Download an analysis on predictive analytics and data mining in

International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,
Volume-4, Issue-2, Feb.-2016
AN ANALYSIS ON PREDICTIVE ANALYTICS AND DATA MINING IN
TELECOM &AUTOMOBILE IN MAHARASHTRA
1
SHAILESH A KAHALEKAR, 2DR. SUNIL PATIL
1
Ph.D Scholer, Pacific University, Udaipur
Principal, AG Patil Engg and Technology, solapur
2
Abstract- It is indeed required to establish relations on the various factors that are affecting the areas of business intelligence
widely in Maharashtra, mostly being used in the IT and Telecom and in the automobile industry all over maharshtra.
Data mining aids predictive analysis by providing a
record of the past that can be analyzed and used to
predict which customers are most likely to renew,
purchase, or purchase related products and services.
Business intelligence data mining is important to your
marketing campaigns. Proper data mining algorithms
and predictive modeling can narrow your target
audience and allow you to tailor your ads to each
online customer as he or she navigates your site. Your
marketing team will have the opportunity to develop
multiple advertisements based on the past clicks of
your visitors.
I. INTRODUCTION
Prediction in BI and BW is now a need of hour , in
todays telecom and automobile scenario - such
factors affecting the BI related with Engg and IT
related questions are considered at the heart in a case
of the data centre envoirnment in IT companies and
BSNL in particular where BI systems are
implemented as an ETL tool for reporting and ERP is
implemented Automobile companies such as
Endurance systems limited is considered where ERP
and BI both are implemented has been surveyed and
certain conclusions are drawn .
When used together, predictive analytics and data
mining can make marketing more efficient. There are
many techniques and methods, including business
intelligence data collection.
Predictive analytics can aid in choosing marketing
methods, and marketing more efficiently. By only
targeting customers who are likely to respond
positively, and targeting them with a combination of
goods and services they are likely to enjoy, marketing
methods become more efficient. In the best cases,
predictive analytics can reduce the amount of dollars
spent to close a sale.
What is Business Intelligence Data?
Business intelligence is a decision support system
where information is gathered for the purpose of
predictive analysis and support for business
decisions. Prior to the widespread availability of data
marts and reporting software, business intelligence
data was gathered manually. Collecting information
across corporate departments such as finance, sales
and production, and correlating it into meaningful
presentations created further time delays.
Current availability of business intelligence data in
computer-readable form, both within a company and
from online sources, make incorporation of business
intelligence data into business operations more
dynamic and bring it closer to real time. Instead of
having to wait a week, or even a month for data,
managers are now able to mine data and perform
predictive analysis from multiple sources daily.
At its most effective, business intelligence data
mining can help marketing professionals anticipate
and prepare for customer needs, rather than just
reacting to them. And data mining can present data on
demographics which may have been previously
overlooked. For example, have your loyal customers
gotten older or younger? Are they now shopping for
maternity clothing, instead of clothing to wear to a
club? Are they more hip or environmentally aware?
Any combination of those changes in your customer
demographics could be useful in determining what
newspaper or magazine is the best venue for your
print campaign and what type of campaign it should
be. When applied to marketing strategy, predictive
analytics and data mining can help managers to bring
in more sales, while spending less on campaigns.
What Are Predictive Analytics?
Predictive analytics is using business intelligence data
for forecasting and modeling. It is a way to use
predictive analysis data to predict future patterns. It is
used widely in the insurance, medical and credit
industries. Assessment of credit, and assignment of a
credit score is probably the most widely known use of
predictive analytics. Using events of the past,
managers are able to estimate the likelihood of future
events.
II. CASE STUDY OF IBM FOR DATA
ANALYSIS WITH SPSS
In today’s data-driven landscape, the ability to
analyze information to drive decision-making and
solve problems is fundamental for success. IBM
SPSS Statistics is commonly used tool by
statisticians/analytics professionals for ad hoc
An Analysis On Predictive Analytics And Data Mining In Telecom &Automobile In Maharashtra
88
International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,
analysis and hypothesis testing. Rich in Extensive
data preparation, analysis and visualization
features, IBM SPSS Statistics provides insight into a
sample of data and tools for prediction and
forecasting. Generally speaking, it helps in:
This is because IBM SPSS Modeler adds exceptional
predictive analytics capabilities, along with a range of
invaluable procedures for cleansing and preparing
data. Business intelligence software can discover a
great deal of information – painting a detailed picture,
for example, of which customers are buying what
products, and where and when they’re doing so.
