Research on Course Evaluation Based Genetic Algorithm with Chi-Square Test Using Student Feedback

In recent days, Course evaluation is a systematic process utilized in educational institutions for assessing course quality, student learning experiences to improve teaching methodologies. Traditional approach like Standard Genetic Algorithm (SGA) technique for course evaluation had faced challenges including low response rates and selection problem of courses. Therefore, this research proposes Genetic Algorithm-Chi-Square test (GA-Chi-Square test) to resolve the issue of course selections through optimization. Initially, the data is collected from Student Feedback (SF) dataset which consists of student feedback on various course related aspects and then preprocessed by punctuation and stop-words removal techniques. After that, Term Frequency-Inverse Document Frequency (TF-IDF) method is used to extract course scores based on term frequency and fed into GA. This technique integrated Two-Direction Crossover (TDC) and Grouped Mutation (GM) to optimize course selections for each student. Subsequently, Chi-Square test is used for course evaluation by testing optimized course selections. The proposed GA-Chi- Square model attained better results in terms of accuracy (96.2%), precision (95.3%), recall (96.3%), and F1-score (95.8^%) when compared with existing Multi-task Learning (MTL) - Roberta.