论文部分内容阅读
本文依据不同服务模式下图书馆用户参与度的主要评价指标及影响因素,基于BP神经网络算法构建了图书馆用户参与度的预测模型。通过对某市公共图书馆用户参与度的预测与一元非线性回归预测模型对比验证了该模型具有更高的预测精度。分析认为:在全面把握影响用户参与外部条件与内部因素的基础上进行预测模型参数优化,可进一步提高预测结果的可信度。
Based on the main evaluation indexes and influential factors of library users’ participation in different service modes, this paper constructs a predictive model of library user participation based on BP neural network algorithm. The prediction of users’ participation degree in a public library in a city is compared with the one-dimensional nonlinear regression prediction model to verify that the model has higher prediction accuracy. According to the analysis, optimizing the parameters of the forecasting model based on the full understanding of external factors and internal factors that affect users’ participation can further enhance the credibility of the forecasting results.