Design of Personalized Recommendation System for College Education Based on Multivariate Hybrid Criteria Fuzzy Algorithm

Abstract

The learner model's design is a most important aspect of a personalized college student education recommendation system. Currently, most learner models need more scientific focus, relying on a single method to collect dimensions and feature attributes with low computing costs. Hence, this paper introduced a Multivariate Hybrid Criteria Fuzzy Algorithm (MHCFA) for personalized college education recommendation. The MHCFA is trained by Social Feedback Artificial Tree (SFAT), where SFAT is the combination of Social Optimization Algorithm (SOA) and Feedback Artificial Tree (FAT). In addition, Deep Fuzzy Clustering (DFC) is utilized to group college education content. The RV-Coefficient is employed to select the best content. Moreover, the feature is extracted by All Caps and numerical for further personalized recommendations. In addition, the test results show that SFAT_MHCFA performs better in Precision, Recall, and F-Measure where the values gained 0.989, 0.878, and 0.859, respectively.