11.direct marketing with the application of data mining
1 like458 views
This document summarizes a research paper on using data mining techniques for direct marketing. It discusses how direct marketing focuses on specific customer groups rather than mass marketing. Data mining algorithms like decision trees are used to classify customers as loyal or unloyal based on attributes in customer data. This helps direct marketing efforts towards the most beneficial customers. The document also outlines some common problems in classification for direct marketing like imbalanced data and issues with predictive accuracy, and provides solutions like lift analysis.
![Journal of Information Engineering and Applications www.iiste.org
ISSN 2224-5758 (print) ISSN 2224-896X (online)
Vol 1, No.6, 2011
3. PROBLEMS IN CLASSIFICATION
According to Charles X.Ling and Chenghui Li,the classification of data base involves the following
situations:
In the first situation, some (say X%) of the customers in the database have already bought the product,
through previous mass marketing or passive promotion. X is usually rather small, typically around 1.Data
mining can be used to discover patterns of buyers, in order to single out likely buyers from the current non-
buyers,(100-X%)of all customers. More specifically, data mining for direct marketing in the first situation
can be discovered in:
1. Get the database of all customers, among which X% are buyers.
2. Data mining on the data set based on Geo-demographic information, transforming address and area
codes, deal with missing values, etc.
3. Applying algorithm to prepare objects, classes .
4. Evaluate the patterns formed by applying dmt on testing set.
5. Use the patterns found to predict likely buyers among the current non-buyers
6. Promote to the likely buyers(called rollout).
In the second situation, a brand new product is to be promoted to the customers in the data base, so none
of them are buyers. In this case, pilot study is conducted, in which a small portion(say 5%) of the customers
is choosen randomly as the target of promotion. Again, X% of the customers in the pilot group may
respond to the promotion. Then data mining is performed in the pilot group to find the likely buyers in the
whole database.
Specific problems encountered while data mining on data sets for direct marketing are
1. Imbalance class distribution: Because only a small amount of buyers are likely means positive but most
of the algorithms can work on this type of sets. they assume that 100% are unlikely. Many data mining and
machine learning researchers have recognized and studied this problem in recent years(Farwett &
Provost,19s96;Kubat, Holte, & Matwin; Lewis & Catleltl; Pazzani, Merz, Murphy, Ali, Hume & Brunk).
2. Predictive accuracy cannot be used as a suitable evaluation criterion for the data mining process.
Classifying can be difficult. Means considering likely buyers as non-buyers and non-buyers as buyers
should be avoided.
4.SOLUTIONS
Ranking of non-buyers makes it possible to choose any number of likely buyers for the promotion.It also
provides a fine distinction among chosen customers to apply different means of promotion.
Lift analysis has been widely used in database marketing previously(Hughes,1996).A lift reflects the
redistribution of responders in the testing set after the testing examples are ranked.
5. CONCLUSION
Direct marketing is widely used in the fields of marketing like telemarketing,direct mail marketing,email
marketing etc.,data mining is applied on this marketing strategy to avoid human flaws in classifying the
customers based on their loyalty.We discussed the problems one faces in applying the datamining for direct
marketing and discussed their solutions.
6. ACKNOWLEDGEMENTS
This work was supported by Mrs. T Anuradha Assoc.professor and Mr. M Suman Assoc. Professor
,Department of Electronics and Computer Science Engineering, KLUniversity.
7. REFERENCES
[1] J.R. Quinlan, Morgan Kaufmann, C4.5 Programs for Machine Learning. 1993.
3](https://image.slidesharecdn.com/11-directmarketingwiththeapplicationofdatamining-120513002009-phpapp01/85/11-direct-marketing-with-the-application-of-data-mining-3-320.jpg)
![Journal of Information Engineering and Applications www.iiste.org
ISSN 2224-5758 (print) ISSN 2224-896X (online)
Vol 1, No.6, 2011
[2] A.Berson, K. Thearling, and S.J. Smith, Building Data Mining Applications for CRM. McGraw-Hill.
1999.
[3] G.K.Gupta, Introduction to data mining with case study ,Prentice Hall of India.2006.
[4] Mehmed Kantardzic, (2003), Data Mining: Concepts, Models,Methods, and Algorithms, John Wiley &
Sons.
