Importance of Data Sharing for the Validation of Interaction-based Emotion Recognition for User Experience Evaluation

In the era of big data and AI, we have the resources to provide website users with easy-to-use, efficient, error-free, and satisfying experiences through automated usability tests. Automated usability tests allow companies to evaluate user interactions at scale, significantly reducing the time and resources needed for traditional testing methods and making them more accessible to a broader range of companies. One way to enable automated usability tests for websites is by collecting user interactions for emotion recognition. However, current research on AI to automate user experience evaluation by detecting emotions based on keyboard, mouse, and touch interaction is not reproducible as most researchers do not publish the data they used for the experiments along with the results. Therefore, we focus on validating interaction-based emotion recognition approaches for usability testing. We will elaborate a verifiable method for automatically testing a website's usability and user experience, considering only the user interactions created while using the website. Furthermore, we want to provide a tool suite for an automated approach to testing a website's usability and user experience, considering only the user interactions created while using the website. We want to offer a continuous feedback loop by integrating automated usability tests into the development cycle. This fosters a culture of continuous improvement and enables teams to iterate quickly based on real-time user feedback. Our approach can be applied by any organization that utilizes a website. Overall, automated usability testing represents a critical advancement in the field, aligning with the industry's need for efficiency, effectiveness, and enhanced user satisfaction.