THE EFFECTIVENESS OF DATA MINING TECHNIQUES IN BANKING

Computer Applications: An International Journal (CAIJ), Vol.4, No.1/2/3/4, November 2017
DOI:10.5121/caij.2017.4401 1
THE EFFECTIVENESS OF DATA MINING
TECHNIQUES IN BANKING
Yuvika Priyadarshini
Researcher, Jharkhand Rai University, Ranchi.
ABSTRACT
The aim of this study is to identify the extent of Data mining activities that are practiced by banks, Data
mining is the ability to link structured and unstructured information with the changing rules by which
people apply it. It is not a technology, but a solution that applies information technologies. Currently
several industries including like banking, finance, retail, insurance, publicity, database marketing, sales
predict, etc are Data Mining tools for Customer . Leading banks are using Data Mining tools for customer
segmentation and benefit, credit scoring and approval, predicting payment lapse, marketing, detecting
illegal transactions, etc. The Banking is realizing that it is possible to gain competitive advantage deploy
data mining. This article provides the effectiveness of Data mining technique in organized Banking. It also
discusses standard tasks involved in data mining; evaluate various data mining applications in different
sectors
KEYWORDS
Definition of Data Mining and its task, Effectiveness of Data Mining Technique, Application of Data
Mining in Banking, Global Banking Industry Trends, Effective Data Mining Component and Capabilities,
Data Mining Strategy, Benefit of Data Mining Program in Banking
1. DEFINITION OF DATA MINING AND ITS TASK
Data mining is an essential step in the knowledge discovery in databases (KDD) process that
produces useful patterns. The terms of KDD and data mining are different. KDD refers to the
overall process of discovering useful knowledge from data. Data mining refers to discover new
patterns from a wealth of data in databases to extract useful knowledge. KDD process consists of
iterative sequence methods as follows:
1. Selection: Selecting data relevant to the analysis task from the database
2. Preprocessing: Removing noise and inconsistent data; combining multiple data sources
3. Transformation: Transforming data into appropriate forms to perform data mining
4. Data mining: Choosing a data mining algorithm which is appropriate to pattern in the
data; Extracting data patterns
5. Interpretation/Evaluation : Interpreting the patterns into knowledge by removing
redundant or irrelevant patterns; Translating the useful patterns into terms that human
understandable

2
Data mining has two primary objectives of prediction and description. Prediction involves using
some variables in data sets in order to predict unknown values of other relevant variables
(e.g.classification, regression, and anomaly detection).
Data mining task define six main functions of Data mining
1. Classification is finding models that analyze and classify a data item into several
predefined classes
2. Regression is mapping a data item to a real-valued prediction variable
3. Clustering is identifying a finite set of categories or clusters to describe the data
4. Dependency Modeling (Association Rule Learning) is finding a model which describes
significant dependencies between variables
5. Deviation Detection (Anomaly Detection) is discovering the most significant changes in
the data
6. Summarization is finding a compact description for a subset of data
2. EFFECTIVENESS OF DATA MINING TECHNIQUE
Data mining is the process of analyzing data from different perspectives and summarizing it into
useful information. DM techniques are the result of a long process of research and product
development.
Data mining consists of five major elements; to extract, to transform, and to load transaction data
onto the data warehouse system, to store and manage the data in a multidimensional database
system, to provide data access to business analysts and in format Analyze the data by application
software, and finally to present the data in a useful format, such as a graph or table
DM techniques usually fall into two categories, predictive or descriptive. Predictive DM uses
historical data to infer something about future events. Predictive mining tasks use data to build a
model to make predictions on unseen future events. Descriptive DM aims to find patterns in the
data that provide some information about internal hidden relationships.
Descriptive mining tasks characterize the general properties of the data and represent it in a
meaningful way.
Data sources of a data mining system can be divergent information repositories like database,
data warehouse or other repository. Data mining engine is the core of data mining system. The
functional modules od Data mining algorithms and rules are kept in the engine. Database stores
the knowledge that is used to guide the data mining process and provides the data that pattern
evaluation module needs to validate the result of data mining.
3. APPLICATION OF DATA MINING IN BANKING
Data Mining can help by contributing in solving business problems by finding patterns,
associations and correlations which are hidden in the business information stored in the data bases
The banks who have realized the importance of data mining are in the process of reaping huge

