Data Mining Concepts

Wipro Technologies
 Agenda
► The Data mining Technology
► Data mining Process
 Data Preparation
 Data Mining Models
► Data Mining Techniques
► Data Mining Applications & Tools
► Data Mining Methodologies
The Data mining
Technology
 A Problem...
► You are a marketing manager for a brokerage company
 Problem: Churn is too high
►Turnover (after six month introductory period ends) is 40%
 Customers receive incentives (average cost: $160)
when account is opened
 Giving new incentives to everyone who might leave is very
expensive (as well as wasteful)
 Bringing back a customer after they leave is both difficult
and costly
 … A Solution
► One month before the end of the introductory period is over,
predict which customers will leave
 If you want to keep a customer that is predicted to churn, offer
them something based on their predicted value
►The ones that are not predicted to churn need no attention
 If you don’t want to keep the customer, do nothing

► How can you predict future behavior?

 Tarot Cards
 Magic 8 Ball

► Data Mining
 Data Mining : Why now ?
► Changes in the Business Environment
 Customers becoming more demanding
 Markets are saturated
 Replace statistician  Better models, less grunge work
 Many different data mining algorithms / tools available
 Statistical expertise required to compare different techniques
 Build intelligence into the software

► Drivers
 Focus on the customer, competition, and data assets
► Enablers
 Increased data hoarding
 Cheaper and faster hardware
 Growing Base of data

• Data doubling every

20 months in the
world
• Businesses feel there
is value in historical
data
• Automated knowledge
discovery is only way to
explore this data
1970 1980 1990 2000
 Improved Data Collection

► Data Collection  Access  Navigation  Mining

► The more data the better (usually)
 Improved Algorithms
► Techniques have often been waiting for computing
technology to catch up
► Statisticians already doing “manual data mining”
► Good machine learning is just the intelligent
application of statistical processes
► A lot of data mining research focused on tweaking
existing techniques to get small percentage gains
Convergence of Three
Key Technologies

Increased
Computing
Power

DM
Statistical Improved
and Learning Data Collection
Algorithms and Mgmt
Motivation for doing Data
Mining
► Investment in Data Collection/Data Warehouse
► Add value to the data holding
► Competitive advantage
► More effective decision making
► OLTP =) Data Warehouse =) Decision Support
► Work to add value to the data holding
► Support high level and long term decision making
► Fundamental move in use of Databases
 Data Mining - Definition
► Data mining is the automated detection for new, valuable and
non trivial information in large volumes of data.
► It predicts future trends and finds behavior that the experts
may miss because it lies outside their expectations
 Data mining lets you be proactive
 Prospective rather than Retrospective
► DataMining Leads to simplification and automation of the
overall statistical process of deriving information from huge
volume of data.
 Data Mining Introduction
► DM - what it can do
 Exploit patterns & relationships in data to produce
models
 Two uses for models:
►Predictive
►Descriptive

► DM - what it can’t do
 Automatically find relationships
►withoutuser intervention
►when no relationships exist
 Data Mining Introduction
► Data Mining and Data Warehousing
 Data preparation for DM may be part of the Data
Warehousing
 Data Warehouse not a requirement for Data Mining
► DM and OLAP
 OLAP = Classic descriptive model
 Requires significant user input
 Example : Beer and diaper sales
►An OLAP tools shows reports giving sales of different items
►A data mining tool analyses the data and predicts ‘how
many times beer and diapers are sold together
 Data Mining Introduction
► DM and Classical Statistics
 Classical statistics based on elegant theory
and restrictive data assumptions
 Fine if data sets small and assumptions met
 Modeler plays active role - specifying model
form, interactions, etc
 In newer tools, pattern finding is data-driven
rather than user-driven
Data Mining : Introduction
 Data Mining is Not ...
► Data warehousing
► SQL / Ad Hoc Queries / Reporting
► Software Agents
► Online Analytical Processing (OLAP)
► Data Visualization
Data Mining Environment
Data mining - Analysis
Examples of Data Mining

Conditional Logic ► If profession = athlete

then age < 30 in 90 %
cases

Associations ► When Paint is sold, Paint

brushes are also sold
85% times

Trends & Variations

► Golf balls sales are
seasonal with Summer
peak and Winter low
Examples of Datamining contd....

✠How many people can be

Outcome Prediction expected to respond to a mailer
campaign?

