基于关键词挖掘的客户细分方法

1)深圳大学管理学院,广东深圳 518060; 2)深圳大学人因工程研究所,广东深圳 518060

人工智能; 自然语言处理; 知识工程; 客户细分; 关键词挖掘; 客户特征; 数据挖掘

A keyword-based mining method for customer segmentation
Chen Xingyu1, Zhou Zhan1, Huang Junwen1, and Tao Da2

Chen Xingyu1, Zhou Zhan1, Huang Junwen1, and Tao Da21)College of Management, Shenzhen University, Shenzhen 518060, Guangdong Province, P.R.China2)Institute of Human Factors and Ergonomics, Shenzhen University, Shenzhen 518060, Guangdong Province, P.R.China

artificial intelligence; natural language processing; knowledge engineering; customer segmentation; keyword mining; customer characteristics; data mining

DOI: 10.3724/SP.J.1249.2017.03300

备注

提出一种基于关键词的数据挖掘方法对客户群进行细分,采用自然语义处理的方法从原始客户信息文本中提取客户特征关键词.再通过人工标记一些与内在特征维度相关的关键词,基于这些关键词找到特征客户.最后以特征客户作为训练集,获得更多关于某个维度内客户特征的关键词,再进行新一轮的客户细分.经此模式学习过程,得到基于内在特征维度的客户细分群体.通过与采用随机选择特征关键词的基准化方法进行自动客户细分结果对比,发现采用基于关键词数据挖掘的自动客户细分结果得到的准确度更高,结果更稳健.

We propose a novel customer segmentation method using keyword-based data mining approach. First, keywords about customer characteristics from original customer information are extracted by natural semantic processing. Then, keywords related to intrinsic characteristics are tagged. Based on the keywords, customers with the specific characteristics are identified. Finally, we use the identified customers as the training samples to obtain more keywords about the customer characteristics, and conduct a new round of customer segmentation. After the learning process, customer segmentation groups based on intrinsic characteristics are obtained. Compared with the benchmarking method of random selection of feature keywords for customer segmentation, the feasibility and validity of the proposed method are verified by a case study where a high level of accuracy rate and robustness is observed in the customer segmentation results.

·