针对目前高校图书馆普遍存在的读者失信问题,根据高校图书馆的实际情况构建相应的读者信用指标体系,将读者信用评价与信用积分制结合起来。信用分由读者的借还行为以及读者在馆内的阅读行为记录共同决定。给各种指标赋予不同的权值,经过加权计算来进行信用评分。采用神经网络的方法对读者信用评价模型进行设计,通过样本测试证明信用评价模型的有效性。 Aiming at the problem of loss of trust in universally existing university libraries, this paper constructs the corresponding readers ’credit index system based on the actual situation of university libraries, and integrates the readers’ credit evaluation and credit integration. Credit points by the reader’s borrowing behavior and reader’s reading behavior records in the hall jointly decided. Give various indicators different weights and conduct weighted calculations for credit scores. The method of neural network is used to design the reader’s credit evaluation model, and the effectiveness of the credit evaluation model is proved through the sample test.