3
profits and considerable competitive advantage. According to the regulations given by Reserve
Bank of India, the banks have to Provide Off-site Monitoring Surveillance (OSMOS) reports on
regular basis in electronic format only and Regulatory requirement of filing of statutory returns
such as the one under Section 42 of the Reserve Bank of India Act, 1934 for working out Cash
Reserve Ratio (CRR) and Statutory Liquidity Ratio (SLR) obligations in electronic format.
According to the Committee formed by Reserve Bank of India Headed by Dr. A. Vasudevan to
go through the details of this topic, gave his report on 17th July,1999, the committee highlighted
that by the use of data mining techniques, data available at various computer systems can be
accessed and by a combination of techniques like classification, clustering, segmentation,
association rules, sequencing, decision tree various ALM reports such as Statement of Structural
Liquidity, Statement of Interest Rate Sensitivity etc. or accounting reports like Balance Sheet and
Profit & Loss Account can be generated instantaneously for any desired period/ date . Trends can
be analyzed and predicted with the availability of historical data and the data warehouse assures
that everyone is using the same data at the same level of extraction, which eliminates conflicting
analytical results and arguments over the source and quality of data used for analysis. In short,
data warehouse enables information processing to be done in a credible, efficient manner. The
Committee recognizes the need for data warehouses and data mining both at the individual bank
level and at industry level . The implication of adopting such technology in a bank would be as
under :
• All transactions captured at the branch level would get consolidated at a central location. Such a
central location could be called the Data Warehouse of the concerned bank. For this to happen,
one of the requirements would be to establish connectivity between the branches on the one hand
and the Data Warehouse platform on the other.
• For banks with large number of branches, it may not be desirable to consolidate the transaction
details at one place only. It can be decentralized
• By way of data mining techniques, data available at various computer systems can be accessed
and by a combination.
4. GLOBAL BANKING INDUSTRY TRENDS
The ongoing global financial crisis, with its historic dimensions, will have a lasting impact on the
global banking industry and the world economy. Banks are looking for growth opportunities2, but
their success is very much dependent on their ability to build critical mass and successful
operations in these economic times.
The regulatory landscape has strengthened significantly, with governments in many markets
implementing much more stringent rules—such as minimum capital requirements—putting
pressure on firms to raise capital.
This financial pressure has increased focus on operational efficiencies and is driving investments
in automating credit underwriting platforms with scorecards and models, fraud and collection
analytics, and enhancing data and risk analytics capabilities. Some banks are progressing from
using qualitative or analytical models to the deployment of predictive models for risk and
customer analytics.

4
Amid the increased requirements on capital adequacy and optimal utilization of capital, economic
capital management and risk adjusted return on capital (RAROC) have become a top priority for
banks. Loss forecasting and stress testing have gained increased importance in the current
economic scenario and are the norm today.Other important trends in the global banking industry
are:
■ Business intelligence practices are being integrated with customer relation management
(CRM) and there is a stronger focus on predictive and behavioral modeling tools. This
combination is fast becoming the central cog for cross selling, customer lifetime value
management, risk management and recovery management.
■ Banks have started to realize the potential of social media analytics and content tracking and
are aligning their CRM tools with Twitter and Facebook.
■ With the growth of fraudsters and hackers, security threats for all firms but especially banks
have mushroomed. Among the big drivers of payments security are new encryption initiatives and
efforts to bring interoperability to point-to-point transaction encryption.3
■ With the emergence of mobile banking, banks have to restructure their business model and
increase collaboration with firms in cards, payments and telecom domains.
■ Banks will be ramping up their personal finance management (PFMs) offerings to help
customers realize their financial goals and to avoid disintermediation
5. EFFECTIVE DATA MINING COMPONENT AND CAPABILITIES
A comprehensive EDM program comprises of a host of capabilities. In order to enable these
capabilities, there are a few components that first need to be in place. The pre-requisite
components and key capabilities are listed below.
5.1 PRE-REQUISITE COMPONENTS
■ DATA MANAGEMENT VISION – an organization needs to describe the vision and principles or
core values around which its enterprise data management program is based.
■ DATA MANAGEMENT GOALS – goals of an EDM program need to be related to strategic
business goals, objectives and priorities. These, furthermore, need to be adopted by and
communicated to key stakeholders.
■ GOVERNANCE MODEL – An EDM program needs to adopt an enterprise-wide mechanism by
which the program is managed, funded and implemented.
■ ISSUES MANAGEMENT AND RESOLUTION – the organization has the ability to identify, triage,
track, and update status for all data and integration issues identified during “business as usual”
(BAU) activities or ongoing data management initiatives.
■ MONITORING AND CONTROL – Collective capabilities for measuring and reporting on the
quality and effectiveness of the data management program as it operates as part of the BAU
environment.

