TY - CONF T1 - An Automated Decision System for Computer Adaptive Testing Using Genetic Algorithms T2 - Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing, 2008. SNPD'08. Ninth ACIS International Conference on Y1 - 2008 A1 - Phankokkruad, M. A1 - Woraratpanya, K. AB - This paper proposes an approach to solve the triangle decision tree problem for computer adaptive testing (CAT) using genetic algorithms (GAs). In this approach, item response theory (IRT) parameters composed of discrimination, difficulty, and guess are firstly obtained and stored in an item bank. Then a fitness function, which is based on IRT parameters, of GAs for obtaining an optimal solution is set up. Finally, the GAs is applied to the parameters of the item bank so that an optimal decision tree is generated. Based on a six-level triangle-decision tree for examination items, the experimental results show that the optimal decision tree can be generated correctly when compared with the standard patterns. JF - Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing, 2008. SNPD'08. Ninth ACIS International Conference on PB - IEEE ER -