A Novel AI Approach for the Systematic Creation of Empathy Maps

Abstract

The utilization of Empathy Maps in User Experience and Design Thinking is becoming commonplace. They can be seen as a valuable tool to help developers better understand the user’s requirements. During the interaction design, it is essential to have a thorough knowledge of what the user wants, which might be difficult to gather or portray to others. In this sense, Empathy Maps visually represent the user’s information, attitude, and behavior. Nevertheless, creating Empathy Maps can become tedious, as this usually requires the involvement of a group of people with expert knowledge, deeply analyzing the information elicited from the user. This is the reason why most Empathy Maps are created manually. A first literature analysis reports that there are still no specific tools to generate Empathy Maps automatically, although there are technologies to take on this challenge. In this paper, we present a tool for automatically generating an initial version of an Empathy Map that can help speed up the capture of explicit knowledge, and then it can be further worked on by the development team. The tool creates the Empathy Map from an audio interview, featuring audio transcription technology and natural language processing. More specifically, this paper includes the architecture, a proof of concept, and user testing, reporting numerical evidence and promising results to corroborate the proposal’s feasibility, reaching an overall accuracy of 86% at classification.