Research on online test question recommendation technology based on multi-dimensional user dynamic portrait

Abstract This paper proposes a research method of personalized recommendation test questions for users by fully collecting and studying the user’s characteristic data in multiple dimensions, establishing dynamic user portrait, mining analyzing information question bank. Firstly, collect data, calculate similarity behavior sequence, use cosine algorithm to text questions. Then, tag system is established analyze activity, TF-IDF used proficiency different keyword classification Based on above information, hybrid mechanism adopted provide with services.