Predictive analytics adds foresight to the insights
gained through business intelligence. In this way,
companies can anticipate future purchasing patterns
and make attractive offers at just the right times and
places. Business Analytics IBM Software IBM SPSS
Modeler and IBM Cognos Business Intelligence
Reliably predicting outcomes While BI solutions
excel at revealing how we are doing, and why,
predictive analytics provides a window into the future
by predicting what is likely to happen next. Predictive
analytics connects data to effective action by drawing
reliable conclusions about current conditions and
future events. IBM SPSS Modeler offers techniques
for clustering and segmentation, to reveal groups
within the data; techniques for association, to reveal
connections and sequences of events; and techniques
for predicting outcomes. An association technique is
often used to look at all supermarket purchases and
identify what products are purchased together. This
technique, called “market basket analysis,” could
reveal that there are many different “baskets” bought
by different groups of customers.
 Understanding how to leverage the data collected
and being able to fill in the gaps
 Understanding the underlying relationships in
data from the data warehouse or various
repositories and how they impact decisions
 Recognizing what influences business outcomes
and to what degree by recognizing what variables
drive outcomes
 Testing assumptions before applying the theory
to real world events. Also when to adjust or
abandon these assumptions
IBM SPSS Statistics offers the full scope of statistical
and analytical capabilities that organizations require.
It’s an easy-to-use, comprehensive software solution
that:
1.
2.
3.
4.
5.
6.
Volume-4, Issue-2, Feb.-2016
Addresses the entire analytical process from
planning and data preparation to analysis,
reporting and deployment
Allows application of statistical information to
data to improve understanding and decision
quality throughout the enterprise.
Provides advanced analysis using statistical
algorithms (for regression, correlation,
segmentation etc.), and display of those results.
Enterprise scale and extensibility as part of the
SPSS suite of products.
Works with all common data types, external
programming languages, operating systems
and file types
Use with Cognos BI for statistical information
in reports, dashboards and scorecards.
IBM SPSS Modeler can go beyond this simple
technique to examine the customers who buy
particular baskets and identify the unique
characteristics that define their preferences. Then,
using these characteristics as a model, it can predict
which products they are likely to buy next. For
example, using IBM Cognos Business Intelligence
scorecards and dashboards, a supermarket product
manager might discover that sales of a certain
premium brand of coffee are declining. The analysis
and reporting features might show that customers
seem to prefer to buy other, lower-margin coffee
brands. The question, then, is what to do to increase
sales of the high-margin product. The answer is to use
IBM SPSS Modeler to perform a market basket
analysis. This could reveal that a group of customers
who frequently buy the premium coffee brand also
buys, say, fresh meat, certain varieties of fruit and
vegetables and “artisanal” bread. On the basis of this
predictive intelligence, a product manager could
reliably conclude that premium coffee sales would
increase by making a special offer – and strategically
placing display units near the produce, meat and
bread counters. 3 IBM SPSS Modeler and IBM
Cognos Business Intelligence The predictive
intelligence produced by IBM SPSS Modeler – for
example, information about the products people are
likely to buy next, and their propensity to purchase –
can be integrated directly into the IBM Cognos
Mining for intelligence One straightforward BI
application would be to analyze sales data to discover
which products are selling well (or badly) in which
geographies. Armed with this intelligence, sales
managers could take action to boost sales in the
affected geographies – for example, by conducting
further research to find out why some products are
not selling, or by deciding to concentrate on those
that are. Other applications include:
• Tracking shifts in the market for a product or
service
• Comparative analysis against a competitor’s
products/services
• Analysis of the market for a new product/service
• Financial analysis • Profitability analysis IBM
Cognos Business Intelligence is an immensely
powerful solution for guiding business decisions –
but when combined with the advanced functionality
of IBM SPSS Modeler it is even more powerful.
An Analysis On Predictive Analytics And Data Mining In Telecom &Automobile In Maharashtra
89
International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,
Business Intelligence framework and made available
to the reporting environment immediately.
Then product managers have a significantly more
comprehensive view into their business – including a
view into the future – all from their familiar, intuitive
IBM Cognos Business Intelligence dashboard. In this
way, marketers can create personalized, targeted
offers, and focus effort and expenditure.
