CRM AND DATA MINING TECHNIQUES: A LITERATURE REVIEW & CLASSIFICATION

CRM AND DATA MINING TECHNIQUES: A LITERATURE REVIEW & CLASSIFICATION

Dr.LAKSHMI.P

Assistant Professor, Department of Management IFIM College, Bangalore

Abstract

Prof.VEENA.N

Assistant Professor, Department of Computer Sciences IFIM College, Bangalore

Data Mining is a continuous business process for exploring large amounts of data to discover meaningful patterns and rules, which readily fits in with other strategies for understanding markets and customers .The key is to recognize the flow of customers and the commonalities of data across the enterprise. The current paper is an attempt to conduct an in depth review of different articles pertaining to data mining technologies to customer relationship management, which consists of four dimensions like Customer Identification, Customer Attraction, Customer Retention, Customer Development.

Keywords: CRM, Data Mining, business, customer

Introduction

Information systems and technology are contributing to ever-growing volumes of user profiles of consumers. Even so, most of the profit-generating knowledge about consumers and their habits is still shrouded in mystery. Coupled with the current focus on consumer management, data mining has an excellent opportunity to help the role of marketing to determine which business strategies to follow. Data mining helps organisations to find previously unknown customers, predict future actions, and allow organisations to proactively gather knowledge. Once predictive analyses can process large amounts of new and evolving data in real time, these simulations can include information that was never before possible, such as planning for the future. Data mining tools satisfy complex questions that previously needed in-depth investigation. In any sector, understanding the needs of the customers is of utmost importance, no matter what it is the business you have to give them. Many different models and techniques were generated from the process of segmenting clients according to statistical properties (such as total amount of money received), followed by data mining models (like classification models), then going on to include neural networks and beyond. When we talk about customers, we also include the CRM department. Most major corporations apply their own model to analyse consumers' trends, define their behaviours, and eventually discern their desires. Towards the end of the decade, numerous creative works and new technologies were created in this area. Rogiels, Wang and Yen 2002: Many data mining techniques from discovery, predictive modelling, and forensic analysis have been identified in their review in 2012. In terms of business analytics, they start from the premise that customer relationship management is achievable thanks to data mining techniques that have become actual tools in the response to

our questions. they say that the priorities of managers have changed from product to customer management. There are two ways to accomplish this goal: to focus on what customers value the most, and to identify the methods to maximize customers equity. Shaving the same idea, the focus on customers value, Verhoef, Doorn and Dorotic (2007) make a wide literature review and read to the conclusion that customer lifetime value is a research area that has too few studies. The research in data mining has gained a high attraction due to the importance of its applications and the increasing generation information(Radhakrishnan et al., 2013). Data Mining technology allows marketing organizations to better understand their customers and respond to their needs.

Objectives

1. To analyse the possibilities of data mining techniques in improving customer relations.
2. To examine the web personalization strategies adopted by major e-commerce sites

Literature review

Customers whether they are individuals, households, or business, change over time. (Sheth et al.,2000). The business relationship with a customer evolves over time. Although each business is different, the customer relationship places customers into five major phases like Prospects, Responders, New customers, Established customers and former customers. Business processes move customers from one phase of the customer life cycle to the next. These business processes are important because they lead to making customers more valuable over time. Customer Acquisition is the process of attracting prospectus and turning them into customers. Albeit this is often done by advertising and word of mouth ,data mining is also play an important role. Available data limits the role that data mining can play. Response modeling is used for channels such as direct mail and telemarketing, where the cost of contact is relatively high. The goal is to limit the contacts to prospects that are more likely respond and become good customers and become good customers data available for such endeavours fall into 3 categories source of prospect appended individual household data appended demographic data at a geographic level

.Data mining can play many roles in prospecting the most important of these are identifying good prospects choosing a communication channel reaching prospects picking appropriate messages for different groups of prospects .

