Orientation: Business Analytics

Business Analytics
Orientation
www.proschoolonline.com
Agenda
• What is Business Analytics?
• Some BA Examples
• BA Process
• Job Profiles
• Our Course
 Why?
 What?
 How?
• Evaluation
Why BA ?
• Credit ratings/targeted marketing:
 Given a database of 100,000 names, which persons are the least likely to default on their credit
cards?
 Identify likely responders to sales promotions
• Fraud detection
 Which types of transactions are likely to be fraudulent, given the demographics and transactional
history of a particular customer?
• Customer relationship management:
 Which of my customers are likely to be the most loyal, and which are most likely to leave for a
competitor? :
What is Business Analytics?
• Analytics is the use of:
 Data,
 Information technology,
 Statistical Analysis,
 Quantitative Methods,
 Mathematical or Computer-based models
 To help managers gain improved insight about their business operations and make better, fact-based
decisions.
What is Business Analytics?
• Process of semi-automatically analyzing large databases to find patterns that are:
 valid: hold on new data with some certainity
 novel: non-obvious to the system
 useful: should be possible to act on the item
 understandable: humans should be able to interpret the pattern
• Also known as Knowledge Discovery in Databases (KDD)
Applications
• Banking: loan/credit card approval
 Predict good customers based on old customers
• Customer relationship management:
 Identify those who are likely to leave for a competitor.
• Targeted marketing:
 Identify likely responders to promotions
• Fraud detection: telecommunications, financial transactions
 From an online stream of event identify fraudulent events
• Manufacturing and production:
 Automatically adjust knobs when process parameter changes
• Medicine: disease outcome, effectiveness of treatments
 Analyze patient disease history: find relationship between diseases
• Molecular/Pharmaceutical: identify new drugs
• Scientific data analysis: Identify new galaxies by searching for sub clusters
• Web site/store design and promotion: Find affinity of visitor to pages and modify layout
Some Basic Operations
• Predictive:
 Regression
 Classification
• Descriptive:
 Clustering / similarity matching
 Association rules and variants
Classification (Supervised learning)
Classification
• Given old data about customers and payments, predict new applicant’s loan eligibility.
Previous customers Classifier Decision rules
Salary > 5 L
Age Good/
Salary Prof. = Exec bad
Profession
Location
Customer type
New applicant’s data
Nearest Neighbor
• Define proximity between instances, find neighbors of new instance and assign majority class
• Case based reasoning: when attributes are more complicated than real-valued.
Decision Trees
• Tree where internal nodes are simple decision rules on one or more attributes and leaf nodes are
predicted class labels.
Salary < 1 M
Prof = teacher Age < 30
Good Bad Bad Good
Neural networks
• Useful for learning complex data like handwriting, speech and image recognition
Decision Boundaries:
Linear regression Classification tree Neural network
Clustering(Supervised Learning)
Clustering
• Unsupervised learning when old data with class labels not available e.g. when introducing a new product.
• Group/cluster existing customers based on time series of payment history such that similar customers in
same cluster.
• Key requirement: Need a good measure of similarity between instances.
• Identify micro-markets and develop policies for each
Applications
• Customer segmentation e.g. for targeted marketing
 Group/cluster existing customers based on time series of payment history such that similar
customers in same cluster.
 Identify micro-markets and develop policies for each
• Collaborative filtering:
 Group based on common items purchased
• Text clustering
• Compression
Association
Association rules
• Given set T of groups of items
Milk, cereal
• Example: set of item sets purchased
Tea, milk
• Goal: find all rules on itemsets of the form a-->b such that
 Support of a and b > user threshold s Tea, rice, bread
 Conditional probability (confidence) of b given a > user threshold c
• Example: Milk --> bread
• Purchase of product A --> service B cereal
Prevalent  Interesting
• Analysts already know about prevalent rules 1995
• Interesting rules are those that deviate from prior expectation Milk and
• Mining’s payoff is in finding surprising phenomena cereal sell
together!
Zzzz... 1998 Milk and

