基于数据挖掘的呼叫中心数据分析与研究

发布时间：2018-10-26 07:32

【摘要】：伴随着三网融合，网络技术日益成熟和普及所产生的大量互联网信息数据逐渐被人们所关注，海量信息数据所蕴含的巨大价值也受到越来越多的重视，，而数据挖掘作为解决海量信息数据价值转换的一个重要手段，已成为当前最热门的研究领域。从巨大的、复杂的数据中获取隐藏的信息的过程，就是数据挖掘，比如说对客户进行分类、聚类、识别欺诈行为、挖掘潜在顾客等，大多应用在零售业、金融业、医疗机构、政府机构、公司财务等领域。但是，海量的信息数据在展示出巨大的商业活动信息同时也带来了一系列挑战：一是海量信息数据大得无法想象，难以被有效的利用起来；二是难以辨别信息真伪，而虚假信息的产生源于互联网数据过于开放；三是由于信息表现形式不一致，导致难以对其进行统一处理，正是这些挑战推动着数据挖掘技术的革新和完善。呼叫中心在我国起源于上世纪80年代，并成为企业与顾客之间最直接的沟通渠道，市场经济迅速发展，呼叫中心产生的数据越来越繁杂和巨大。但通过调查发现，大量企业对呼叫中心业务数据仅仅是进行简单的备份和存储，忽视了这些数据中隐藏的客户价值，并未对数据信息进行有效的开发和利用。面对日益激烈的市场竞争，企业如何利用这些数据进一步挖掘高质量目标客户、精细化企业客户分类、制定精准的营销策略、提高核心竞争力，从而为企业管理决策提供有效的支持，已成为各大企业的当务之急。本文在大量阅读了国内外文献和进行企业实例调查的基础上，结合前人研究成果，进一步完善了数据挖掘技术在呼叫中心领域的运用。首先介绍了数据挖掘常用算法的原理，并就呼叫中心应用中如何开发数据挖掘工具进行描述和说明。然后在对国内外数据挖掘应用研究进行归纳总结的基础上，根据呼叫中心数据特点找到一种高效的K-means聚类算法，设计出符合呼叫中心业务数据特点的挖掘系统。最后，论文以移动话费营销呼叫中心为例通过该系统对呼叫中心数据进行了有效准确的分析，以客户的数据业务消费信息为对象进行数据挖掘，找出了可能的高价值客户信息。
[Abstract]:With the integration of three networks, a large number of Internet information data, which is produced by the increasingly mature and popularization of network technology, has gradually been concerned by people, and the huge value of massive information data has also been paid more and more attention. As an important means to solve the value conversion of massive information data, data mining has become the most popular research field. The process of extracting hidden information from huge, complex data is data mining, such as categorizing customers, clustering, identifying fraud, mining potential customers, etc., mostly in retail, financial, medical, etc. Government agencies, corporate finance, etc. However, the huge amount of information data also brings a series of challenges: first, the massive information data is too big to be imagined, and it is difficult to be used effectively; Second, it is difficult to distinguish the authenticity of information, and the generation of false information from the Internet data is too open; The third is that it is difficult to deal with information in a uniform way because of the inconsistent forms of information expression. It is these challenges that promote the innovation and improvement of data mining technology. Call center originated in 1980s in our country and has become the most direct communication channel between enterprises and customers. With the rapid development of market economy, the data generated by call center is more and more complicated and huge. However, through the investigation, it is found that a large number of enterprises only backup and store the call center business data simply, ignoring the customer value hidden in these data, and not effectively developing and utilizing the data information. In the face of increasingly fierce market competition, how to use these data to further tap high quality target customers, refine the classification of enterprise customers, formulate accurate marketing strategies, improve the core competitiveness, In order to provide effective support for enterprise management decisions, has become an urgent matter for major enterprises. On the basis of reading a large number of domestic and foreign literature and carrying out enterprise case investigation, this paper further consummates the application of data mining technology in the field of call center, combined with the previous research results. This paper first introduces the principle of common algorithms of data mining, and describes how to develop data mining tools in the application of call center. Then, on the basis of summarizing the domestic and foreign data mining application research, we find an efficient K-means clustering algorithm according to the characteristics of call center data, and design a mining system which accords with the characteristics of call center business data. Finally, the paper takes the mobile call center as an example to analyze the data of the call center effectively and accurately, and takes the customer's data consumption information as the object to mine the data, and finds out the possible high-value customer information.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP311.13

【参考文献】