抄録
This paper presents a case study on the application of Natural Language Processing（ NLP） and Artificial Intelligence（ AI） to improve the benchmarking and decision-making processes in Thailand’s External Quality Assurance（ EQA） system. This study attempted to address this issue by creating an NLP-based data extraction pipeline adapted to EQA benchmarking requirements through the mix-methods research. Stakeholder surveys found that approximately 87.5% of respondents identified the need for automated NLP techniques for transforming unstructured data into actionable insights, implying that the response emphasizes the practical relevance of creating technology to expedite and improve the benchmarking process in education. Leveraging these findings, the NLP pipeline was built using regular expression, pattern matching, and Named Entity Recognition（ NER） to capture the desired text from complicated documents. Thereafter utilizing TF-IDF to vectorize and analyze meaningful insights with high accuracy, reaching a 98.33% match with annotated datasets and an F1 score of 1.0, the system effectively extract data while also obtaining critical data to support advanced analytics and visualizations revealed hidden performance patterns for both regulatory and collaborative benchmarks.
