Constructing Usability Evaluation Indicators For Computer-aided Translation Software

Computer-aided translation technology (CAT) has been widely used in the language industry and attracted considerable attention from scholars, translators, as well as toolmakers in this field. Different from machine translation, CAT is based on human-machine interaction and can help translators leverage translation memory. With the use of CAT, translation efficiency can be improved and translators can also utilize such functions of mainstream CAT tools as collaborative translation and project management. However, with so many CAT tools available, it is difficult for translators to find the appropriate CAT software to help them, and sometimes the software is considered as less usable. In order to facilitate the comparisons of CAT tools, this study looks into usability evaluation indicators.The author has a review on researches of computer-aided translation, usability evaluation and user experience, and a questionnaire has been made and the research has been conducted with the combination of theories and experience. From the results of the research, the influence of satisfaction, errors, learnability and efficiency on the users' choice are verified. Besides, the author works out a model for the usability evaluation indicators of CAT software, under the guide of Nielsen's usability theory and Garrett's user experience theory, as well as the results of the research. There are three parts included in the evaluation system-"appearance - satisfaction", "design-efficiency" and "information - ease of use". The author also illustrates how to work out a detailed model and give weight values to different indicators.In this research, the author has tried a lot of methods to study the users' needs of CAT software, but there are still some deficiencies, for example, the research coverage is not large enough; the methods to evaluate the indicators are whether suitable. Future studies might make efforts in such directions. Besides, as the computer technology develops, CAT software will have more probabilities in its performance in future, so that there is always a large research gap in this field.