基于IRT模型的BP神经网络降维法参数估计及其应用  

The Reduced Dimension Method of Parameter-Estimation Using BP Neural Network and Its Application Based on Item Response Theory

作　　者：谭云兰[1] 丁树良[2] 辛锐铭[2] 

机构地区：[1]井冈山学院计算机系,江西吉安343009 [2]江西师范大学计算机信息工程学院,江西南昌330027

出　　处：《江西师范大学学报（自然科学版）》2004年第6期485-488,共4页Journal of Jiangxi Normal University(Natural Science Edition)

基　　金：国家自然科学基金 (6 0 2 6 30 0 5 );全国教育科学规划重点课题 (DBB0 1 0 5 0 1 );江西省自然科学基金 (0 4 1 1 0 2 1 )资助项目

摘　　要：该文对应用BP神经网络和降维法相结合在 0 - 1记分模式下估计项目参数和考生能力的方法作了概述 。The item parameters and the person ability parameter in Logistic Models based on IRT are estimated with Back-Propagation Neural Network. The dimension of scoring matrices X is reduced by using scoring rate, passing rate, coefficient of correlation, guess rate when estimating corresponding ability parameter and item parameters. The results drawn from Monte Carlo study show that the item parameters estimation is more precise than the current prevailing softwares.The focus of this paper is to discuss how to apply this estimation method to practice.

关 键 词：降维法 考生 IRT 实际 能力 记分 模式 参数估计 对应 

分 类 号：O241[理学—计算数学] O178[理学—数学]