Research on Personalized Recommendation Based on Improved Whale Optimization Algorithm

In order to solve the problem of insufficient information mining and low degree of personalization due to the single data dimension of cognitive diagnosis test question recommendation, this paper proposes a personalized exercise question recommendation method based on multi-dimensional data cognitive diagnosis optimization model. Firstly, the improved whale optimization algorithm with adaptive probability distribution is used to solve the problem that the initial clustering center of the K-means algorithm is sensitive and easy to fall into the local optimum, and secondly, the optimized k-means algorithm was used to cluster the exercises to generate a specific group of test questions. Then, the students' answers to specific test groups were recorded to form a multi-dimensional test question set. Finally, based on the multi-dimensional question bank, the method proposed in this paper and the traditional GDINA cognitive model algorithm are used to test the average accuracy of student sampling, and the results show that the algorithm proposed in this paper has stronger information mining ability and is more suitable for accurate recommendation of personalized test question resources matching learning needs and abilities.

I. Introduction

With the explosion of Internet information, college students need to face massive information and diverse choices, which cannot be quickly and accurately matched with needs, and are prone to the problem of "information overload". Information overload not only leads to the consumption of time and energy by college students when screening information, but also greatly affects their decision-making efficiency. In recent years, scholars have applied the recommendation system to the recommendation of exercises, which enables students to efficiently acquire and learn knowledge, and has achieved good results ^[1]. The recommender system flowchart is shown in Figure 1.