Research on the Recommendation Method of Postgraduate Supervisor Based on Natural Intelligence Information Integration

In response to the difficulty of choosing a supervisor for prospective postgraduate, this research proposes a method for recommending postgraduate supervisors. The method starts with prospective postgraduate describing their good wishes for selecting a supervisor, initially drawing up the candidate supervisors of the college they apply to, and then using the crowd familiarity level with the candidate supervisors as natural intelligence perceptron, let each natural intelligence perceptron independently rank the candidate supervisors’ recommendations, and through the information integration algorithm, obtain the consensus ranking of the candidate supervisors, and prospective postgraduate can select supervisors based on the consensus ranking. To encourage positiveness and professionalism among assessors, the method draws on positive-sum game mechanisms. To encourage evaluators’ positiveness and professionalism, this method draws on the positive sum game mechanism to design the recommendation contribution level algorithm, which quantifies the assessor’s contribution to the recommendation assessment. And pay the assessor according to the level of contribution, allowing the assessor to be paid negatively. The research also designed an effectiveness feedback algorithm, which assesses the actual effectiveness of prospective postgraduate in selecting a supervisor by the recommendation results, and secondary distribution of the reward to the assessors based on the effectiveness feedback. This research demonstrates the calculation process of the postgraduate supervisor recommendation method using instance.