[5] Farwett & Provost,19s96;Kubat, Holte, & Matwin;Lewis & Catleltl;Pazzani,Merz,Murphy,Ali, Hume &
Brunk.
[6]data-mine.com/white_papers/direct_marketing.
[7] The Application of Data Mining For Direct
Marketing Purushottam R Patil, Pravin Revankar, Prashant Joshi
Second International Conference on Emerging Trends in Engineering and Technology, ICETET-09.
4](https://image.slidesharecdn.com/11-directmarketingwiththeapplicationofdatamining-120513002009-phpapp01/85/11-direct-marketing-with-the-application-of-data-mining-4-320.jpg)
More Related Content
What's hot (18)
Similar to 11.direct marketing with the application of data mining (20)
More from Alexander Decker (20)
Recently uploaded (20)
11.direct marketing with the application of data mining
- 1. Journal of Information Engineering and Applications www.iiste.org ISSN 2224-5758 (print) ISSN 2224-896X (online) Vol 1, No.6, 2011 Direct Marketing with the Application of Data Mining M Suman, T Anuradha, K Manasa Veena* KLUNIVERSITY,GreenFields,vijayawada,A.P *Email: [email protected] Abstract For any business to be successful it must find a perfect way to approach its customers. Marketing plays a huge role in this. Mass marketing and direct marketing are the two types of it. Mass marketing targets everybody in the society and thus it has less impact on valued customers where direct marketing concentrates mainly on these valued and un loyal customers and promotes only to them which in turn makes profits. For this separation of customers based on their loyalty data mining algorithms and tools are used. In this paper we discussed the approach of implementation of data mining for direct marketing. We mainly concentrated and studied on why we apply data mining for direct marketing, how we apply and problems one faces while applying data mining concept for direct marketing and the solutions for them in direct marketing. Keywords: Direct marketing, Data mining,Decision tree. 1. INTRODUCTION In marketing a product ,there are several forms of sub disciplines, and one of them is direct marketing which involves messages sent directly to consumers usually through email, telemarketing and direct mail.As the traditional forms of advertising (radio, newspapers, television,etc.) may not be the best use of their promotional budgets, many companies or service providers with a specific market use this method of marketing. 1.1 Where we use direct marketing For example, a company which sells a hair loss prevention product or life insurance policy would have to find a radio station whose format appealed to older male listeners who might be experiencing this problem. There would be no guarantee that this group would be listening to that particular station at the exact time the company's ads were broadcast. Money spent on a radio spot (or television commercial or newspaper ad) may or may not reach the type of consumer who would be interested in a hair restoring product. This is where direct marketing becomes very appealing. Instead of investing in a scattershot means of advertising, companies with a specific type of potential customer can send out literature directly to a list of pre-screened individuals. Direct marketing firms may also keep addresses of those who match a certain age group or income level or special interest. Manufacturers of a new dog shampoo might benefit from having the phone numbers and mailing addresses of pet store owners or dog show participants. Direct marketing works best when the recipients accept the fact that their personal information might be used for this purpose. Some customers prefer to receive targeted catalogues which offer more variety than a general mailing. 2. DATA MINING Data mining is the process of extracting hidden patterns from data. It is commonly used in a wide range of profiling practices, such as marketing, surveillance, fraud detection and scientific discovery. Generally, data mining (sometimes called data or knowledge discovery) is the process of analyzing data from different perspectives and summarizing it into useful information - information that can be used to increase revenue, cuts costs, or both. Data mining software is one of a number of analytical tools for analyzing data. It allows users to analyze data from many different dimensions or angles, categorize it, and summarize the relationships identified. Technically, data mining is the process of finding correlations or patterns among dozens of fields in large relational databases. 2.1. Data mining in direct marketing 1
