Biography
Biography pending.
Conference contributions 1
MACHINE LEARNING–BASED MODELLING OF ARTIFICIAL INTELLIGENCE-SUPPORTED COGNITIVE IMPROVEMENT OF CONSTRUCTION MANAGEMENT STUDENTS
This study investigated the extent to which an AI teaching assistant enhances planning, problem solving, decision making and logical reasoning among 72 construction management students using a pre-test and post-test quasi experimental design. Students completed eight cognitively demanding and discipline specific assessment tasks covering construction planning, managerial problem solving, decision analysis and critical reasoning. Parallel test forms were administered before and after structured exposure to an AI teaching assistant integrated into learning and practice activities. Descriptive analysis, paired samples t tests, Wilcoxon signed rank tests, repeated measures ANOVA and repeated measures MANOVA were used to evaluate baseline trends, statistical attribution of improvement and overall multivariate cognitive change. Practical educational impact was quantified using Cohen's d and partial eta squared. Gain score analysis and machine learning models were further applied to predict learning improvement, identify key cognitive predictors, detect learner profiles and classify students by improvement level. Results show statistically significant and practically large gains across all four competencies, with balanced improvement patterns and a very strong overall multivariate effect. Decision making and logical reasoning demonstrated the largest improvements and were the strongest predictors of total cognitive gain. Distinct learner profiles were identified, indicating heterogeneous learning responses. The findings provide robust empirical evidence that AI supported learning can strengthen integrated professional cognition in construction management education and inform adaptive instructional design.