cereal sell
together!
What makes a Rule Surprising?
• Does not match prior expectation
Correlation between milk and cereal remains roughly constant over time
• Cannot be trivially derived from simpler rules
Milk 10%, cereal 10%
Milk and cereal 10% … surprising
Eggs 10%
Milk, cereal and eggs 0.1% … surprising!
Expected 1%
Applications of fast itemset counting
• Find correlated events:
Applications in medicine: find redundant tests
Cross selling in retail, banking
Business Analytics in Practice
Application Areas
Industry Application
Finance Credit Card Analysis
Insurance Claims, Fraud Analysis
Telecommunication Call record analysis
Transport Logistics management
Consumer goods promotion analysis
Data Service providers Value added data
Utilities Power usage analysis
Why Now?
• Data is being produced and warehoused
• The computing power is available and affordable
• The competitive pressures are strong
• Commercial products are available
• Growth in Data: The size of the data from the beginning of time to 2003 is now generated in five minutes
• BA Industry is $15 billion currently and is growing at 9 to 10% per annum
• Data Scientist – Sexiest Job – HBR
• Approx 4.4 million data scientists will be required by end of 2015
Analytics in Use
• The US Government uses Data Mining to track fraud
• A Supermarket becomes an information broker
• Basketball teams use it to track game strategy
• Cross Selling
• Target Marketing
• Holding on to Good Customers
• Weeding out Bad Customers
Some Success stories
• Network intrusion detection using a combination of sequential rule discovery and classification tree on 4
GB DARPA data
 Won over (manual) knowledge engineering approach
 http://www.cs.columbia.edu/~sal/JAM/PROJECT/ provides good detailed description of the entire
process
• Major US bank: customer attrition prediction
 First segment customers based on financial behavior: found 3 segments
 Build attrition models for each of the 3 segments
 40-50% of attritions were predicted == factor of 18 increase
• Targeted credit marketing: major US banks
 Find customer segments based on 13 months credit balances
 Build another response model based on surveys
 Increased response 4 times -- 2%
Another Success Story
• Example 1.5 A Sales-Promotion Model
• In the grocery industry, managers typically need to know how best to use pricing, coupons and
advertising strategies to influence sales.
• Using Business Analytics, a grocer can develop a model that predicts sales using price, coupons and
advertising.
Decision Models
Sales = 500 – 0.05(price) + 30(coupons)

+0.08(advertising) + 0.25(price)(advertising)
Business Analytics Process
Problem Solving and Decision Making
• BA represents only a portion of the overall problem solving and decision making process.
• Six steps in the problem solving process
1. Recognizing the problem
2. Defining the problem
3. Structuring the problem
4. Analyzing the problem
5. Interpreting results and making a decision
6. Implementing the solution
1. Recognizing the Problem
• Problems exist when there is a gap between what is happening and what we think should be happening.
• For example, costs are too high compared with competitors.
2. Defining the Problem
• Clearly defining the problem is not a trivial task.
• Complexity increases when the following occur:
 Large number of courses of action
 Several competing objectives
 External groups are affected
 Problem owner and problem solver are not the same person
 Time constraints exist
3. Structuring the Problem
• Stating goals and objectives
• Characterizing the possible decisions
• Identifying any constraints or restrictions
4. Analyzing the Problem
• Identifying and applying appropriate Business Analytics techniques
• Typically involves experimentation, statistical analysis, or a solution process
Much of this course is devoted to learning BA techniques for use in Step 4.
5. Interpreting Results and Making a Decision
• Managers interpret the results from the analysis phase.
• Incorporate subjective judgment as needed.
• Understand limitations and model assumptions.
• Make a decision utilizing the above information.
6. Implementing the Solution
• Translate the results of the model back to the real world.
• Make the solution work in the organization by providing adequate training and resources.
Job Profiles
Project Mechanism
• Recognizing the problem
Domain Expertise
• Defining the problem
Domain Expertise
• Structuring the problem
A technology expert with the domain expert with someone who knows Stats
• Analyzing the problem
Statistical Expertise
• Interpreting results and making a decision
Statistical Expertise along with Domain Expertise
• Implementing the solution
Domain Expertise
Job Profiles
• The Data / Technology Profile
Comfortable with Analytics
RDBMS, Operating Systems
Data Gatherer
• The Domain Profile
Comfortable with Analytics and IT, Can do basic Analytics
Knows the business, industry
Liaises with the Data Gatherer and the Stats Expert
• The Statistics Profile
Deep knowledge of Statistics and Modeling
Comfortable with IT
Our Course
Our Course – Why?
• Makes you very comfortable with Analytics and IT
• Focus is on
Popular Statistical Techniques
Commonly – used Tools and Technologies
Brief Overview of Domains
• Based on your Background
IT
Domain
Stats
Our Course – What?
• Basic Statistics with Sampling and Hypothesis Testing
Lay down the foundation
• Linear and Multiple Regression
Most common numeric prediction technique
• Logistic Regression and Classification
Most common application of BA
• Clustering
The backbone of Customer Segmentation
• Market Basket Analysis
Highly effective in the Retail Industry
Our Course – What?
• MS – Excel
90% of the organizations do 90% of their work on Excel, need we say more!
• R
The open source Data Mining Language is being used and accepted by more and more companies as
we speak.
• SAS
Old and Still strong Mining tool. The best tool for Visualization
• Overview of Big Data, Mobile BA
What to expect from the future
• 3 Case Studies from 3 Domains
Introduction to the domains and common applications within it
Our Course – How?
• Each Technique is taught with at least one tool and at least one dataset
• Example – Basic Statistics
Course taught in Excel
Class Data Set
Home Assignment Data Set
Evaluation
• IMS Proschool Certification
End Term Test (80%)
• NSE Certification
2 hour
75 MCQs
Thank You.

Orientation: Business Analytics

Uploaded by

Copyright:

Available Formats

You might also like

Orientation: Business Analytics

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Orientation: Business Analytics

Uploaded by

Copyright:

Available Formats

Business Analytics

Previous customers Classifier Decision rules

New applicant’s data

Prof = teacher Age < 30

Good Bad Bad Good

Linear regression Classification tree Neural network

Zzzz... 1998 Milk and

Sales = 500 – 0.05(price) + 30(coupons)

You might also like