Test Paper Generation System using Multi Agents

Objective: To build a test-paper generation system for the welfare of students to improve their knowledge thereby developing their skills. Methods: In this paper, Multi Agent System is employed to generate the test paper for students. Different intelligent agents such as Question Selector Agent, Performance Analysis Agent and Co-ordinating Agent are designed, which interact with each other, to find the performance of the student in the test. The proposed algorithms assign utility values to the questions stored in the database and based upon the subject preferences given by the students, test paper is generated. Findings: A new intelligent system is proposed which helps the students to understand their knowledge level in the preferred subjects. The developed test paper generation system overcomes the drawback of normal class room test by reducing the problem in crowd management and also by making the test paper evaluation process easier and faster. The student can take the test from any place and the system is designed in such a way that each time when the user takes the test, he would get different set of questions. The difficulty level of the test paper is automatically computed for every test the student takes, based upon his previous test performance. Applications: The test paper generation system is a helping hand for students' community to improve their study skills in their subject areas. The proposed system using Multi Agents improve the efficiency of the learning process of students.