Analysing the Customer Behaviour based on CBA(DataMining Approach)

K.V.Nagendra,
G.Srujana Kumari M-Tech,(cs) student G.RaviKumar, Asso.Professor, CMR College of engg & Tech.

Asst.professor,Dept.Of Computers, ASCET, Gudur. Mail:kvnagendramtech@gmail.com

HOD, Dept of Computers

ASCET,

C.Rajendra,

ASCET ,Gudur.

Gudur.

Mail: sruji7@gmail.com, Medchal,HYD.

Abstrac:t With the rampant competition in the domestic and international business the customer relationship management has become one of the matters of concern to the enterprise.CRM takes the customer as the center. CRM can be defined as the process of classifying customer behavior and selecting actions to influence that behavior to benefit the company. It gives a new life to the enterprise organization system and optimizes the business process. Customer satisfaction can also be improved through more effective marketing.In an effort to help enterprises understand their customer shopping behaviourand the ways to retain valued customers we propose datamining techniques.Data mining, the extraction of hidden predictive information from large databases, is a powerful new technology with great potential to help companies focus on the most important information in their data warehouses. Data mining tools predict future trends and behaviors, allowing businesses to make proactive, knowledge-driven decisions. As the organization mainly concentrates on the behaviour of the potential customers for any new proposals.Here the potential customers of a commercial bank are predicted and finding the type of loan they are taking by taking the loan as the class label using the classification based on association.

Usually, companies build a customer prediction model to find the prospects for a specific product. In this case, we classify prospects into either purchasing or non-purchasing groups. This kind of knowledge may create a variety of marketing opportunities for the company such as one-to-one marketing, direct mailing, and sales promotion via telephone or e-mail. As a rising subject, data mining is playing an increasingly important role in the decision support activity of every walk of life, as in [2]. Data mining uses sophisticated statistical processing or artificial intelligence algorithms to discover useful trends and patterns from the extracted data so that it can yield important insights including prediction models and associations that can help companies understand their customer better. Examining and analyzing the data can turn raw data into valuable information about customers needs. For example, faced with the increasingly serious situation in customer churning, enterprises need data mining technology to analyze the churning in order to take measures to maintain valuable customers, and reduce customers churning to lower economic losses. Therefore customer classification and prediction is the base of the implement of CRM. Its the precondition to analyze and forecast customers pattern of consumption, and the premise of personalized marketing services and management. In this study, we suggest a customer classification and prediction model in commercial bank that uses collected As a rising subject, data mining is playing an increasingly important role in the decision support activity of every walk of life, as in [2]. Data mining uses sophisticated statistical processing or artificial intelligence algorithms to discover useful trends and patterns from the extracted data so that it can yield important insights including prediction models and associations that can help companies understand their customer better. Examining and analyzing the data can turn raw data into valuable information about customers needs. For example, faced with the increasingly serious situation in customer churning, enterprises need data mining technology to analyze the churning in order to take measures to maintain valuable customers, and reduce customers churning to lower economic losses. Therefore customer classification and prediction is the base of the implement of CRM. Its the precondition to analyze and

Keywords- customer relationship management ; customer classification and prediction; data mining;classification based on assosiation I. INTRODUCTION With the rampant competition in the domestic and international business, the Customer Relationship Management(CRM) has become one of matters of concern to the enterprise. CRM can be defined as the process of predicting customer behavior and selecting actions to influence that behavior to benefit the company, as in [1], usually leveraging on information technology and databaserelated tools. This important concept has been given a new lease of life because of the growth of the Internet and Ebusiness. Customer Relationship Management takes the customers as the center. It gives a new life to the enterprise organization system and optimizes the business process.By predicting customer needs in advance, business can then market the right products to the right segments at the right time through the right delivery channels. Customer satisfaction can also be improved through more effective marketing. One of the important issues in customer relationship management is customer classification and prediction, by which a company classifies its customers into predefined groups with similar behavior patterns.

