Business platform models often require continual adaptation and agility to create and deliver new experiences to customers. To understand user behavior in online systems, research has taken advantage of the combination of traditional analysis techniques with the latest ones. Earlier studies have shown that user behavior monitoring data, as obtained by mouse tracking, can be used to improve the user experience. While there are many mouse tracking solutions, the vast majority are proprietary and open source packages do not provide the resources and data needed to support this work. Given this, this paper presents: (i) the development of an interaction monitoring application titled Web Tracer; (ii) validation of the tool in a case study done on the website of the Brazilian federal revenue service; (iii) the definition of a new relationship pattern of variables that determine user behavior; (iv) the construction of a fuzzy inference system using the defined variables and the data captured in the case study. The results indicate, through a comparison of the applied quantitative methodologies, that the developed architecture was able to infer UX scores similar to those reported by users in questionnaires.
|
