Modeling the Examinee Ability on the Computerized Adaptive Testing Using Adaptive Network-Based Fuzzy Inference System

The computerized adaptive testing attempts to acquire an online evaluation that best indicates an examineepsilas ability based on his/her latent traits and to provide the most suitable follow up item or question dynamically for the examinee to take. Although maximum likelihood estimation (MLE) and Bayesian likelihood estimation (BLE) have been proposed to solve the problem of estimating the examinee ability, little attention has been paid to the issue that an item response may not be consistent with the examineepsilas ability nor the shifts of the ability estimation. We hypothesized that the adaptive-network-based fuzzy inference system (ANFIS) can be used to infer flexible examineepsilas ability estimate automatically by analyzing the relevant data of the examinee in a test. Consequently, the study presents a novel learning ability model using ANFIS, which can adaptively choose questions based on the item response theory. By taking into account the item parameters of discrimination, difficulty, guessing, as well as the estimated examinee ability before he/she answers a question, the proposed method could infer the adjustment of the examinee ability and improve the ability estimation. Simulations study was conducted to evaluate the performance of our approach. The experimental results demonstrate that our method is more effective in estimating the examinee ability than MLE and BLE when the value of the test information is within the range of some precision.