forecast customers pattern of consumption, and the premise of personalized marketing services and management. II. LITERATURE REVIEW In recent years, data mining has gained widespread attention and increasing popularity in the commercial world, as in [3],[4]. According to the professional and trade literature, more companies are using data mining as the foundation for strategies that help them outsmart competitors, identify new customers and lower costs, as in [5]. In particular, data mining is widely used in marketing, risk management and fraud control, as in [6]. Although recent surveys found that data mining had grown in usage and effectiveness, data mining applications in the commercial world have not been widely. Literature about data mining applications in the fields about economic and management are still few. With the realization of importance of business intelligence, we need to strengthen the research on data mining applications in the commercial world.
III. CLASSIFICATION BASED ON ASSOCIATION

the derived rules are analyzed and used for classification.One of the simplest algorithms for associative classification is CBA.(Classification Based on Association). CBA uses an iterative approach to frequent itemset mining,similarly as for apriory,where multiple passes are made over the data and the derived frequent itemsets are used to generate and test longer itemsets.In general the number of passes made is equal to the length of the longer rule found.The complete set of rules satisfying minimum confidence and minimum support thresholds are found and then analyzed for inclusion in the classifier. CBA uses a heuristic method to construct the classifier wher the rules are organizedaccording to decreasing precedence based on their confidence and support.If a set of rules has the same antecedent ,then the rule with the highest confidence is selected to represent the set.
Suppose commercial bank wants to find out the potential customers in the bank and wants to find out in which loans they are preferring the most,so that according to this information the bank management wants to provide some extra features for that particular loans in order to attract the customers so that the customers will be benifited by which the organization will be benifited. For this there is a large number of valuable customer information in huge amounts of data accumulated by commercial banks,which is used to identify potential customers and provide decision support We wish to predict the class label of an unknown sample using classification based on association,given the training data as Table1. The data samples are described by the attributes:gender,age,student. The class label attribute ,type of loan has three distinct values(namely,{education,car,housing}). Table 1 Training data from the customer database

IV.AN EXAMPLE

Frequent patterns and their corresponding associations or correlation rules characterize interesting relationships between attributes conditions and class labels and thus have been recently used for effective classification.Association rules show strong associationsbetween attribute value pairs that occure frequently in agiven data set. Association rules are commonly used used to analyze the purchasing patterns of customers in any business application areas. Such analysis is useful in many decision making process such as product placement,catalog design and cross marketing. In this paper we look at associative classification ,where associative rules are generated and analyzed for use in classification. The general idea is that we can search for strong associations between frequent patterns and class . labels.Because associative rules explore highly confident associations among multiple attributes,this approach may overcome some constraints introduced by decision tree induction,which considers only one attribute at a time. In many surveys also associative classification has been found to be more accurate than some traditional classification methods.Association rules are mined in atwo step process consisting of frequent itemset minning,followed by rule generation. The first step searches for pattern of attribute value pairs that occure repeatedly in a dataset where each attribute value pair is considered an item.The resulting attribute value pairs form frequent itemsets. The second step analyzes the frequent itemsets in order to generate association rules.All association rules must satisfy certain criteria regarding their accuracy (or confidence) nad the proportion of the dataset that they actually represent(referred to as support). Methods of associative classification differs primarily in the approach used for frequent itemset mining and in how

Gender male male female male female male male

Age 16-25 30-40 16-25 16-25 30-40 16-25 30-40

Student yes no yes no no yes no

Type of loan education car education car housing education housing

In this each tuple is taken as a transaction with the attribute value in that tuple as an item.The attribute with the

