Estimating learner’s comprehension with Cellular Neural Network for associative memory

In self-directed learning like e-learning, it is very important for learners to recognize their characteristics, and the system should appropriately support them. Accuracy rate and time required to answer are objective measures for judging learnerpsilas comprehension in test. The problem that learnerpsilas comprehension is classified based on objective information (score, time) can be treated as a classification of ambiguous data. On the other hand, cellular neural network (CNN) has been reported to be effective for associative memory. This paper, proposes an estimation system of learnerpsilas comprehension. A classification rate of CNN was about 95.6% from experimental results. Moreover, a comparison with MLP illustrated a high CNN performance.