摘要
本文研究讨论了神经网络在情报处理与信息咨询中的应用及其用于预测问题的范围和限制，利用ＢＰ算法提出了时间序列和信息预测方法，在此基础上对大豆单产进行了预测，并对比证实了本方法比统计回归模型有较强的预测能力。
This study discusses the applications of neural networks to information processing and consulting,and the scope and limitations for forecasting problems as well.An information forecasting method for time series is presented using BP arithmetic and on this grounds an example for forecasting per unit area yield of soybean is provided.By comparing the method used in the example here and statistical regression model it is confirmed that the former possesses more powerful forecasting ability.
出处
《情报学报》
CSSCI
北大核心
1999年第2期105-112,共8页
Journal of the China Society for Scientific and Technical Information
关键词
人工神经网络
预测
情报分析
artificial neural network,forecasting,information analysis.