same value in multiple tuple represents the same item,but if the attribute is bearing different values in different tuples it used represent different items.So in this example also we are considering the gender=male-----------------------I11 Gender=femle---------------------I12 Age=16-25------------------------I21 Age=30-40------------------------I22 Student=yes-----------------------I31 Student=no------------------------I32 Type of loan=education---------I41 Type of loan=car-----------------I42 Type of loan=housing-----------I43 Here we are considering support as 2% i.e each item should be present in atleast 2 transaction.If it is not present then it will be rejected from the frequent item set.Thus the number of times the above mensioned items present in all the above tuples are as followes ONE ITEMSET: I11=5,I12=2 I21=4,I22=3 I31=3,I32=4 I41=3,I42=2,I43=2 Now in the above one item set as all the items are satisfying the support value ,thus no item is eliminated.Now similarly we have to find the two item set and find out whether they satisfy the support or not. TWO ITEMSET: I11,I21=3 I21,I41=3 I11,I22=2 I21,I42=1 I11,I31=2 I21,I43=0 I11,I32=3 I22,I31=0 I11,I41=2 I22,I32=3 I11,I42=2 I22,I41=0 I11,I43=1 I22,I42=1 I12,I21=1 I22,I43=2 I12,I31=1 I31,I41=3 I12,I32=1 I31,I42=0 I12,I41=1 I31,I43=0 I12,I42=0 I32,I41=0 I12,I43=1 I32,I42=2 I12,I31=3 I32,I43=2 I21,I32=1 In the above two item set the bold items are not satisfying the support value,thus they have to be eliminated.Now we have to find the three item set. THREE ITEMSET I11,I21,I31=2 V.

I11,I22,I32=2 I11,I21,I41=2 I21,I31,I41=1 I22,I32,I42=1 I22,I32,I43=2 I11,I31,I41=2 I11,I32,I42=2 In the above three item set there are two bold items which are not satisfying the support.Thus they are eliminated from the list.Now the four item set is as followes FOUR ITEMSET I11,I21,I31,I41=2 I11,I22,I3,I43=1 I11,I22,I32,I42=1 In the above four itemset also the two bold items which are not satisfying support value are eliminated and as the further combinations are not possible.Thus the final frequent item is I11,I21,I31,I41. From this we have to generate a n association rule placing all the attributes with their values on antecedent separated by and operator and the class label with its value on consequent side as followes. I11^I21^I31=>I41 i.e gender=male^age=1625^student=yes=>loans=education is acting as the potential customer for our example.But generally there will me more than one frequent items generated.Thus at that time more than one association rules are generated.Though all are frequent among them we have to find the strong association rules by checking whether the association rules are satisfying confidence threshold.If they are satisfied they are acting as the strong association rules.According to the confidence value they are prioritized. Thus in our example the c onfidence for the generated association rule is (support count of antecedent and consequent items) X 100 support count of antecedent item i.e 2x100 =100% 2 Therefore we predict that the potential customers for the given commercial bank are male with age of 16-25 and who are students and they are mostly preferring education loans. CONCLUSION

Data mining provides the technology to analyze mass volume of data and/or detect hidden patterns in data to convert raw data into valuable information. This paper mainly focused on the research of the potential customer classification and prediction in Customer Relation Management concerned with data mining based on Classification based on Association algorithm, which have a try to the optimization of the business process. The study will help the company to analyze and forecast customers pattern of consumption, and the premise of personalized marketing services and management. Although the paper focuses mainly on the banking industry, the issues and applications discussed are applicable to other industries, such as insurance industry, retail industry, manufacture industries, and so on. REFERENCES
[1] Jenkins D(1999). Customer relationship management and the data warehouse [J]. Call center Solutions, 1892):88-92 [9] .

[2] Chung HM and P Gray(1999). Data mining [J]. Journal of MIS,16(1):11-13 [3]HOKEY MIN, Developing the Profiles of Supermarket Customers through Data Mining[J], The Service Industries Journal, Vol.26, No.7, October 2006, pp.747763 [4] Balaji Padmanabhan, Alexander Tuzhilin , On the Use of Optimization for Data Mining: Theoretical Interactions and eCRM Opportunities[J], Management Science . Vol. 49, No. 10, October 2003, pp. 1327 [5] Davis B(1999). Informationweek, 751:86-88 Data mining transformed 1343 [J].

[6] Kuykendall L(1999).The data-mining toolbox[J]. Credit Card Management,12(6):30-40 [7] Han, J. and Kamber M., (2001) Data Mining: Concepts and Techniques[M], San Francisco, CA: Morgan Kaufmann Publishers [8] Kantardzic, M. (2003) Data Mining: Concepts, Models, Methods, and Algorithms[M], Piscataway, NJ: IEEE Press