Forecasting ✠What will be the total sales of this

product range in next quarter
taking into account seasonal and
long term trends?

Deviation Detection ✠Is this insurance claim likely to be

a fraud?

✠When a person is fired, he is

Link Analysis
likely to default on credit card
payments.
 Data mining - Users
► Executives - need top-level insights and spend far
less time with computers than the other groups.

► Analystsmay be financial analysts, statisticians,

consultants, or database designers.

► Endusers are sales people, scientists, market

researchers, engineers, physicians, etc.
The Data Mining
Process
Mining Technology is Just
One Part
Data Mining Cycle
 The DM Process
► Understand the Objective of the Data
Mining Effort
 Determine the Business Problem
 Translate Business Problem into a Data
Mining Objective
Data Mining Process
Data Preparation
 Data is the Foundation for
Analytics
► If you don’t have good data, your analysis will suffer
 Rich vs. Poor
 Good vs. Bad (quality)
► Missingdata
► Sampling
 Random vs. stratified
► Data types
 Binary vs. Categorical vs. Continuous
 High cardinality categorical (e.g., zip codes)
► Transformations
Don’t Make Assumptions
About the Data
 Data Preparation
► Data preparation – takes usually over 90% of our time
 Collection
 Assessment
 Consolidation and Cleaning
►table links, aggregation level, missing values, transformations, etc
 Data selection
►active role in ignoring non-contributory data?
►outliers?
►Use of samples
►visualization tools
Data Preparation
Data Preparation
 Data Preparation

► Cross Validation

 Break up data into groups of the same size

 Hold aside one group for testing and use the rest to build model
 Repeat
Data Mining Process
Data Mining Models
 The DM Process
► Model building
 an iterative process - different for supervised and unsupervised
learning
►Supervised Model
 Driven by a real business problems and historical data
 Quality of results dependent on quality of data
 Want to build a predictive model
►Unsupervised Model
 Want to find groups of things with similar characteristics
 Relevance often an issue
 Useful when trying to get an initial understanding of the data
 Non-obvious patterns can sometimes pop out of a completed data analysis project
 Types of Models
► Are We Trying to Predict What Will Happen or Describe
the State of the World?
► Descriptive modeling
 Clustering
 Association
 Sequence discovery
► Predictive modeling
 Classification
 Regression
 Time series
 Types of Models
► Descriptive Models for Clustering and Associations
 Clustering algorithms: K-means, Kohonen
 Association algorithms: apriori, GRI

► Prediction Models for Regression and Classification

 Regression algorithms: neural networks, rule induction, CART (OLS regression,
GLM)
 Classification algorithms: CHAID, C5.0 (discriminant analysis, logistic regression)

► Some models are better than others

 Accuracy
 Understandability
 Determining the Model
► Hierarchy of Data Mining Solutions:
► Business Goal
 Data Mining Goal -
►Type of model (predictive, descriptive)
 Algorithm

► Oncethe Data Mining Goal of the Data

Mining Effort is Determined, the Technique
Used to Meet that Goal Falls into Place
 Determining the Model
► Hierarchy of Data Mining Solutions
Step Num Step in Hierarchy Value Chosen

Step 1 Business Goal Increase

Revenues
Step 2 Data Mining Goal Identify
customers who
are likely to buy

Step 3 Type of model Predictive

Step 4 Algorithm Rule Induction
 Model Building
► The choice of model depends on model type which
will influence data preparation step
► The essence of model building is to train the
model with a subset of data, test it with an
independent subset
► Some algorithms inherently test as part of model
construction:
►e.g., neural networks, rule induction
► Others require user designation of train/test data
 Model Building
► Evaluation of model: how well it performed on test data
► Methods and criteria depend on model type:
►e.g. mean error rate with regression models
► Interpretation
of model: important or not, easy or hard
depends on algorithm
How are Data Mining Models
Built and Used?
 The DM Process
►
Use the model on application
deploy models on line

on the web

against the database


►
Model monitoring
Determine if the model still ‘works’

Data Mining Process
Data Mining Models
Types of Models
 Clustering
► Divide the data into a
number of different groups
► Determine the attributes
that characterises a group
automatically
► Can be used for
classification of new cases
 K-Means Clustering
► User starts by specifying the number of
clusters (K)
► K data points are randomly selected
► Repeat until no change in specific clustering
100

statistics
Age

0 Dose (cc’s) 1000

46
 Association
► Identifies
the items that occur
together in a given event
 If ‘A’ occurs then x% (confidence factor) of
the times ‘B’ occurs. This is found in y%
(Support) of the data.
► Used for ‘Market Basket Analysis’
 Association Rules
► Finds relations among attributes in the
data that frequently co-occur .
 E.g., Association Rules. Popular for basket
analysis Then
Buy diapers
on Buy beer
Friday night