It should be noted that customers drive every organization's revenues. Acquiring new customers, delighting existing customers and predicting buyer behaviour will improve the availability and profitability of products and services. Therefore the ultimate objective of any DM exercise in e-commerce is to improve processes which contribute to the value of the end customer. Consider an online store such as http:www.amazon.com in which the customer can select a product of their choice, order the same, track movement and pay for the product and services. Amazon has the potential to make the retail experience outstanding with the technology behind such a website. The details available in web log files will, in the most basic level, illuminate what prospective customers want from a platform. Are they shopping or just browsing intentionally? They're familiar with buying something or something they know nothing about? Are they shopping from home, work or a hotel? The information in the log files is frequently utilised for determining what profiling in the context can be dynamically

processed and indexed into the dynamic HTML generation and how the servers and network perform to help customer service and to make e-business interaction efficient. New computer systems have been designed to warn consumers of major incidents of concern. The paper by Jeng and Drissi (2000) describes a tool called PENS that not only notifies customers of things that happen, but also is able to anticipate scenarios and events. The event notification system in PENS (Planning, Engagement, Systems, Registries, and Collaboration Infrastructure) has the following components: Association rules and cluster analysis are the algorithms used for forecasting events. This e-commerce business introduces the PENS method to help itself better predict the needs of product category demand. How the data science of shopping has been used to evaluate how consumers respond to rewards offered by a credit card company (Zhang et al 2003). Fuzzy and conditional logic can be used to determine whether-or-and-and laws for personalised product strategy, Niu et al (2002) describe a method of building customer profiles using product hierarchy They sort each customer by demographic data, then by behavioural results, and lastly by association rules. Since customers are segmented according to their interests, the authors produce two kinds of recommendations, interests based recommendations. They propose a specialised search/matching data structure called a "tree"

Data Mining in web personalization

A Guide to Customizing the Personalization and on Site Use, Mobasher (2004) provides an overview of how to customise the web from the perspective of Web Mining. Preprocessing and combining data from different sources are standard practises in this process. Any visitor's navigational behaviour is represented in the log data captured by the Web and application servers.

Data mining techniques in CRM assist the organisation in identifying and selecting appropriate data. This information can then be used to create a simple picture of the consumer life cycle. Customer detection, attraction, retention, and growth are all part of the life-cycle. The more data in the database, the more detailed the models generated would be, and hence the greater the value gained. Predictive modelling, forecasting, and descriptive modelling techniques are popular techniques used in data mining. CRM allows a company to participate in a variety of useful practises in the era of data analytics. You can control customer satisfaction, select the appropriate segments, establish optimal pricing strategies, and rank suppliers according to your requirements.

According to Swift (2001), Parvatiyar and Sheth (2001, p. 5) and Kracklauer, Mills, and Seifert (2004, p. 4), CRM consists of four dimensions: (1) Customer Identification; (2) Customer Attraction; (3) Customer Retention; (4) Customer Development

Applications of Data Mining in CRM

Basket Analysis: Discover which products customers often buy together. Stocking, store layout plans, and promotions will all benefit from this experience.( Ailawadi et al., 2006) Sales Forecasting: Businesses can make better re-stocking decisions by looking at time-based trends. It also aids supply chain management, financial management, and provides full control over internal operations. (Hiles, 2004)

Database Marketing: Retailers can build consumer profiles based on demographics, tastes, interests, and purchasing habits. It will also help the marketing team create the best marketing plans and promotional deals possible. This would result in increased efficiency, optimum

resource utilisation, and a favourable return on investment. (Tynan & Drayton,1987).

Market Segmentation: Find out who would be interested in buying your products. Create marketing campaigns and promotions that cater to their tastes and desires. Since you won't be approaching consumers who aren't interested in your product, this will improve productivity and yield the desired ROI.

Product Customization: Manufacturers have the ability to customise goods to suit the unique needs of consumers. They must be able to anticipate which features should be bundled to satisfy consumer demand in order to accomplish this.

Fraud Detection: You can avoid potential fraudulent transactions by reviewing previous transactions that turned out to be fraudulent. This function would greatly benefit banks and other financial institutions by lowering the amount of bad debts.

Techniques of Data Mining for CRM

Anomaly Detection: Anomaly detection is when you look for information that doesn't fit expected behaviour or a predicted trend. Since they differ from the average in the data collection, anomalies may provide useful information.

Association:The goal of association is to create relationships between things that are found together in a record. (Ahmed, 2004; Jiao, Zhang, & Helander, 2006; Mitra et al.,2002). Association modelling is widely used in market basket research and cross-selling systems, to name a few examples. Statistics and apriori algorithms are popular methods for association modelling.

