Domain Model Discovery from Textbooks for Computer Programming Intelligent Tutors

We present a novel approach to intro-to-programming domain model discovery from textbooks using an over-generation and ranking strategy. first extract candidate key phrases each chapter in Computer Science textbook focusing on then rank those concepts according number of metrics such as the standard tf-idf weight used information retrieval produced by other text algorithms. Specifically, we conduct our work context developing intelligent tutoring system for source code comprehension which specification programming is needed - monitors students' performance scaffolds their learning process until they show mastery concepts. Our experiments with concept instruction Java indicate that statistical methods KP Miner method are quite competitive compared more sophisticated methods. Automated models will lead scalable Intelligent Tutoring Systems (ITSs) across topics domains, major challenge needs be addressed if ITSs widely millions learners many domains.