Chatbot em contexto: design de experiência do usuário aplicado à recuperação da informação no catálogo de teses e dissertações da CAPES

Abstract

This research aims to present the construction of a chatbot conversational interface for information recovery in CAPES's theses and dissertation catalog database. It is based on methodologies of user studies anchored in the theoretical references of informational practices, which articulate the socially constructed elements and that influence the formation of subjectivity of individuals, while changing reality. The methodological contribution of the research has directed to the challenge of building a technological product (artifact) user context - a chatbot called Kika - with the use of Design Science Research (DSR). The built artifact recovered data from the theses and dissertations catalog, made available as data opened by CAPES for the period 2013 to 2020, and was generated iteratively over three virtual meetings, with the participation of volunteers divided into three target groups: undergraduate, postgraduate and researchers; Composed of both people who did not know the Capes portal and those who had knowledge of it and used it. It was sought to measure aspects of the user experience regarding reliability, ease of interaction in pseudodial with chat for access and recovery of information. Finally, it is concluded that the construction of a conversational artifact, based on the perspectives of informational practices, contributes to the interaction of the informational subject with the built artifact, as it considers its evolution during the design process. To this end, it is argued that the theoretical constructs of users articulated users' studies of the use of artifact, centered on the interaction of their users, are congruent and contribute as another strategy for data recovery in an agile, functional and applicable way.