Intelligent Test Paper Generation with Genetic Algorithm

Abstract

In this study, the solution of the problem of generating an intelligent test paper with a genetic algorithm is presented depending on the required criteria in a question bank. Generating the intelligent test paper is considered as a multi-parameter optimization problem, depending on whether each question in the question bank has many attributes. A genetic algorithm is a heuristic search algorithm with parallel search feature which is often used to solve optimization problems. In the study, the changes in the crossover and mutation operators of the standard genetic algorithm increased the performance of the genetic algorithm and created the test papers in the required quality. Experimental results show that the improved genetic algorithm is more effective when compared to the standard genetic algorithm in the same conditions. In the study, a web-based user interface application was developed in which users can set the criteria for genetic algorithm and test paper and can run the algorithm

Keywords

• Genetic algorithm
• Question bank
• Test page generation
• User interface
• Web-based application