Clustering:The process of segmenting a heterogeneous population into a number of more homogeneous clusters is known as clustering. In contrast to classification, clusters are uncertain at the start of the algorithm. In other words, no predefined clusters exist. Neural networks and discrimination analysis are two popular clustering methods. (Ahmed, 2004; Berry & Linoff, 2004; Carrier & Povel, 2003).

Classification:In data mining, classification is one of the most popular learning models. (Ahmed, 2004; Berry & Linoff, 2004; Carrier & Povel, 2003). Its aim is to construct a model that can predict future customer behaviour by categorising database records into a collection of predefined classes based on certain parameters. (Ahmed, 2004; Berson et al., 2000; Chen, Hsu, & Chou, 2003; Mitra et al., 2002). Neural networks, decision trees, and if then-else rules are popular classification methods.

Regression:One of the more advanced data mining techniques in CRM is regression analysis. The aim is to identify the relationships between various data items and map out which variables are influenced by others. This method is used to figure out how satisfied customers are and how it affects consumer loyalty. Each data object is mapped to a real value using regression, a statistical estimation technique that provide a value for prediction (Carrier & Povel, 2003; Mitra et al., 2002). Linear regression and logistic regression are two popular regression tools.

Sequence discovery:The discovery of connections or patterns over time is known as sequence discovery. (Berson et al., 2000; Carrier & Povel, 2003; Mitra et al., 2002). Its aim is to model the states of the process that generates the series, as well as to extract and report deviations and trends over time (Mitra et al., 2002). Tools for sequence discovery are statistics and set theory. Visualization :Visualization refers to the presentation of data that users can view in complex patterns (Shaw et al., 2001). It's used in combination with other data mining models to provide a deeper understanding of the trends or relationships that have been discovered (Turban et al., 2007). Visualization model are 3D graphs, ‘‘Hygraphs” and ‘‘SeeNet” (Shaw et al., 2001)

The following are some widely used algorithms in Data Mining

1. Association rule;
2. Decision tree;
3. Genetic algorithm;
4. Neural networks;
5. K-Nearest neighbour;
6. Linear/logistic regression.

Table 1: Distribution of articles according to Data mining Techniques

CRM Dimensions	CRM Eleme nts	Data Mining Functions	Data Mining Techniques	References
Customer Identification	Segme ntation	Classification	Decision tree	Kim, Jung, Suh, and Hwang (2006)
			Self-organizing map, decision tree and Markov Chain Model	Ha, Bae, and Park (2002)
		Clustering	K-means	Dennis, Marsland, and Cockett (2001), Raja et al.,(2020)
			Data envelopment analysis, self organizing map and decision tree	Lee and Park (2005)
			Pattern based cluster	Yang and Padmanabhan (2005)
			Self-organizing map	Bae, Park, and Ha (2003), Verdú, García, Senabre, Marín, and Franco (2006)
		Regression	Logistic regression	Hwang, Jung, and Suh (2004)
	Target custom er	Classification	Decision tree	Chen et al. (2003), Wu, Kao, Su, and Wu (2005),
				Zhang, and Zhang (2005)
		Clustering	Self-organizing map	Lee, Suh, Kim, and Lee (2004)
		Visualization	Customer map	Woo et al. (2005)
Customer Attraction	Direct market ing	Regression	Logistic regression	Prinzie and Poel (2005)
		Classification	Bayesian network classifier	Baesens et al. (2002)
			Decision tree	Buckinx et al. (2004)
			Genetic algorithm	Ahn et al. (2006), Chiu (2002)
			Neural network and genetic algorithm	Kim and Street (2004)
		Clustering	Outlier detection	He et al. (2004)
Customer Retention	Compl aints manag ement	Clustering	Self-organizing map	Bae et al. (2005)
		Sequence discovery	Survival analysis	LarivièRe and Poel (2005)
	Loyalt y progra m	Classification	Decision tree	Cox (2002), Douglas et al. (2005), Larivie‘Re and Poel (2005), Chen et al.,(2020)
			Genetic algorithm	Kim et al. (2003)
			Logical analysis of data	Lejeune (2001)
			Neural network, K-nearest	Datta et al. (2000)

