The purpose of this article is to design an intelligent tutoring system. The learning model of the proposed system includes features of academic self-efficacy and learning style. Academic self-efficacy has been automatically identified by designing of a fuzzy system based on learners' behavior and learning style through the questionnaire of Felder-Silverman which contains 44 question. After identification of these features, Proportional education strategies are presented and implemented in tutoring system in a real environment. The effectiveness of the proposed tutoring system is evaluated in terms of learners' operation by investigation of their satisfaction from system. The results show that considering functional characteristics in learning model, presenting some learning objects and proportional recommendations to the characteristics, results in 75% learners' educational progress and their educational satisfaction. Moreover, evaluation of the time passed in the e-Learning environment before and after using Perles doesn’ t show a significant difference. Results show that the designed intelligent tutoring system based on the learner model and educational strategies, has led not only to the educational success of the learners but also to increase in their enthusiasm in using the system. Considering other effective and cognitive features in learning is highly recommended inorder to provide a personalized environment.