Fuzzy Logic-Based System for the Estimation of the Usability Level in User Tests

Abstract

Starting from the challenge of obtaining the usability level in numerical and linguistic terms within a user test based on the three attributes that define usability, the development of a system based on fuzzy logic is proposed for estimating the level of output usability in a test with users based on the ISO 9241-11 standard. The attributes are effectiveness, efficiency, and satisfaction, which have different metrics and numerical scales. For the development of the system, five methodological stages were defined: characterization of the structure of a usability test, definition of membership functions for the inputs and outputs of the system, design of the inference rules that relate the inputs and outputs, design and implementation of the fuzzy system, and development of the case study. The proposed system was implemented using the FCL (Fuzzy Control Language) and the jFuzzyLogic API that takes as inputs the values calculated for the attributes of effectiveness, efficiency, and satisfaction, and obtains the usability level as output considering the membership functions of the inputs and outputs, as well as a set of inference rules defined by a set of experts. As a case study, the proposed fuzzy system was validated from the results obtained in a usability test with 5 users which was developed on the Sigma Electrónica website. From the results obtained in the case study, it could be concluded that the implemented system is adequate in terms of obtaining a level of usability in numerical and linguistic terms in conventional usability tests developed in a usability laboratory considering the attributes of ISO 9241-11.