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。     

An Algorithm for the Fixed-Charge Assigning Users to Sources Problem

Given a set of sources with known capacities and a set of users with known demands for a product, the objective is to minimize the cost of assigning the users to the sources. The restrictions are that each user's demand is satisfied by exactly one of the sources, and that the total amount assigned to each source does not exceed its capacity. A fixed-charge version of this problem is formulated as a set partitioning problem and solved by a column-generating branch and bound procedure. Computational results are given.