In the realm of innovation and entrepreneurship education for university students, the concept of industry-education integration has gained increasing prominence. This paper designs and implements a knowledge graph-based fragmented knowledge management system and a personalized question recommendation system based on student profiling. Through knowledge tag extraction algorithms and fuzzy cognitive diagnosis models, it enhances learners’ knowledge management capabilities and the accuracy of learning diagnostics. From an industry-education integration perspective, three major reform pathways are proposed. Empirical research validates the positive effects of the new teaching model. Following the teaching practice, scores across all educational factors remained at relatively high levels. Student satisfaction with their own innovative thinking and entrepreneurial awareness reached the highest average score of 3.343 ± 0.456. Academic output factors scored at a moderately high level, with overall student satisfaction toward the institution’s innovation and entrepreneurship education (SC4) significantly higher than satisfaction with student entrepreneurship rates (SC1) (P < 0.05). The deep integration of industry-education collaboration with intelligent technologies holds significant practical value for effectively linking educational chains, talent chains, industrial chains, and innovation chains.
