International Journal of Modern Education Research 2016; 3(1): 1-5 http://www.aascit.org/journal/ijmer ISSN: 2375-3781 Keywords Intelligent Tutoring System, Autonomous Model, Q-Learning, Convergence Improvement, Tabu Search Received: November 8, 2015 Revised: November 30, 2015 Accepted: May 13, 2016 Improvement of Q-Learning Algorithm Convergence with Intelligent Tutoring Systems and Tabu Search Éverton de Oliveira Paiva 1 , Marcus Vinicius Carvalho Guelpeli 2 1 Management of Educational Institutions - GIED, Federal University of Jequitinhonha and Mucuri Valleys – UFVJM, Diamantina, Brazil 2 Department of Computing – DECOM, Federal University of Jequitinhonha and Mucuri Valleys – UFVJM, Diamantina, Brazil Email address evertonpaiva@gmail.com (E. O. Paiva), marcus.guelpeli@ufvjm.edu.br (M. V. C. Guelpeli) Citation Éverton de Oliveira Paiva, Marcus Vinicius Carvalho Guelpeli. Improvement of Q-Learning Algorithm Convergence with Intelligent Tutoring Systems and Tabu Search. International Journal of Modern Education Research. Vol. 3, No. 1, 2016, pp. 1-5. Abstract Using computer systems as a complement or replacement for the classroom experience is an increasingly common practice in education, and Intelligent Tutoring Systems (ITS) are one of these alternatives. Therefore, it is crucial to develop ITS that are capable of both teaching and learning relevant information about the student through artificial intelligence techniques. This learning process occurs by means of direct, and generally slow, interaction between the ITS and the student. This article presents the insertion of meta-heuristic Tabu search with the purpose of accelerating learning. Computer simulations were conducted in order to compare the performance of traditional randomized search methods with the meta-heuristic Tabu search. Results obtained from these simulations strongly indicate that the introduction of meta-heuristics in exploration policy improves the performance of the learning algorithm in STI. 1. Introduction An Intelligent Tutoring System (FREEDMAN, 2000) [1] (ITS) is a broad term that includes any program displaying intelligence which can be employed as a learning tool. The ITS evolved from Computer-Assisted Instruction (CAI) and differs from its predecessor by the addition of knowledge learning strategies and because it keeps an updated model of the learner’s activities. Traditional E-learning systems (LI; ZHOU, 2015) [2] were criticized for its limitations, since they always presented the same material and topics for students, regardless of their previous knowledge, level of comprehension of the subject, or learning ability. Conversely, ITS uses a database that contains knowledge expertise on the subject, learning strategies and heuristics and should be able to dynamically select relevant teaching material and thus choose different pedagogical pathways, examples and exercises for different students. Intelligent Tutoring Systems offer flexibility in the way the material is presented and greater ability to attend to students’ needs. In addition to teaching, they seek to learn relevant information about the student, thus creating an individualized learning process. ITS have been presented as highly efficient for improving student performance and motivation (PALOMINO, 2013) [3]. In order for students to acquire this ability, Artificial Intelligence (AI) techniques, such as