CRM Dimensions	CRM Eleme nts	Data Mining Functions	Data Mining Techniques	References
			neighbor and decision tree
			Classification and regression tree and multivariate	Lee et al. (2006)
			adaptive regression splines
Customer Development	Lifeti me value	Classification	Bayesian network classifier	Baesens et al. (2004)
		Clustering	Neural network	Drew et al. (2001)
			Survival analysis	Rosset et al. (2003)
		Forecasting	Markov chain model	Etzion et al. (2005)
		Regression	Linear regression	Verhoef and Donkers (2001)
	Market basket analysi s	Association	Association rules	Aggarval and Yu (2002), Brijs et al. (2004), Jukic and Nestorov(2006)
			Markov chain model	Giudici and Passerone (2002)
		Sequence discovery	Association rules	Chen, Tang, Shen, and Hu (2005), Kubat et al. (2003)
	Up/cro ss selling	Association	Neural network and association rule	Changchien et al. (2004)
		Sequence discovery	Mixture transition distribution	Prinzie and Poel (2006)

Table No: 2 Quantification of Articles by CRM and Data mining models

CRM Dimensions	CRM Elements	Data Mining Model	Number of Articles
Customer Identification	Customer Segmentation		9
		Classification		2
		Clustering		6
		Regression		1
	Target Customer Analysis		5
		Classification		3
		Clustering		1
		Visualization		1
					14
Customer Attraction	Direct marketing		7
		Regression		1
		Classification		5
		Clustering		1
					7
Customer Retention	Complaints Management		2
		Clustering		1
		Sequence Discovery		1
	Loyalty program	Classification	8	8
					10
Customer Development	Customer lifetime Value		5
		Classification		1
		Clustering		2

CRM Dimensions	CRM Elements	Data Mining Model	Number of Articles
		Forecasting		1
		Regression		1
	Market basket		6
		Association		4
		Sequence Discovery		2
	Up/cross selling		2
		Association		1
		Sequence Discovery		1
					13

> CRM elements identified with customer identification are customer segmentation and target customer analysis.The data mining models used for customer segmentation and target analysis are classification,clustering,regression and visualisation.14 articles were reviewed u Discussion

The current study gives an in-depth review of different articles pertaining to techniques of data mining in CRM. For this study around 46 articles have been reviewed. Each article was carefully analysed and separately classified according to the four categories of CRM dimension and seven categories of data mining models, which serves as a comprehensive base for understanding different data mining techniques and functions. Even though there is no specific definition for CRM many scholars have tried to define it in their own ways. CRM as a ‘‘enterprise approach to understanding and shaping customer behaviour by meaningful interactions in order to improve customer acquisition, retention, and satisfaction customer profitability and loyalty” (Ling & Yen, 2001; Ngai, 2005). Swift (2001, p. 12). CRM is described as "the strategic use of knowledge, processes, technology, and people to manage your customer's relationship with your company (Marketing, Sales, Services, and Support) across the customer life cycle."(Kincaid (2003, p. 41). In today’s world E-commerce site are widely used, profitability of E-Commerce sites is linked with customers. So a prolific CRM is essential. The current study analyses different techniques and tools which can be used to boost the sales of E-commerce sites.

Conclusion

Application of data mining techniques in CRM is an developing trend in the industry. It has enticed the interest of practitioners and academics all round the globe. CRM which is an emerging trend which can be linked to profitability of an enterprise includes a set of processes and supporting systems to boost up marketing strategies to build long term relationships with customers who are in turn stake holders of the company. Data mining is used in organization for decision making and forecasting of prospective customers. We have studied recent literature related to use of data mining techniques for CRM. Based on review of the contemporary literature, we analyzed different data mining techniques employed in different types of business, corporate sectors and organizations. We draw a critical review table which provides the problem addressed, proposed techniques, significance, limitations and suggested possible improvements for each proposed techniques review during this study. The critical review of the data mining techniques which are being used for CRM is provided in this paper.

