CAJ | 학술논문

根据BP神经网络强处理非线性问题和遗传算法具有全局寻优的特点，总结出用遗传算法优化BP神经网络的初始权值和阈值，建立了基于遗传算法和BP神经网络的多维客户行为细分模型，对客户进行细分，挖掘出潜力客户，有效降低营销成本。最后，利用Matlab对多组数据进行测试，并与纯BP算法进行分析比较，结果表明该方法是有效可行的。
근거BP신경망락강처리비선성문제화유전산법구유전국심우적특점，총결출용유전산법우화BP신경망락적초시권치화역치，건립료기우유전산법화BP신경망락적다유객호행위세분모형，대객호진행세분，알굴출잠력객호，유효강저영소성본。최후，이용Matlab대다조수거진행측시，병여순BP산법진행분석비교，결과표명해방법시유효가행적。
To take advantage of the BP neural networks's strong to deal with nonlinear problem and the genetic algorithm's globe optimal searching, summarize the initialized weights and threshold of BP neural network was optimized with GA, established multidimensional customer behavior segmentation model, subdividing the customer, dig up the potential customers, effectively reduce the cost of marketing. Finally, multi-group data were used to test by matlab, compared to BP algorithm, the results show that GA-BP is an effective method.