8

Introduction to Data
Mining

Business Analytics,1e
By Sanjiv Jaggia, Alison Kelly, Kevin Lertwachara, and Leida Chen

Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
8/26/2020 written consent of McGraw-Hill Education.
8-1
Chapter 8 Learning Objectives (LOs)
LO 8.1 Describe the data mining process.
LO 8.4 Conduct principal component analysis.

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-2

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-2
Introductory Case: Social Media Marketing
• Alissa Bridges is the marketing director of FashionTech. Alissa has hired a
social media marketing firm, MarketWiz, to develop predictive models that
would help FashionTech acquire new customers as well as increase sales from
existing customers.
• Using FashionTech’s historical social media marketing and sales data,
MarketWiz develops two types of predictive models.
– A classification model that predicts potential customers’ purchase probability from FashionTech
within 30 days of receiving a promotional message in their social media account.
– Two prediction models that predict the one-year purchase amounts of customers acquired
through social media channels.
• In order to assess the performance of the predictive models, Alissa’s team
would like to use the validation data set to:
– Evaluate how accurately the classification model classifies potential customers into the purchase
and no-purchase classes.
– Compare the performance of prediction models to estimate the one-year purchase amounts of
customers acquired through social media channels.

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-3

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-3
8.1: Data Mining Overview (1/8)
• Applications of computer software used to obtain insightful solutions that
traditional data analysis techniques may not be able to achieve.
• Artificial Intelligence (AI)
– Demonstrate human-like intelligence and cognitive functions
– Deduction, pattern recognition, and interpretation of complex data
– Examples: Deep Blue playing chess, Watson
• Machine Learning
– Application of AI that allows the computer to learn automatically
– Uncover hidden patterns and relationships
– Use self-learning algorithms to evaluate results and improve performance over time
– Example: predict rider demand to strategically dispatch drivers for Uber
• Data Mining
– Process of applying a set of analytical techniques for machine learning and AI
– Uncover hidden patterns and relationships in data
– Data segmentation, pattern recognition, classification, prediction
– Example: group customers into segments for customized promotions
• These are often grouped together or used interchangeably.

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-4

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-4
8.1: Data Mining Overview (2/8)
• Data mining is a complex process of examining large
sets of data for identifying patterns and then using them
for valuable business insights.
• Established standards with defined steps
– Cross-Industry Standard Process for Data Mining (CRISP-
DM): SPSS, TeraData, Daimler AG, NCR, and OHRA
– Sample, Explore, Modify, Model, and Assess (SEMMA): SAS
– CRISP-DM found to be more popular
– Not every step needed in every application
– Data preparation and modification are the most important
– 80% of time related to data
• It is always important to fully understand the
socioeconomic climate, business goals, and underlying
issues.
BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-5
Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-5
8.1: Data Mining Overview (3/8)
CRISP-DM

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-6

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-6
8.1: Data Mining Overview (4/8)
• Business understanding: situational context, specific objectives,
project schedule, deliverables
• Data understanding: collecting raw data, preliminary results,
potential hypotheses
• Data preparation: record and variable selection, wrangling,
cleaning
• Modeling: selection and execution of data mining techniques,
convert or transform data to formats/types needed for certain
analyses, document assumptions, cross-validation
• Evaluation: evaluate performance of competing models, select
best models, review and interpret results, develop
recommendations
• Deployment: develop a set of actionable insights and a strategy
for deployment/monitoring/feedback
BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-7
Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-7
8.1: Data Mining Overview (5/8)
SEMMA

• Sample: identify appropriate variables, merging and/or

dividing, sample
• Explore: exploratory data analysis
• Modify: variables are selected, created, and/or
transformed
• Model: analysis techniques and models are chosen and
applied
• Assess: results from different models are presented to end
users, compare outcomes and performance, new
observations are scored

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-8

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-8
8.1: Data Mining Overview (6/8)
• Data mining algorithms are classified into two types of
techniques depending on the way they learn about data
• Supervised
– Target is known/exists
– Regression, k-Nearest Neighbors, naïve Bayes, Decision Trees
– Classification model: target is categorical, objective is to predict
class
– Prediction model: target is numeric, objective is to predict the target
for a new case
• Unsupervised
– Target is not known
– Principal components, clustering
– Dimension reduction: convert high-dimensional data to lower
– Pattern recognition: recognize patterns using machine learning

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-9

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-9
8.1: Data Mining Overview (7/8)
• In supervised data mining, the target (response) is known.
• Historical values of the target variable exist, can examine
the impact of the predictor variables on the target.
• Common algorithms are based on statistical techniques.
• Example: linear and logistic regression
– Predict the target (response) based on several predictor variables
– Data are denoted as 𝑦, 𝑥1 , 𝑥2 , … , 𝑥𝑘
– Trained or supervised because of the known target
• Classification example: classify stock buy, hold, or sale
• Prediction example: spending of a customer

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-10

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-10
8.1: Data Mining Overview (8/8)
• In unsupervised data mining, the target (response) is not known.
• Allow the computer to identify complex processes without guidance
• Used in exploratory data analysis and descriptive analytics
• Used prior to supervised learning
– Understand the data
– Formulate questions
– Summarize data
• Dimension reduction
– Convert a set of high-dimensional data (large number of variables) into data
with lesser dimensions without losing much of the information
– Reduce information redundancy
– Improve model stability
• Pattern recognition
– Recognize patterns using machine learning
– Recurring sequences, frequent combinations, recognizable features, common
characteristics

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-11

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-11
8.4: Principal Component Analysis (1)
• High-dimensional data
– Variables are redundant or highly correlated which causes model instability
• Principal Components Analysis (PCA) is a dimension reduction
technique used to reduce the number of variables without losing
important information.
– Transforms a large number of possibly correlated variables into a smaller
number of uncorrelated variables (principal components).
– Summarize the data with less variables that collectively explain most of the
variation in the data.
– Use the principal components in other unsupervised learning methods such
as clustering.
– Use the principal components in other supervised learning such as
regression (as predictors).

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-12

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-12
Use TAM
• Put all variables that you want to group
• How many components?
• Are they all on separate columns?
• Take a look at BI1. Where should it be
placed?

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-13

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-13
Use Health
• Put all variables that you want to group
• How many components?
• Are they all on separate columns?

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-14

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-14
Exploratory Factory Analysis
• Similar to PCA
• Input all variables
• How many components?
• Choose Maximum Likelihood for
Extraction
• Choose for Promax Rotation
• You can change number of factors (if you
already have an idea)
BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-15
Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-15
Confirmatory Factor Analysis
• You already have a set of scales given the
model.
• Create one factor at a time.
• Take a look at the final model.

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-16

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-16
Reliability Analysis
• For Likert Scale questions only
• Use TAM (compute for reliability for each
major variable)
• Use Online Gaming Study (Q1 to Q8)

BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-17

Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-17
Exercises
• Please use only continuous variables.

• Baseball Players (PCA, EFA, CFA)

• Internet Addiction (PCA, EFA, CFA)
• Happiness (PCA, EFA, CFA)

• For the CFA, use the results of PCA for

number of factors (just call them Factor 1, 2
et al)
BUSINESS ANALYTICS, 1e | Jaggia, Kelly, 8-18
Copyright © 2021 McGraw-Hill Education. AllLertwachara,
rights reserved.Chen
No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
Copyright © 2021 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education. 8-18