- 2. Journal of Information Engineering and Applications www.iiste.org ISSN 2224-5758 (print) ISSN 2224-896X (online) Vol 1, No.6, 2011 As we already discussed, in Direct marketing, concentrates on a particular group of customers( not loyal and beneficial).So, the data mining technique called the Supervised Classification is used to classify the customers for marketing. 2.2 The Process DMT will extract customer data, append it with extensive demographic, financial and lifestyle information, then identify hidden, profitable market segments that are highly responsive to promotions. 2.3 Decision tree A Decision tree is a popular classification technique that results in flowchart like tree structure where each node denotes test on a attribute value and each branch represents an outcome of test. The leaves represent classes. Using Training data Decision tree generate a tree that consists of nodes that are rules and each leaf node represents a classification or decision. The data usually plays important role in determining the quality of the decision tree. If there are number of classes, then there should be sufficient training data available that belongs to each of the classes. Decision trees are predictive models, used to graphically organize information about possible options, consequences and end value. They are used in computing for calculating probabilities. Example- CUSTOMER DATA LOYAL UNLOYAL ACTION A ACTION B Fig 1:A decicion tree based on customer’s loyalty 2.4 Building A Decision Tree In Direct Marketing Decision-tree learning algorithms, such as ID3 or C4.5 are among the most powerful and popular predictive methods for classification. So here in direct marketing we classify the customers on basis of their attributes like sex, age, location, purchase history, feedback details etc. 2.5 Algorithm C4.5 Builds decision trees from set of training data using the concept of Information entropy. The training data is a set S = s1, s2... of already classified samples. Each sample si = x1, x2... is a vector where x1, x2... represent attributes or features of the sample. The training data is augmented with a vector C = c1, c2... where c1, c2... represent the class to which each sample belongs. At each node of the tree, C4.5 chooses one attribute of the data that most effectively splits its set of samples into subsets enriched in one class or the other. Its criterion is the normalized information gain (difference in entropy) that results from choosing an attribute for splitting the data. The attribute with the highest normalized information gain is chosen to make the decision. The C4.5 algorithm then recurses on the smaller sublists. In general, steps in C4.5 algorithm to build decision tree are: 1. Choose attribute for root node 2. Create branch for each value of that attribute 3. Split cases according to branches 4. Repeat process for each branch until all cases in the branch have the same class. 2
- 3. Journal of Information Engineering and Applications www.iiste.org ISSN 2224-5758 (print) ISSN 2224-896X (online) Vol 1, No.6, 2011 3. PROBLEMS IN CLASSIFICATION According to Charles X.Ling and Chenghui Li,the classification of data base involves the following situations: In the first situation, some (say X%) of the customers in the database have already bought the product, through previous mass marketing or passive promotion. X is usually rather small, typically around 1.Data mining can be used to discover patterns of buyers, in order to single out likely buyers from the current non- buyers,(100-X%)of all customers. More specifically, data mining for direct marketing in the first situation can be discovered in: 1. Get the database of all customers, among which X% are buyers. 2. Data mining on the data set based on Geo-demographic information, transforming address and area codes, deal with missing values, etc. 3. Applying algorithm to prepare objects, classes . 4. Evaluate the patterns formed by applying dmt on testing set. 5. Use the patterns found to predict likely buyers among the current non-buyers 6. Promote to the likely buyers(called rollout). In the second situation, a brand new product is to be promoted to the customers in the data base, so none of them are buyers. In this case, pilot study is conducted, in which a small portion(say 5%) of the customers is choosen randomly as the target of promotion. Again, X% of the customers in the pilot group may respond to the promotion. Then data mining is performed in the pilot group to find the likely buyers in the whole database. Specific problems encountered while data mining on data sets for direct marketing are 1. Imbalance class distribution: Because only a small amount of buyers are likely means positive but most of the algorithms can work on this type of sets. they assume that 100% are unlikely. Many data mining and machine learning researchers have recognized and studied this problem in recent years(Farwett & Provost,19s96;Kubat, Holte, & Matwin; Lewis & Catleltl; Pazzani, Merz, Murphy, Ali, Hume & Brunk). 2. Predictive accuracy cannot be used as a suitable evaluation criterion for the data mining process. Classifying can be difficult. Means considering likely buyers as non-buyers and non-buyers as buyers should be avoided. 