Transaction Day Item 1 Item 2 Item 3 …

No.
100 Fri Beer Diaper Chocolate

101 Fri Milk Chocolate Shampoo

102 Thu Beer Wine Vodka

103 Fri Beer Cheese Diaper
104 Fri Ice Cream Diaper Beer
 Sequence discovery
► Finds association among the items
► The related items are spread over the
time
 If surgical procedure ‘A’ is followed, then
the infection ‘B’ occurs at a later time
 If stock ‘A’ rises by 12% then stock ‘B’ also
rises by 8% within 2 days
► Requiresinformation on transactions
and transactors
 Predictive Modeling

►A “black box” that makes predictions about the future

based on information from the past and present

► Large number of inputs usually available

 How Good is a Predictive
Model?

► Response curves
 How does the response rate of a targeted
selection compare to a random selection?
 Classification
► Goal of classification is to build structures from
examples of past decisions that can be used to make
decisions for unseen cases.
► Predicts the cluster in which a new case fits in
► The characteristics of the groups can be defined by
an expert or fed from historic data
► Often referred to as supervised learning.
► Decision Tree and Rule induction are popular
techniques
► Neural Networks also used
 Regression
► Forecasts the future values based on
existing values
► Types
 Simple - one independent variable
 Multiple - more than one independent
variables
Example 1: Regression
Name Income Age Order
Amt
a 23000 30 83
b 51100 40 131
c 68000 55 178
d 74000 46 166
e 23000 47 117
Pattern:

This will result into Amount=

Linear graph 0.001*Income+2*Age
Example 2: Regression
100
Age

0
Dose (cc’s) 1000
 Time series
► Forecasts the future trends
► Model includes time hierarchy like
year, quarter, month, week etc.
► Considers impact of seasonality,
calendar effects such as holidays
► What is the expected price for
Microsoft’s stock by the end of this
year?
Data Mining
Techniques
 Techniques
► Neural Networks
► Decision Trees
► Rule Induction
► K nearest neighbour
 Neural Networks
► Parameter adjustment systems
► Interconnected elements (neurons)
► Train the net on a training data set
► Use Trained net to make predictions
► Can deal with only numeric data
 Neural Networks -
functioning
Adjust Weights

Feedback
Prediction Learning
Module

Output Actual
Middle Layer Data
Layer
Input
(functions)
Layer
 (Feed Forward) Neural
Networks

► Very loosely based on biology

► Inputs transformed via a network of simple processors
► Processor combines (weighted) inputs and produces an output value
► Obvious questions: What transformation function do you use and how
are the weights determined?
Types of Neural Networks
 Applications and Issues
► Neural Networks can be used for predictive modeling
► Kohonen networks - a type of neural networks can be used for
clustering
► Drawback - Activities are completely black box
► Key problem: Difficult to understand
 The neural network model is difficult to understand
 Relationship between weights and variables is complicated
► Graphical interaction with input variables (sliders)
 No intuitive understanding of results
► Training time
 Error decreases as a power of the training size
► Significant pre-processing of data often required
 Decision Trees
►A way of representing series of rules
► Trees are grown through an iterative
splitting of data
► Goal is to maximize the distance
between groups at each split
► Can have two branches (binary tree)
or more than two
 Decision Trees
► Each node is called a decision node
► A branch leads to another decision
node or to the bottom of the tree
► By moving along the tree one can
reach a decision by deciding at each
node which branch to take
 Decision Trees
► Decision trees are of two types
 Classification trees
►used to predict categorical values
 regression trees
►used to predict continuous variables
Decision Trees : Examples

