A novel neuro-fuzzy approach for evaluating educational programme quality and institutional performance in higher education
Year: 2025
Type: conference-paper
Abstract: This paper presents a novel methodology for evaluating the quality of educational programmes and institutional performance in higher education institutions using advanced artificial intelligence techniques, specifically the Adaptive Neuro-Fuzzy Inference System (ANFIS) and multi-layer neural networks. The primary objectives of the study were to address the challenges of subjectivity in self-assessment processes and proactively identify potential issues and deficiencies in educational activities prior to accreditation reviews. The proposed approach utilised student ratings on a four-level assessment scale as input data for the multi-layer neural network, while the criteria for assessing educational programme quality served as input variables for the ANFIS model. The underlying hypothesis was that students with higher academic performance would provide more objective assessments of the quality criteria. The results demonstrated that the multi-layer neural network exhibited superior predictive accuracy compared to the ANFIS model. This paper suggests that the proposed methodology can equip higher education leaders with high-quality forecasts to ascertain the calibre of educational services and pinpoint potential problems in advance of accreditation examinations. However, the authors acknowledge the necessity for further research with larger datasets to enhance the predictive capabilities of the models.
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: bronze
OpenAlex ID: W7166019689