Processing

Please wait...

Settings

Settings

Goto Application

1. CN106096052 - Consumer clustering method facing Wechat marketing

Office China
Application Number 201610497893.4
Application Date 25.06.2016
Publication Number 106096052
Publication Date 09.11.2016
Publication Kind A
IPC
G06F 17/30
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
30Information retrieval; Database structures therefor
G06Q 30/02
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
30Commerce, e.g. shopping or e-commerce
02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
G06Q 50/00
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
50Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
CPC
G06Q 50/01
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
50Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
01Social networking
G06F 16/95
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
90Details of database functions independent of the retrieved data types
95Retrieval from the web
G06Q 30/0201
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
30Commerce, e.g. shopping or e-commerce
02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
0201Market data gathering, market analysis or market modelling
Applicants CHINA TOBACCO ZHEJIANG INDUSTRIAL CO., LTD.
Inventors GAO YANGHUA
LU HAILIANG
SHAN YUXIANG
YU GANG
Title
(EN) Consumer clustering method facing Wechat marketing
(ZH) 一种面向微信营销的消费者聚类方法
Abstract
(EN)
The invention relates to the field of social network data processing, in particular to a consumer clustering method facing Wechat marketing. The method comprises the following steps that 1, a complete data set S is subjected to searching, and random sampling is conducted on the complete data set J times; 2, k-means algorithm clustering is conducted on a sample data set obtained after random sampling is conducted every time to obtain a clustering center, J times of sampling is conducted, and J clustering centers can be obtained in total; 3, the optimal clustering center is found by means of a sum-of-squared-error criterion function and output; 4, a k-means algorithm is conducted on the complete data set by taking the optimal clustering center found in the third step as the initial clustering center and taking K as an input parameter, wherein K is larger than J; 5, in the generated K clusters, the two clusters with the smallest distance are merged, the clustering center after merging is recalculated till the number of the clusters is decreased to R, merging is stopped, and the whole algorithm is ended. By means of the consumer clustering method facing Wechat marketing, the speed of the consumer data clustering process is increased, and the stability of the consumer data clustering process is improved.

(ZH)
本发明涉及社交网络数据处理领域,尤其涉及一种面向微信营销的消费者聚类处理方法,该方法包括以下的步骤:1)对数据全集进行搜索,通过对数据全集随机取样次;2)对每次随机取样后的样本数据集进行k‑means算法聚类,获得一组聚类中心,次取样,共可获得组聚类中心;3)利用误差平方和准则函数,寻找到最优的一组聚类中心,并输出;4)以步骤3)寻找到的最优聚类中心为初始聚类中心,为输入参数(),对数据全集执行k‑means算法;5)在产生的组聚类中,合并距离最近的两组,重新计算合并后的聚类中心;直到聚类数目减少到,停止合并;整个算法结束。该方法提高了消费者数据聚类过程的速率与稳定性。