Volume-4, Issue-2, Feb.-2016
REFERENCES
[1]
[2]
Extending the value of enterprise data IBM Cognos
Business Intelligence delivers a complete, consistent
and enterprise-wide view of information to support
decision-making. IBM SPSS Modeler can read that
enterprise view directly, preserving the structure and
the metadata, so that users are interacting with data in
the way that is already familiar to them. Once the
IBM SPSS Modeler analysis is complete, the results
can be written back to that same consistent view
using familiar tools and technologies. Because the
predictive intelligence is then a part of the underlying
data, managers can combine, view and manipulate
those results from within the familiar IBM Cognos
Business Intelligence interface to leverage the data
views that are driving the organization.
[3]
[4]
[5]
[6]
[7]
[8]
In case of the SAP environment data centers such as
Accenture, BSNL , MTNL and other IT companies in
Pune and in the vorrac,Tata motors , Endurance
system limited , a wide interview \ Of the factors
affecting Business Intelligence are found . Thefeed
backs are taken in the questioners , which includes
Engineering and IT related factors and it is found that
certain factors are found to be helping each other and
certain other are opposite .
[9]
[10]
[11]
[12]
SPSS analysis of factors affecting Engineering and IT
related factors are very important for decision making
purpose
The analysis shows that positive response from
various companies have been achieved and a
comparative model of these factors may be developed
in future .
[13]
[14]
CONCLUSION
[15]
A survey made for group of companies in
Maharashtra for predictive BI leads to certain
decisions in Engineering and IT related factors.
Buchda, S., Rulers for Business Intelligence and
Competitive Intelligence: An Overview and Evaluation of
Measurement Approaches. Journal of Competitive
Intelligence and Management, 2007. 4(2)
Yeoh, W. and A. Koronios, CRITICAL SUCES
FACTORS FOR BUSINES INTELIGENCE SYSTEMS.
Journal of Computer Information Systems, 2010. 50 (3): p.
23-32. Vol. I - ISSUE – III Page – 7 Worldwide
International Inter Disciplinary Research (A Peer
Reviewed) ISSN – 2454 – 790
Watson, H.J., Wixom, B. H., Hoffer, J. A.,
AndersonLehman, R., & Reynolds, A. M., Real-Time
Business Intelligence: Best practices at Continental
Airlines. Information Systems Management, 2006. 23(1):
p. 7– 18.
Jourdan, Z., R.K. Rainer, and T.E. Marshall, Business
Intelligence: An Analysis of the Literature. Information
Systems Management, 2008. 25(2): p. 121-131.
Gessner, G.H.V., L., Quick Response Improves Return on
Business Intelligence Investments. Information Systems
Management, 2005. 22(3): p. 66–74.
R.Carvalho, M.F., Using information technology to
support knowledge conversion processes. Information
Research, 2001. 07.
Choo, C.W., The Knowing Organization. 1998: Oxford:
Oxford University Press.
Wingyan Chung, H.C., Jay F. Nunamaker Jr. Business
Intelligence Explorer: A Knowledge Map Framework for
Discovering Business Intelligence on the Web. in 36th
Hawaii International Conference on System Sciences.
2002. Hawaii: Computer Society-IEEE.
L. Fuld, K.S., J. Carmichael, J. Kim, and K. Hynes,
Intelligence Software Report 2002. 2002, Fuld &
Company Inc: Cambridge, MA, USA.
Popovič, A.T., T. Jaklič, J., Conceptual model of business
value of business intelligence systems. Konceptualni
model poslovne vrijednosti sustava poslovne inteligencije,
2010. 15(1): p. 5-29.
Jorge García, R.A., Technology Evaluation Centers, 2011
business intelligence buyers guide:" bi for everyone".
Technology Evaluation Centers, 2011. 1: p. 5-15. [
Wu, J., Critical Success Factors for ERP System
Implementation. IFIP International Federation for
INTERNATIONAL JOURNAL OF SCIENTIFIC &
TECHNOLOGY RESEARCH VOLUME 2, ISSUE 2,
FEBRUARY 2013 ISSN 2277-8616, Amin Babazadeh
Sangar, Noorminshah Binti A.Iahad.
Volume 3, No. 3, May-June 2012 International Journal of
Advanced Research in Computer Science CASE STUDY
AND REPORT Available Online at www.ijarcs.info ©
2010, IJARCS All Rights Reserved 786 ISSN No. 09765697 Business Intelligence For Mobile Users,Dr
Prabhune,Dr S.A.Deshmukh
N. A. Patil Shreee Warana Sahakari Dudh Utpadak Sangh
Ltd., NCI2 TM: 2014 ISBN: 978-81-927230-0-6.

An Analysis On Predictive Analytics And Data Mining In Telecom &Automobile In Maharashtra
90