Age < 35 Age ≥ 35

Dose < 100 Dose ≥ 100 Dose < 160 Dose ≥ 160
Decision Trees : Examples
 Advantages & Limitations
► Advantages
 Models can be built very quickly
 Suitable for large data sets
 Easy to understand
 Gives reasons for a decision taken
 Handle non-numeric data very well
 Minimum amount of data transformation
► Limitations
 Leads to an artificial sense of clarity
 Trees left to grow without bound take longer to build and become unintelligible
 May overfit the data
 Algorithms used for splitting are generally univariate - using single independent
variable at a time
 Rule Induction
► Extraction of useful if-then rules from
data based on statistical significance.
► Completely a machine driven process.
► Can discover very general rules which
deal with both numeric and non-numeric
data.
► Translating the rules into a usable model
must be done either by the user or a
decision trees interface
 Rule Induction Examples
► If Car = Ford and Age = 30…40
Then Defaults = Yes ,Weight = 3.7
► If Age = 25…35 and Prior_purchase = No
Then Defaults = No, Weight = 1.2
► Not necessarily exclusive (overlap)
► Start by considering single item rules
 If A then B
►A = Missed Payment, B = Defaults on Credit Card
 Is observed probability of A & B combination greater than
expected (assuming independence)?
►If It is, rule describes a predictable pattern
 Rule Induction: Important
points
► Look at all possible variable combinations
► Compute probabilities of combinations
► Expensive!
► Look only at rules that predict relevant behavior
► Limit calculations to those with sufficient support
► When moving onto larger combinations of variables like
n3, n4, n5, ...
 Support decreases dramatically, limiting calculations
 K Nearest Neighbor
►A classification technique
► Decides in which class to put a new
case in
► Criterion is to find a maximum number
(k) of neighbors having most similar
properties
► Assigns a new case to the same class
to which most of the neighbors belong
 K Nearest Neighbor
► Set of already classified cases selected to use
as a basis.
► Neighborhood size, in which to do the
comparisons, decided.
► Decided how to count the neighbors (assigning
weights to neighbors, may give more weight to
nearer neighbor than a farther one).
 K Nearest Neighbor Model

100
Age 0 Dose (cc’s) 1000

► Use entire training database as the model

► Find nearest data point
► Very easy to implement. More difficult to use in production.
► Disadvantage: Huge Models
K Nearest Neighbor Diagram
 Developing a Nearest
Neighbor Model
► Model generation:
 What does “near” mean computationally?
 Need to scale variables for effect
 How is voting handled?
► Confidence Function
► Conditional probabilities used to calculate weights
► Optimization of this process can be mechanized
► Developing a Nearest Neighbor Model
 Example of a Nearest
Neighbor Model
► Weights:
 Age: 1.0
 Dose: 0.2
∆ Age + 0.2 × ∆ Dose
► Distance =

► Voting: 3 out of 5 Nearest Neighbors (k = 5)

► Confidence = 1.0 - D(v) / D(v’)

 Comparison of Algorithms
►K Nearest Neighbor
 Quick and easy
 Models tend to be very large
► Neural Networks
 Difficult to interpret
 Can require significant amounts of time to train
► Rule Induction
 Understandable
 Need to limit calculations
► Decision Trees
 Understandable
 Relatively fast
 Easy to translate into SQL queries
 Supervised / Unsupervised
► Supervised Learning:

1. As the name applies Techniques are supervised to work or perform. These models
are first TRAINED using data whose RESPONSE variable or result is already
KNOWN.
2. Predictive models (classification, regression etc) fall under this category –They have
to train and then test. Estimated value is to compared with known value

► Unsupervised Learning:

1. As the name applies Techniques are unsupervised. There is no already known

result to guide the algorithm.
2. Descriptive Techniques clustering, Association etc falls under this category .
 Data Mining
Applications & Tools
 Applications
► Customer Profiling
► Target Marketing
► Market Basket Analysis
► Fraud Detection
► Medical Diagnostics
► Direct mail marketing
► Web site personalization
► Bioinformatics
► Anti Money Laundering
► Churn Analysis
 Some Uses of DM
► DM can be used to discover new purchasing trends, plan
investment strategies, and detect unauthorized expenditures in
the accounting system.
► It can improve marketing campaigns and the outcomes can be
used to provide customers with more focused support and
attention.
► Many law enforcement and special investigative units, whose
mission is to identify fraudulent activities and discover crime
trends, have also used data mining.
► Like identifying the critical behavior patterns in the
communication interactions of narcotics organization
► The monetary transactions of money laundering and insider
trading operations
► The movement of serial killers, and the targeting of smugglers at
border crossings.
► Government that maintains large data sources as part of
activities relating to matters of national security.
 Applications

Source: IDC 1998

 CHURN ANALYSIS

► Road Map for Minimizing Churn Rate

Churning means switching – A PROBLEM

Increasingly competitive environment customer retention

has surfaced as one of the key problem faced by mobile
service provider .

