基于粗糙集约简与概率图的认知诊断模型研究

Research on Cognitive Diagnosis Model Based on Rough Set Reduction and Probability Graph

随着孤独症儿童的数量不断增加,准确且及时地对其进行认知诊断变得愈发重要.构建基于粗糙集约简算法的孤独症诊断知识库,研究基于概率图的认知诊断模型,以提高对孤独症儿童诊断的准确性和效率.实验结果表明:该认知诊断模型的均方根误差值范围为0.10~0.11,平均绝对误差值范围为0.009~0.115,在孤独症儿童的认知诊断中具有较高的准确性和稳定性.

As the number of children with autism continues to increase,accurate and timely cognitive diagnosis has become increasingly important.The diagnosis knowledge base of autism was constructed based on rough set reduction algorithm,and the cognitive diagnosis model was developed based on probability graph to improve the accuracy and efficiency of the diagnosis of autistic children.The experimental results show that the root mean square error value of the proposed model ranged from 0.10 to 0.11,and the average absolute error value ranged from 0.009 to 0.115.The cognitive diagnosis model has high accuracy and stability in the cognitive diagnosis of autistic children.