and which is not. Therefore, means to organize data more adequately become necessary to facilitate the identification of the desired information. One way to carry this out is the use of taxonomies or ontologies. The definition of this kind of structure is interesting for the information retrieval process, but it is costly and incompatible with the huge amount of information available on the web, since this is a complex task for most users. For that reason, the tagging technique is becoming popular once its main advantage is the ease with which users categorize the information. However, with a significant increase in the amount of categorizations another information overload is generated on users at the moment of the information retrieval. For this reason, in this dissertation we propose the use of the benefits of tagging process for the information categorization and the benefits of ontologies as a structure for helping in the information retrieval process. Our proposal uses an unsupervised approach for the ontologies emergence from a person's tagging data - its personomy. This technique aims to shape the user's mental model relating to his/her tagspace to solve problems inherent to the tags use in the information retrieval process, considering some cognitive psychology studies.