Business Objective: TO MINIMISE CHURN RATE

 CHURN ANALYSIS

Data Mining Goal

IDENTIFIED CUSTOMER WITH DELIQUENT
Scope
NATURE.

•Assign Churn Score to all customers in order to identify those who are
most likely to churn (Quarter etc).

•Determine the most relevant parameter that influences the inclination

to churn.

•Define Clearly segments that are strongly divided by their churn relating
Behavior
 CHURN ANALYSIS
► Basic Understanding

There are mainly two types of churn

1. Customer Request (CRQ)

2. Forced Churn (Defaulters)

 CHURN ANALYSIS

► Information Sources

 Call Statistics (CDR)

 Credit History
 Billing History
 Revenue History
 Payment History
 Survey Data
 Demographic data
 Complaint information
 CHURN ANALYSIS

► Suggested Analysis
Pareto analysis
Also called 80/20 Analysis. Its been observed that 80%
of the revenue profit comes from 20 % of the
customer. Key Business Improvement was identifying
those 20% and serves them better.
Techniques/ Reports/Algorithms

Characterization and summarization, Top 10 report ,

List ,Cross tab Reports , Graph Charts etc.
 CHURN ANALYSIS
► Suggested Analysis
Loyalty Analysis
A loyal customer is worth new customer. If it is possible to identify the
loyal customer and increase that volume. A loyal customer is defined
as the one who is with the company for last six months. This analysis
will give insight in to the complete details of various customer bases.

Techniques/ Reports/Algorithms

Characterization and summarization, Top 10 report , List ,Cross tab Reports ,

Graph Charts etc.
 CHURN ANALYSIS
► Suggested Analysis
Customer Profit Analysis
dentifying wining and loosing customer. A wining customer is one who
giving increasing revenue month after month and vice versa. Identify
he characteristic and reason for better decision.

Techniques/ Reports/Algorithms

Characterization and summarization, Top 10 report , List ,Cross tab Reports ,

Graph Charts etc.
 CHURN ANALYSIS
► Suggested Analysis
Trend Analysis

It’s a Visualization Technique. This Technique uses parallel

Coordinate system to show the trend of various measures over
different time period

Techniques/ Reports/Algorithms

Parallel Coordinate graph

 CHURN ANALYSIS
► Suggested Analysis

Customer profiling
active accounts, Light user, risky customer Active accounts, Loss makin
ofit making accounts. This segment helps in mapping with the predictiv
gment.

Techniques/ Reports/Algorithms

List, Cross Tab, clustering , Graph Charts

 CHURN ANALYSIS
► Suggested Analysis
LTV Analysis

Called Life Time value Analysis .Revenue projected over 25 yrs and
Projected Churning loss and rate.

Techniques/ Reports/Algorithms

List report, Line Graphs, graph Charts

 CHURN ANALYSIS
► Suggested Analysis
Churn Modeling

st of acquiring new customer is more than retaining one. Classification o

urn models. Assign churn score and predictive modeling.

Techniques/ Reports/Algorithms

Scoring, Decision Tree, neural network, Clustering.

 CHURN ANALYSIS
► Suggested Analysis
Survival Analysis
This predicts how long the customer would continue with existing
service in terms of time. What measures can be taken. One of the
Popular Technique are K. Hazard Analysis .

Techniques/ Reports/Algorithms

K.Hazard technique
 CHURN ANALYSIS
Suggested Approach

Derived Customer Segmentation mapped them against their

probability of
churning and expected Profits. Each Segment has significantly
different
Assign
usageChurn
characteristic,
score and demographic
create churn models
and channel selection.

