Evaluation of influencing factors of China university teaching quality based on fuzzy logic and deep learning technology

• Published in: PloS one Vol. 19; no. 9; p. e0303613
• Main Author: Yu, Jie
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 06.09.2024; Public Library of Science (PLoS)
• Subjects: Accuracy; Algorithms; Artificial intelligence; China; Classrooms; College faculty; Colleges & universities; Curricula; Datasets; Decision making; Deep Learning; Educational objectives; Educational research; Evaluation; Fuzzy algorithms; Fuzzy Logic; Fuzzy sets; Fuzzy systems; Higher education; Humans; Information theory; Long short-term memory; Machine learning; Methods; Neural networks; Prediction models; Quality assessment; Quality management; Social aspects; Student participation; Students; Teachers; Teaching; Teaching methods; Technology assessment; Universities; China
• DOI: 10.1371/journal.pone.0303613
• ISSN: 1932-6203

Nowadays, colleges and universities focus on the assessment model for considering educational offers, suitable environments, and circumstances for students’ growth, as well as the influence of Teaching Quality (TQ) and the applicability of the skills promoted by teaching to life. Teaching excellence is an important evaluation metric at the university level, but it is challenging to determine it accurately due to its wide range of influencing factors. Fuzzy and Deep Learning (DL) approaches must be could to build an assessment model that can precisely measure the teaching qualities to enhance accuracy. Combining fuzzy logic and DL can provide a powerful approach for assessing the influencing factors of college and university teaching effects by implementing the Sequential Intuitionistic Fuzzy (SIF) assisted Long Short-Term Memory (LSTM) model proposed. Sequential Intuitionistic Fuzzy (SIF) can be used sets to assess factors that affect teaching quality to enhance teaching methods and raise the standard of education. LSTM model to create a predictive model that can pinpoint the primary factors that influence teaching quality and forecast outcomes in the future using those influencing factors for academic growth. The enhancement of the SIF-LSTM model for assessing the influencing factors of teaching quality is proved by the accuracy of 98.4%, Mean Square Error (MSE) of 0.028%, Tucker Lewis Index (TLI) measure for all influencing factors and entropy measure of non-membership and membership degree correlation of factors related to quality in teaching by various dimensional measures. The effectiveness of the proposed model is validated by implementing data sources with a set of 60+ teachers’ and students’ open-ended questionnaire surveys from a university.