Knowledge Graph Based Recommendation by Adversarial Learning Algorithm in Application of Lifelong Education Data Classification

Abstract

Students can improve their capacity to learn continuously and work together to achieve a common goal through cooperative and explorative coursework in personalized learning. This article presents an effective method for clustering people by preference and a strategy for developing course suggestions for different organizations. This lets us consider student characteristics and courses in a statistically and semantically clear way. First, this article uses specific word articles and Word2Vec to extract factors efficiently. Optimizing efficiency. After that, a slightly modified K-means algorithm and perceptron adversarial learning method classify students into interest-based study clusters. The knowledge graph is created and saved to achieve this. In conclusion, the opinion-based deep learning algorithm used for subject recommendation system design provides advice for appropriate and high-quality results based on the degree of similarity between recommendation results and expert scoring. To do this, the proposed method is approximated against existing machine learning methods and compared to their prediction performance metrics.