Software Usability Evaluation Using Opinion Mining

Usability is critical for any system, but in software it is one of the most important features. In fact, one of the main reasons for software failure is the system lacking to achieve users specified goals and satisfaction. For this reason, usability evaluation is becoming an important part of software development. Software usability evaluation can be costly in terms of time and human. Therefore, automation is promising way to augment existing approaches especially if the evaluation is subjective where the usability concentrated about user's "opinion". This paper proposes to use opinion mining as an automatic technique to evaluate subjective usability. Opinion mining is a research subtopic of data mining aiming to automatically obtain useful opinioned knowledge in subjective texts. We propose a novel model to extract knowledge from opinions to improve subjective software usability. This is the first time opinion mining used in software usability. To evaluate our proposed model, a set of experiments was designed and conducted and we got an average accuracy of 85.41%. Also, we propose to use graphics to visualize user's opinion in software and to compare the usability of two software.