References

1. Aggarval, C. C., & Yu, P. S. (2002). Finding localized associations in market basket data. IEEE Transactions on Knowledge and Data Engineering, 14, 51–62.
2. Ahmed, S. R. (2004). Applications of data mining in retail business. Information Technology: Coding and Computing, 2, 455–459.
3. Ahn, H., Kim, K. J., & Han, I. (2006). Hybrid genetic algorithms and case-based reasoning systems for customer classification. Expert Systems with Applications, 23, 127–144.
4. Ailawadi, K. L., Harlam, B. A., Cesar, J., & Trounce, D. (2006). Promotion profitability for a retailer: the role of promotion, brand, category, and store characteristics. Journal of Marketing Research, 43(4), 518-535.
5. Baesens, B., Verstraeten, G., Dirk, V. D. P., Michael, E. P., Kenhove, V. K., & Vanthienen, J. (2004). Bayesian network classifiers for identifying the slope of
6. Berry, M. J. A., & Linoff, G. S. (2004). Data mining techniques second edition – for marketing, sales, and customer relationship management. Wiley.
7. Berson, A., Smith, S., & Thearling, K. (2000). Building data mining applications for CRM. McGraw-Hill.
8. Buckinx, W., Moons, E., Poel, D. V. D., & Wets, G. (2004). Customer-adapted coupon targeting using feature selection. Expert Systems with Applications, 26, 509–518.
9. Carrier, C. G., & Povel, O. (2003). Characterising data mining software. Intelligent Data Analysis, 7, 181–192.
10. Changchien, S. W., Lee, C. F., & Hsu, Y. J. (2004). On-line personalized sales promotion in electronic commerce. Expert Systems with Applications, 27, 35–52.
11. Chen, C., Geng, L., & Zhou, S. (2020). Design and implementation of bank CRM system based on decision tree algorithm. Neural Computing and Applications, 1-11.
12. Chen, Y. L., Hsu, C. L., & Chou, S. C. (2003). Constructing a multi-valued and multilabeled decision tree. Expert Systems with Applications, 25, 199–209.
13. Chen, Y. L., Hsu, C. L., & Chou, S. C. (2003). Constructing a multi-valued and multi-labeled
14. Cox, L. A. (2002). Data mining and causal modeling of customer behaviors. Telecommunication Systems, 21, 349–381.
15. Data Engineering, 17, 1300–1304.
16. Datta, P., Masand, B., Mani, D. R., & Li, B. (2000). Automated cellular modeling and prediction on a large scale. Artificial Intelligence Review, 14, 485–502.
17. decision tree. Expert Systems with Applications, 25, 199–209.
18. Dennis, C., Marsland, D., & Cockett, T. (2001). Data mining for shopping centres customer knowledge management framework. Journal of Knowledge Management, 5, 368–374.
19. Drew, J. H., Mani, D. R., Betz, A. L., & Datta, P. (2001). Targeting customers with statistical and data-mining techniques. Journal of Service Research, 3, 205– 220.
20. Etzion, O., Fisher, A., & Wasserkrug, S. (2005). E-CLV: A modeling approach for customer lifetime evaluation in e-commerce domains, with an application and case study for online auction. Information Systems Frontiers, 7, 421–434.
21. for decision support. Data Mining and Knowledge Discovery, 7, 321–339.
22. Giudici, P., & Passerone, G. (2002). Data mining of association structures to model consumer behaviour. Computational Statistics and Data Analysis, 38, 533–541.
23. Ha, S. H. (2002). Helping online customers decide through web personalization. IEEE Intelligent Systems, 17, 34–43.
24. Ha, S. H., Bae, S. M., & Park, S. C. (2002). Customer’s time-variant purchase behaviour and corresponding marketing strategies: An online retailer’s case. Computers and Industrial Engineering, 43, 801–820.
25. Ha, S. H., Bae, S. M., & Park, S. C. (2006). Digital content recommender on the internet. IEEE Intelligent Systems, 21, 70–77.
26. Hiles, A. (2004). Business continuity: best practices: world-class business continuity management. Rothstein Associates Inc.
27. Hwang, H., Jung, T., & Suh, E. (2004). An LTV model and customer segmentation based on customer value: A case study on the wireless telecommunication industry. Expert Systems with Applications, 26, 181–188.
28. Jeng, J. J., & Drissi, Y. (2000, October). Pens: a predictive event notification system for e-commerce environment. In Proceedings 24th Annual International Computer Software and Applications Conference. COMPSAC2000 (pp. 93-98). IEEE.