4.SOLUTIONS Ranking of non-buyers makes it possible to choose any number of likely buyers for the promotion.It also provides a fine distinction among chosen customers to apply different means of promotion. Lift analysis has been widely used in database marketing previously(Hughes,1996).A lift reflects the redistribution of responders in the testing set after the testing examples are ranked. 5. CONCLUSION Direct marketing is widely used in the fields of marketing like telemarketing,direct mail marketing,email marketing etc.,data mining is applied on this marketing strategy to avoid human flaws in classifying the customers based on their loyalty.We discussed the problems one faces in applying the datamining for direct marketing and discussed their solutions. 6. ACKNOWLEDGEMENTS This work was supported by Mrs. T Anuradha Assoc.professor and Mr. M Suman Assoc. Professor ,Department of Electronics and Computer Science Engineering, KLUniversity. 7. REFERENCES [1] J.R. Quinlan, Morgan Kaufmann, C4.5 Programs for Machine Learning. 1993. 3
- 4. Journal of Information Engineering and Applications www.iiste.org ISSN 2224-5758 (print) ISSN 2224-896X (online) Vol 1, No.6, 2011 [2] A.Berson, K. Thearling, and S.J. Smith, Building Data Mining Applications for CRM. McGraw-Hill. 1999. [3] G.K.Gupta, Introduction to data mining with case study ,Prentice Hall of India.2006. [4] Mehmed Kantardzic, (2003), Data Mining: Concepts, Models,Methods, and Algorithms, John Wiley & Sons. [5] Farwett & Provost,19s96;Kubat, Holte, & Matwin;Lewis & Catleltl;Pazzani,Merz,Murphy,Ali, Hume & Brunk. [6]data-mine.com/white_papers/direct_marketing. [7] The Application of Data Mining For Direct Marketing Purushottam R Patil, Pravin Revankar, Prashant Joshi Second International Conference on Emerging Trends in Engineering and Technology, ICETET-09. 4
- 5. International Journals Call for Paper The IISTE, a U.S. publisher, is currently hosting the academic journals listed below. The peer review process of the following journals usually takes LESS THAN 14 business days and IISTE usually publishes a qualified article within 30 days. Authors should send their full paper to the following email address. More information can be found in the IISTE website : www.iiste.org Business, Economics, Finance and Management PAPER SUBMISSION EMAIL European Journal of Business and Management [email protected] Research Journal of Finance and Accounting [email protected] Journal of Economics and Sustainable Development [email protected] Information and Knowledge Management [email protected] Developing Country Studies [email protected] Industrial Engineering Letters [email protected] Physical Sciences, Mathematics and Chemistry PAPER SUBMISSION EMAIL Journal of Natural Sciences Research [email protected] Chemistry and Materials Research [email protected] Mathematical Theory and Modeling [email protected] Advances in Physics Theories and Applications [email protected] Chemical and Process Engineering Research [email protected] Engineering, Technology and Systems PAPER SUBMISSION EMAIL Computer Engineering and Intelligent Systems [email protected] Innovative Systems Design and Engineering [email protected] Journal of Energy Technologies and Policy [email protected] Information and Knowledge Management [email protected] Control Theory and Informatics [email protected] Journal of Information Engineering and Applications [email protected] Industrial Engineering Letters [email protected] Network and Complex Systems [email protected] Environment, Civil, Materials Sciences PAPER SUBMISSION EMAIL Journal of Environment and Earth Science [email protected] Civil and Environmental Research [email protected] Journal of Natural Sciences Research [email protected] Civil and Environmental Research [email protected] Life Science, Food and Medical Sciences PAPER SUBMISSION EMAIL Journal of Natural Sciences Research [email protected] Journal of Biology, Agriculture and Healthcare [email protected] Food Science and Quality Management [email protected] Chemistry and Materials Research [email protected] Education, and other Social Sciences PAPER SUBMISSION EMAIL Journal of Education and Practice [email protected] Journal of Law, Policy and Globalization [email protected] Global knowledge sharing: New Media and Mass Communication [email protected] EBSCO, Index Copernicus, Ulrich's Journal of Energy Technologies and Policy [email protected] Periodicals Directory, JournalTOCS, PKP Historical Research Letter [email protected] Open Archives Harvester, Bielefeld Academic Search Engine, Elektronische Public Policy and Administration Research [email protected] Zeitschriftenbibliothek EZB, Open J-Gate, International Affairs and Global Strategy [email protected] OCLC WorldCat, Universe Digtial Library , Research on Humanities and Social Sciences [email protected] NewJour, Google Scholar. Developing Country Studies [email protected] IISTE is member of CrossRef. All journals Arts and Design Studies [email protected] have high IC Impact Factor Values (ICV).