Note

Data Mining is an iterative process. Next Step depends upon the

Outcome of previous result. There cannot be fixed approach.
 Tools and Vendors
► IBM - Intelligent miner
► SPSS - Clementine
► SAS - Enterprise miner
► SGI - Mine Set
► Accrue - Decision Series
► Poly Analyst – Megaputer.
 Comparison of Tools

A lgo rithm s P latfo rm s

N a ïvek- k- T im e W in
T oo l N N T re e B a ye sM nsN N Sta ts P re dSe rie sC lust A ssoc32 U N IXP a r
IB M – Inte llige nt
M ine r √ √ √ √ √ √ √ √ √ √
SP SS –
C le m e nt ine √ √ √ √ √ √ √ √
SA S – E nte rprise
M ine r √ √ √ √ √ √ √ √ √
SGI –M ine Se t √ √ √ √ √ √ √ √ √
A c c rue Softw are –
D e c ision Se rie s √ √ √ √ √ √ √ √ √
 Trends and Market Updates
► ACRM (Analytical Customer Relationship Management) ,Web Mining , ITM
(Intelligent Transaction Mining) and Text Mining.
► Anti Money Laundering
1. An increased criminal (Terrorist) activity has become the major concern with most of the
countries of the world. To Track the funds being used for terrorist activities has become the
major application and challenge in a Finance Sector.

► CIBIL
1. It’s the first credit information bureau being established in India 2003. CIBIL will obtain and
Share data on borrowers both consumer and commercial for sound credit decision
therefore helping to avoid adverse selection.
2. Availability of credit information facilitates credit scoring mechanism and Credit Risk
Analysis will play an important role in that

► Basel II
1. As per the Basel II Accord which serves as a guideline for the banks across 32 countries to
reduce credit risk, Operation risk and business risks. By 2007 all banks should have data
warehouse in place so that information should be available for risk related analysis.
Data Mining
Methodology
Data Mining Methodology
 Methodologies

► Discussion on following methodologies

 CRISP-DM methodology
 Wipro - Data mining methodology
 CRISP - DM
► CRoss Industry Standard Process
model - Data Mining
► Hierarchical approach
 Phase
 Generic Task
 Specialised Task
 Process Instance
CRISP - DM
 Mapping Generic to Specific
Model
► Application domain
 The specific area in which the data mining project
takes place
► Data mining problem type
 The specific class(es) of objective(s) which the
data mining project deals with
► Technical aspect
 covers specific technical issues in data mining
► Tool and technique
The Cycle
 Phases
► Business Understanding
► Data Understanding
► Data Preparation
► Modeling
► Evaluation
► Deployment
 Business Understanding
► Understanding the project objectives
and requirements from a business
perspective
► Converting this requirements into a
data mining problem definition
► Preliminary plan designed to achieve
the objectives.
 Data Understanding
► Initialdata collection
► Identifying data quality problems
► detect interesting subsets to form
hypotheses for hidden information.
 Data Preparation
► Identificationof data at Table, Record,
and Attribute level
► Transformation and cleaning of data
for modeling tools.
► Data preparation tasks are performed
multiple times
 Modeling
► Identificationof the suitable modeling
techniques for the requirement
► Applying the various possible models
on the data set
► Requires stepping back to the data
preparation phase, due to the model
specific requirements
 Evaluation
► Evaluate the model, and review the
steps executed to construct the model,
from a business perspective
► Determining the business issues that
are not sufficiently considered.
► Deciding on the use of the data mining
results
 Deployment
► This can be as simple as generating a report
or as complex as implementing a repeatable
data mining process, depending on the
requirements.
► Normally carried out by the customer and
not the data analyst.
► Customer has to understand up front what
actions will need to be carried out in order to
actually make use of the created models.
Wipro’s - DM Process
Model
Business Understanding
•Domain Understanding
•Problem Selection
•Solution Selection

Building a mining database

•Data collection
•Data description
•Selection
•Data quality assessment and data cleansing
•Consolidation and integration
•Metadata construction
•Load the data mining database

Explore the Data Prepare Data for Modeling

•Outliers •Select Variables
•Summaries •Select Rows
•Visualizations •Construct new variables
•Transform Variables

Evaluation & Deploy Model &

Model Building
Interpretation Results
 Business Understanding
► Understanding the project objectives
and requirements from a business
perspective
► Converting this requirements into a
data mining problem definition
► Preliminary plan designed to achieve
the objectives.
 Building a mining database
► Data collection
► Data description
► Selection
► Data quality assessment and data cleansing
► Consolidation and integration
► Metadata construction
► Load the data mining database
► Maintain the data mining database
 Explore the data
► Identifyoutliers.
► Summarization.
► Visualization.
 Prepare data for modeling
► Select variables
► Select rows
► Construct new variables
► Transform variables
 Building and Deploying
model
► Data mining model building
► Evaluation and interpretation of the
model
► Deploy the Model and Results
Any Questions ?
Thank You for
your time