29. Kim, S. Y., Jung, T. S., Suh, E. H., & Hwang, H. S. (2006). Customer segmentation and strategy development based on customer lifetime value: A case study. Expert Systems with Applications, 31, 101–107.
30. Kincaid, J. W. (2003). Customer relationship management: Getting it right. Upper saddle river. N.J.: Prentice Hall PTR.
31. Larivie‘Re, B., & Poel, D. V. D. (2005). Predicting customer retention and profitability by using random forests and regression forests techniques. Expert Systems with Applications, 29, 472–484
32. Lee, S. C., Suh, Y. H., Kim, J. K., & Lee, K. J. (2004). A cross-national market segmentation of online game industry using SOM. Expert Systems with Applications, 27, 559–570.
33. Lejeune, M. A. P. M. (2001). Measuring the impact of data mining on churn management. Internet Research: Electronic Networking Applications and Policy, 11,375–387.
34. Ling, R., & Yen, D. C. (2001). Customer relationship management: An analysis framework and implementation strategies. Journal of Computer Information Systems, 41, 82–97.
35. Mitra, S., Pal, S. K., & Mitra, P. (2002). Data mining in soft computing framework: A survey. IEEE Transactions on Neural Networks, 13, 3–14.
36. Ngai, E. W., Xiu, L., & Chau, D. C. (2009). Application of data mining techniques in customer relationship management: A literature review and classification. Expert systems with applications, 36(2), 2592-2602.
37. Prinzie, A., & Poel, D. V. D. (2006). Investigating purchasing-sequence patterns for financial services using Markov, MTD and MTDg models. European Journal of Operational Research, 170, 710-734.
38. Radhakrishnan, B., Shineraj, G., & Anver Muhammed, K. M. (2013). Application of data mining in marketing. International Journal of Computer Science and Network, 2(5), 41-46.
39. Raja, N. K., Masthan, K., Balachandra Pattanaik, D. B., & Kumar, M. B. S. (2020). Data Mining Model to Analyse Crm in Banking Sector. PalArch's Journal of Archaeology of Egypt/Egyptology, 17(7), 7170-7181.
40. Research, 156, 508–523.
41. Rosset, S., Neumann, E., Eick, U., & Vatnik, N. (2003). Customer lifetime value models
42. Rygielski, C., Wang, J. C., & Yen, D. C. (2002). Data mining techniques for customer relationship management. Technology in society, 24(4), 483-502.
43. Shaw, M. J., Subramaniam, C., Tan, G. W., & Welge, M. E. (2001). Knowledge management and data mining for marketing. Decision Support Systems, 31, 127–137.
44. Sheth, J. N., Sisodia, R. S., & Sharma, A. (2000). The antecedents and consequences of customer-centric marketing. Journal of the Academy of marketing Science, 28(1), 55-66.
45. Swift, R. S. (2001). Accelarating customer relationships: Using CRM and relationship technologies. Upper saddle river. N.J.: Prentice Hall PTR
46. the customer-lifecycle of long-life customers. European Journal of Operational
47. Tudorache, I. C., & Vija, R. I. (2015). Data Mining and Customer Relationship Management for Clients Segmentation. International Journal of Economic Practices and Theories, 5(5), 571-57
48. Turban, E., Aronson, J. E., Liang, T. P., & Sharda, R. (2007). Decision support and business intelligence systems (Eighth ed.). Pearson Education.
49. Tynan, A. C., & Drayton, J. (1987). Market segmentation. Journal of marketing management, 2(3), 301-335.
50. Verhoef, P. C., Van Doorn, J., & Dorotic, M. (2007). Customer value management: an overview and research agenda. Marketing-Journal of Research and Management, 3(2), 105-120.
51. Woo, J. Y., Bae, S. M., & Park, S. C. (2005). Visualization method for customer targeting using customer map. Expert Systems with Applications, 28, 763–772.
52. Wu, C. H., Kao, S. C., Su, Y. Y., & Wu, C. C. (2005). Targeting customers via discovery knowledge for the insurance industry. Expert Systems with Applications, 29, 291–299.
53. Yang, X. (2021). Algorihm Study of New Association Rules and Classification Rules in Data Mining. In Journal of Physics: Conference Series (Vol. 1732, No. 1, p. 012070). IOP Publishing.
54. Yang, Y., & Padmanabhan, B. (2005). A hierarchical pattern-based clustering algorithm for grouping web transactions. IEEE Transaction on Knowledge and
55. Zhang, S., Zhang, C., & Yang, Q. (2003). Data preparation for data mining. Applied artificial intelligence, 17(5-6), 375-381.