论文部分内容阅读
本文将BP(Back Propagation)神经网络应用到题库试题分值的确定中,以解决目前智能组卷研究中题库试题分值确定的不合理性。在训练网络时,对标准BP算法作了相应改进,以适应该智能模型的建立。通过案例试验,验证了确定试题分值的智能模型的精度是符合实际要求的,在一定程度上为智能化组卷奠定了基础。
In this paper, BP (Back Propagation) neural network is applied to the determination of test questions scores, in order to solve the unreasonableness of determining the score of test questions in the current study. When training the network, the standard BP algorithm has been improved to adapt to the establishment of the intelligent model. Through the case test, it is verified that the accuracy of the intelligent model for determining the score of the test questions meets the actual requirements, which lays the foundation for the intelligent test paper to a certain extent.