Research Of Affective Computing Used In Usability Testing

With the development of computer software technology development and increasing the level of the continuous improvement of the user, the user will require not only the software has a powerful performance, but also the availability of software and other non-functional characteristics have become increasingly demanding. In software usability testing, users of the emotional impact as a user experience and user satisfaction are important factors in more and more importance. Users use the computer to obtain information to complete the process of mission is also to influence computer users, resulting in emotional changes of a process, we pass on the user's emotional changes in research, to find its relationship with software usability and user satisfaction with direct relations.Current customer satisfaction there is a more difficult real-time access to customer satisfaction, accuracy is not high enough, can not carry out quantitative research and so on.In this paper, the use of PAD-Mood Scale to obtain a user emotional states and to make use of neural network users to set up changes and the availability of emotional relations, emotional changes through the user metrics to measure customer satisfaction, integration of the user's emotion models and algorithms, to pass the test of a verification of the model. In this paper, the results of research can be used for software usability testing, and software customer satisfaction measurement, and try to access the computer changes the way of human emotion, guidance and harmonious environment for the construction of human-computer interaction and related areas for future research provided a first-hand data.This article specifically completed the following work:First of all, theoretical research, including finishing reading related literature, the availability of testing theory and methods of the concrete practice of research, affective computing research and related applications.Secondly, based on theoretical studies, combined with the actual testing process in the user's behavior, emotional changes, user modeling emotion, set up an interactive model, HMM model of user behavior, user cognitive - emotional model, the user emotional dimension PAD model. Algorithm is derived to carry out, including: Based on the experience of the process of satisfaction measurement algorithm, based on affective computing the availability of measurement methods.Third, for experimental verification, for a total of 20 completed the 4 groups of people the availability of laboratory tests and to obtain relevant data for the algorithm correctness and reliability of a preliminary validation.Finally, based on the satisfaction derived algorithm prototype usability test system design, including the overall structure, and interface design.