Taxonomy of GUM and Usability Prediction Using GUM Multistage Fuzzy Expert System

The evaluation of quality of software is an important aspect for controlling, managing so that we can be able to enhance the improvement in a software process. For such evaluation, many factors have been identified by a number of researchers. The quality of software is further dependent on many other factors. Usability of software is one of the most significant aspect on which quality of software is dependent. Many researchers proposed a number of software usability models, each model considering a set of usability factors but these models do not include all the usability aspects and it is hard to integrate these models into current software engineering practices. As far as real world is concerned, we are facing many obstacles in implementing any of these proposed models as there is a lack in its precise definition and the concept of globally accepted usability. This paper aims to define the term ‘usability' using the Generalized Usability Model (GUM). GUM is proposed with detailed taxonomy for specifying and identifying the quality components, which brings together factors, attributes and characteristics defined in various Human Computer Interaction (HCI) and Software Models. This paper also shows how to predict the usability of a software application using a fuzzy based expert system which has been implemented using multistage fuzzy logic toolbox.