Avaliação da usabilidade de sites de busca que utilizam técnicas de extração de dados semi-estruturados

Abstract

The Web is turning in an enormous repository of data of the most varied domains. Most of the time those data don't possess defined structure, being called semi-structured data. Due to that heterogeneity of the data, it is more and more difficult to recover and to manipulate information of the Web. With the objective of reducing those difficulties, several tools and techniques of extraction of semi-structured data have been studied and developed, besides models of data and consultation languages. Another great difficulty on the part of the users is that the mechanisms of recovery of existent data (search machines) they are not built taking in consideration the usability of the systems, and not always they come back what the users seek, generating countless dissatisfactions. Then the importance of building tools that, besides efficient and fast, be easy to use and to understand. The objective of this work was to evaluate the usability of a search site that uses techniques of extraction of data semi-structured in the process of recovery of the information, making a comparative one with the current model of the existent search machines.