5
5.2 DATA MINING CAPABILITIES
■ CRITICAL DATA INVENTORY – Critical Data consists of those data elements (and their
business definitions) that the business deems as important for decision making and compliance.
This inventory should be made in consultation with business users. The Critical Data Inventory
helps prioritize or contain the scope of an EDM program.
■ DATA INTEGRATION – This covers the processes and tools for acquisition, composition and
enrichment of data from different sources into a single unified store or view. Data integration
typically is done by building an enterprise data warehouse, from which data is sourced directly
into analytical engines, or into data marts that feed the analytical engines. Data integration also
addresses the control processes that are used to monitor data integrity as data flows from data
producers to data consumers.
■ DATA PROFILING – Data profiling is the examination of data to collect statistics and
characteristics about the structure of available data. It is used to assist in critical data assessment,
data classification, data integration and impact analysis.
■ DATA QUALITY – Data quality measures whether data is ‘fit for intended use’. Data quality is
typically measured along the dimensions of accuracy, completeness, conformity, consistency,
duplication and integrity, with each dimension carrying different weight based on the intended
use of the data. End-to-end data quality allows for comparison of data quality across the data flow
at a point in time as well as across time (trends).
■ METADATA MANAGEMENT – Metadata is information about the data itself. Metadata captures
attributes of data like the type, length, timestamp, source, owner etc., as well as relationships in
data (semantics), and helps with data traceability and lineage. Use of uniform methods and tools
for defining, collecting, and managing information metadata ensures that data is identified
consistently across the enterprise.
■ MASTER DATA MANAGEMENT – Master data or Master file is the single, authoritative and
agreed upon source of data that is critical for business operation. It typically includes persistent
non-transactional data like customer, product, employee etc. Master data management ensures
that there is a single consistent version of critical data used across the enterprise.
■ REFERENCE DATA MANAGEMENT – Reference data is used to classify or categorize data. An
example is the product master which contains the list of all products along with their attributes.
As with metadata and master data, reference data management also plays an important role in
data integrity and consistency.
■ DATA PRIVACY (ANONYMIZATION) – This includes processes, algorithms and technology
platforms which are required to ensure that the contents of any information object (data set) fully
comply with information privacy and protection laws and regulations.
6. DATA MINING STRATEGY
A top-down strategic approach to EDM aligns business priorities to specific EDM components
and capabilities. The approach should also evaluate current state of each relevant capability
against the desired future state. An assessment based on a data management maturity model is a

6
good starting point for an EDM strategy and roadmap definition initiative. A data management
maturity model assesses the above-mentioned capabilities with respect to the readiness of the firm
from a people, policies, technology and adoption perspective. This helps the firm identify gaps
and prioritize initiatives along a well designed road-map to achieve the future state.
6.1 BENEFITS OF DM PROGRAM IN BANKING
The benefits of anDM program to banking institutions are best analyzed from the lenses of the
business capabilities that it supports and enables. Each business function potentially has its own
unique needs around data and hence the benefits correspond to those unique perspectives. A few
illustrative benefits are mentioned below.
■ FOR OPERATIONS, a centralized reference data management system will offer great advantages
in providing accurate, timely and consistent data across systems. This will result in a huge
reduction in reconciliation activities and will increase the efficiency and effectiveness of various
teams.
■ FOR RISK MANAGEMENT, EDM offers among other things the ability to correctly identify
counterparty risk. Accurate measurement and management of enterprise wide risk measurement
and management would be virtually impossible without accurate, reliable and consistent data
provided by an effective EDM.
■ BENEFITS TO FINANCE AND ACCOUNTING from EDM are obvious considering the
performance analysis and management reports they produce that are viewed by external
stakeholders (regulatory and market) and internal consumers (board, senior management and
decision makers). EDM can allow these reports to be certified with a greater degree of
confidence.
■ DATA INTEGRITY AND CONSISTENCY, which allow for greater confidence in the management
reports and decisions, are of great importance from an audit, legal and compliance perspective as
well.
■ SALES AND MARKETING OPERATIONS are immensely benefited from an EDM through the
ability to have a single view of customer that enables effective cross-selling andup-selling.
7. CONCLUSION
The Effective Data Mining has become more important than ever before. A bank or financial
institution embarking on an initiative to launch or revitalize its EDM program needs to keep in
mind the following important aspects:
■ An efficient Effective Data Mining program should be designed to be in tune with the
organization’s own specific and unique business needs.
■ It is necessary to design a program that brings together stakeholders from both business and
technology sides.
■ Technology solutions should be viewed as enablers of business capabilities and should be
driven by business needs.

