空巢老人手机用户的精准识别

以神经网络为计算模型,在合理处理已知数据的异常情况后,依据给定指标制定出(非)空巢老人的标准,以此建立(非)空巢老人集合的核作为学习样本,从而获得初步的识别结果,同时从输入和输出2个方面对其进行了多种检验。这套完整流程,也可用于解决类似的精准识别和其他数据挖掘问题。

Accurate Identification Approach to Empty-Nester Mobile-phone Users

In this page, after a proper data processing, we define a judgment upon empty-nesters based on their habits to use the mobile-phone; build kernels for both empty-nester and non-empty-nester groups according to the judgment; standardize samples marked with those kernels; translate the original data into new scores by using factor analysis in order to reduce the differentiation between those kernels; put those scores into neural networks as learning samples; and hence get a preliminary identification followed by four inspections. This approach is also expected to deal with other similar problems of accurate identification and data mining.