Implementation and evaluation of personalized intelligent tutoring system

Abstract

This research article illustrates a generic architecture for intelligent tutoring system christened as SeisTutor. SeisTutor adapts itself according to the learner learning preferences by determining the learning style and pre knowledge level. The aim of SeisTutor is to mimic similar the human intelligence by implicitly adjudge the tutoring strategy prior to tutoring session and custom-tailored the tutoring concepts to enhance the learning gain. SeisTutor was implemented using I2A2 index of learning style model. An Empirical analysis has been performed for graduation pursing students. The experimental analysis reveals that learning style model were accurately predicted with an accuracy of 61-100%. The applicants found SeisTutor is helpful with an average of 13% learning gain, attains 24% engagement at the beginning of the tutoring session.