An Evaluation Method of Knowledge Graph-Based Recommender System Based on FCE and GA-BP Neural Network

• Main Authors: Xing, Jian-Hao, Liu, Jian-Wei, Dai, Zhi-Hao, Yang, Pu, Mao, Ze-Hui, Ma, Ya-Jie
• Format: Conference Proceeding
• Language: English
• Subjects: Automation; Big Data; Design methodology; fuzzy comprehensive evaluation; genetic algorithm; Knowledge engineering; knowledge graph-based recommender system; Manufacturing; neural network; Neural networks; system assessment; Technological innovation
• ISSN: 2688-0938
• DOI: 10.1109/CAC53003.2021.9728252

In the background of the big data era, the knowledge graph-based recommender system has been widely used in the field of intelligent manufacturing. The purpose of this paper is to fill the gaps in the evaluation of the knowledge graph-based recommender system and evaluate the engineering effect of the system. This paper establishes an evaluation indicator system of the knowledge graph-based recommender system and designs an evaluation method based on fuzzy comprehensive evaluation and GA-BP neural network. The evaluation method uses the basic indicator data of the knowledge graph-based recommender system and it's fuzzy evaluation result vector to train the GA-BP neural network. This method can obtain fuzzy evaluation results directly according to the basic indicator data. Finally, the experiment results show that the fuzzy evaluation results predicted by the GA-BP neural network have small errors. The evaluation method can objectively reflect the engineering effect of the knowledge graph-based recommender system and replace the traditional single fuzzy comprehensive evaluation effectively.