7
■ To develop, sustain and mature an Effective Data Mining program, a comprehensive
framework including governance and control elements is needed.
■ It is important to maintain a balance between strategic long-term objectives and tactical quick
wins.
■ A successful Effective Data Mining program is one which builds strong foundations, and at the
same time allows for continuous evolution as business grows or transforms.
REFERENCES
[1] Dr. Madan Lal Bhasin, 2006. Data Mining: A Competitive Tool in the Banking and Retail Industries
[2] Berson, A., Smith, S., and Thearling, K. (1999). Building Data Mining Applications for CRM.
McGraw-Hill, New York.
[3] Ahmed, S. R. (2004). Applications of data mining in retail business. In In- formation Technology:
Coding and Computing, International Conference on, volume 2, page 455, Los Alamitos, CA, USA.
IEEE Computer Society.
[4] Shaw, M. (2001). Knowledge management and data mining for marketing. Decision Support
Systems, 31(1):127–137.
[5] Giraud-Carrier, C. and Povel, O. (2003). Characterising data mining software. Intell. Data Anal.,
7(3):181192.
[6] Burez, J. and Van den Poel, D. (2009). Handling class imbalance in customer churn prediction.
Expert Systems with Applications, 36:4626–4636.
[7] K. Chitra, B.Subashini, Customer Retention in Banking Sector using Predictive Data Mining
Technique, International Conference on Information Technology, Alzaytoonah University, Amman,
Jordan, www.zuj.edu.jo/conferences/icit11/paperlist/Papers/
[8] K. Chitra, B.Subashini, Automatic Credit Approval using Classification Method, International Journal
of Scientific & Engineering Research (IJSER), Volume 4, Issue 7, July-2013 2027 ISSN 2229-5518.
[9] K. Chitra, B.Subashini, Fraud Detection in the Banking Sector, Proceedings of National Level
Seminar on Globalization and its Emerging Trends, December 2012.
[10] K. Chitra, B.Subashini, An Efficient Algorithm for Detecting Credit Card Frauds, Proceedings of
State Level Seminar on Emerging Trends in Banking Industry, March 2013.
[11] Lin, T. Y. (1994), “Anamoly Detection -- A Soft Computing Approach”, Proceedings in the ACM
SIGSAC New Security Paradigm Workshop, Aug 3-5, 1994,44-53. This paper reappeared in the
Proceedings of 1994 National Computer Security Center Conference under the title “Fuzzy Patterns
in data.
[12] Scott W. Ambler (2001) “Challenges with legacy data: Knowing your data enemy is the first step in
overcoming it”, Practice Leader, Agile Development, Rational Methods Group, IBM, 01 Jul 2001.
[13] Agrawal, R, and R. Srikant, “Privacy-preserving Data Mining,” Proceedings of the ACM SIGMOD
Conference, Dallas, TX, May 2000.

8
[14] Clifton, C., M. Kantarcioglu and J. Vaidya, “Defining Privacy for Data Mining,” Purdue University,
2002 (see also Next Generation Data Mining Workshop, Baltimore, MD, November 2002).
[15] Gartner. Evolution of data mining, Gartner Group Advanced Technologies and Applications Research
Note, 2/1/95.
[16] International Conferences on Knowledge Discovery in Databases and Data Mining (KDD’95-98),
1995-1998.
[17] R.J. Miller and Y. Yang. Association rules over interval data. SIGMOD'97, 452-461, Tucson,
Arizona, 1997.
[18] Zaki, M.J., SPADE An Efficient Algorithm for Mining Frequent Sequences Machine Learning, 42(1)
31-60, 200

THE EFFECTIVENESS OF DATA MINING TECHNIQUES IN BANKING

More Related Content

What's hot (20)

Similar to THE EFFECTIVENESS OF DATA MINING TECHNIQUES IN BANKING (20)

Recently uploaded (20)

THE EFFECTIVENESS OF DATA MINING TECHNIQUES IN BANKING