An Automatic Questionnaire Generation Method Based on Emotional Graph-Driven

Traditional questionnaire method has many shortcomings in capturing user’s emotions and preferences, especially for convergence media, though it can collect user feedback to a certain extent. In order to conquer it, this paper proposes an automatic questionnaire generation method based on emotional graph-driven, which comprehensively uses sentiment analysis, user’s behavior analysis and dynamic questionnaire generation technology to generate personalized questionnaire content, in line with the user’s immediate emotional state by capturing the user’s emotional fluctuations and combining with his behavior data. It can also dynamically adjust the content of the questionnaire according to user’s emotional state and behavior pattern to improve the user’s participation and the accuracy of the data. The method is applied to a recommendation system, and the performance is significantly improved. The experimental results show that the recommended click-through rate has been increased by 7.3%, and the average dwell time of users has been increased by 12.8%.