Proposta de um Framework para Visualização de Dados Agregados por Similaridade para Auxiliar Consultas durante a Navegação na Web

Abstract

In the last decade, several specialized tools have been created upon similarity functions that, given a keyword and a context, determine the degree of similarity (or probability) that information in a dataset corresponds to the user's query. Quite often such tools are meant for experts and require training and knowledge on the application domain to be used. However, given the huge amount of information available on the Web, resolving ambiguities becomes a daily task for most users. In this paper, we present a technique for embedding into a Web browser tools for solving ambiguities between keywords that users might found while navigating the Web. A prototype illustrating such techniques has been developed as a proof of concept. The tool presents the degree of similarity directly on Web pages as a contextual help menu. The overall approach includes different datasets and similarity functions and is flexible enough to support extensions for covering additional contexts of use.