Evaluation of Gazes of Experts and Beginners in Vortex Search of Fluid-Simulation

抄録

In the manufacturing industry, knowledge transfer of experts has become an important issue because of the decreasing number of experts. To support knowledge transfer, various support systems have been developed. To present appropriate knowledge to the user, the user’s skill level needs to be assessed. The purpose of this study is to develop a technique for judging a user’s skill level using gaze-measurement data. The gazes of experts and beginners were measured in relation to their vortex-search skills when searching for vortices of fluid-simulation results. We generated 111 features from the fixation time of each vortex, the transition frequency between vortices, and the average pupil size, calculated the importance of the feature values using the Decision Tree, and extracted the top five feature values. Several machine learning models were examined to estimate the user's skill level by using these five feature values. As a result, by using the K-nearest Neighbor, the model was able to discriminate between experts and beginners with an accuracy of 84.2%. These results imply the gaze shows the difference in the users' skill levels and show the prospect of presenting appropriate knowledge to the user of the design support system by using the gaze-measurement result.