With the progress of science and technology, Internet technology has also been greatly developed, people can communicate through the network, so there are a lot of interpersonal interaction in the network. NSGA-II (Non-dominated Sorting Genetic Algorithm II) is an algorithm for multi-objective optimization. Interpersonal interaction psychological analysis system is a system used to analyze interpersonal interaction behavior. In the interpersonal interaction psychological analysis system, NSGA-II algorithm is used to carry out multi-objective optimization of interpersonal interaction behavior, taking various factors in interpersonal interaction behavior as objective functions (emotional state, communication effect, information transmission, etc.), and then NSGA-II algorithm is used to optimize and analyze these objective functions to find the optimal solution. This can help people better understand interpersonal interaction behavior, improve the effectiveness and efficiency of interpersonal communication. After in-depth data analysis, we find that NSGA-II algorithm shows remarkable accuracy in the analysis of interpersonal interaction psychology. The algorithm has high performance in processing complex psychological data, which greatly improves the efficiency of analysis. Through comparative experiments, we further confirm the advantages and potential of NSGA-II algorithm. The experimental results show that the recall rate and accuracy of feedback and update modules are higher than 90%, respectively 90.3% and 97.2%, and the running time ranks third (4.4s).