Lecture : Business Analytics Primer

By:
Dr. Prithvi Bhattacharya
Lecture Outline

1.1 Introduction
1.2 Basics of Business Analytics
1.3 Life Cycle of Business Analytics
Lecture Outline

1.1 Introduction
1.2 Basics of Business Analytics
1.3 Life Cycle of Business Analytics
 Instructor Details

Lecturer/Coordinator: Dr. Prithvi Bhattacharya

Discipline Leader- Management Sciences

O ffice: UOWD Building Level 4 Room 4.28

E mail: [email protected]
[email protected]

Weekly C o n s u l t a t i o n : Thursdays from 10AM-12Noon (online)

https://uow.webex.com/meet/prithvi
 About Me

• Bachelors in Commerce (Honours), India

• Masters in Information Technology, Monash University
• Education: Australia
• PhD in Management of Information Technology, University of
Melbourne Australia
• Professional Certification in Data Science and Analytics,
MIT USA
• SAP Consultant , Accenture Australia
• Previous • Business Analyst, University of Melbourne Australia
work: • Academic Faculty, HCT, UAE

• Towards a Cross- Sectional Framework for Optimal Modelling for

• Current Research Business Problems
• Aligning enterprise systems capabilities with business strategy using
enterprise architecture
 If you’ve got a question…

MANDATORY:
• Check Moodle and Email regularly for latest information

• Please see me during consultation hours via prior email

• If you prefer to email, please use your UOW email ONLY

• I’ll try my best to get back to you within 24h on a
weekday
• If required, we can meet online
 About you…

• What is your experience with data?

• What do you want to learn?

• Why is your aspiration?

Lecture Outline

1.1 Introduction
1.2 Basics of Business Analytics
1.3 Life Cycle of Business Analytics
 What is Business Analytics ?

“Business analytics refers to a broad use of various

quantitative techniques such as statistics, data mining,
optimization tools, and simulation supported by the query and
reporting mechanism to assist decision makers in making
more informed decisions”
Applications of Business Analytics

You Name it !!!

 Prospects For Business Analytics
• 6 out of 12 Technology Trends for 2022 as projected by Gartner include
‘Analytics’ in some form !
https://www.gartner.com/en/information-technology/insights/top-technology-trends

• Business Analytics is needed by every business function in every industry in

every sector in every country !

• It is like the air (almost) !

• This means… More JOBS!!!

• !
 Jobs After Business Analytics
• Data Analyst
• Data Scientist
• Business Analyst
• Operations Research Analyst
• Market Research Analyst
• Data Engineers
• Data Managers
• Chief Information Officer (CIO) !
Jobs In Business Analytics
 What about ….?
• Artificial Intelligence (AI)
• an umbrella discipline that covers everything related to making machines make decisions
autonomously.

• Machine Learning (ML)

• A subset of AI : a systems can automatically learn and improve without explicitly being
programmed

• Deep Learning
• A subset of ML: a technique that is inspired by the way a human brain filters information
• Uses Neural Networks as a foundation
 What about …..?
• Business Intelligence
• It is the process of collecting, storing and analysing data from business
operations.
• It provides comprehensive business metrics, in near-real time, to support better
decision making
• It does not usually predict the future

• Data Mining
• It is the process of analyzing a large batch of information to discern trends in data
using computer science algorithms

• Data Science
• It combines multiple fields, including statistics, computer science, artificial intelligence
(AI), and data analysis, to extract value from data.
• Data science encompasses preparing data for analysis, including cleansing,
aggregating, and manipulating the data to perform advanced data analysis
 What about …..?
• Data Analytics
• Data Engineering
• ……

“What's in a name? that which we call a rose by any other

name would smell as sweet”
- William Shakespeare
A jungle of Software for Business Analytics
 Different Types of Business Analytics

Categorised into 3 broad types

◦Descriptive Analytics
◦Predictive Analytics
◦Prescriptive Analytics
 Different Types of Business Analytics

◦Descriptive Analytics
◦What has happened and why ?
◦Descriptive analytics looks at data statistically to tell you what happened in the past.
◦This can be in the form of data visualizations like graphs, charts, reports, and dashboards.
◦In a healthcare setting, for instance, say that an unusually high number of people are admitted
to the emergency room in a short period of time.
◦Descriptive analytics tells you that this is happening and provides real-time data with all the
corresponding statistics (date of occurrence, volume, patient details, etc.).
 Different Types of Business Analytics

◦Predictive Analytics
◦What will happen next ?
◦Predictive analytics takes historical data and feeds it into a machine learning model that
considers key trends and
◦The model is then applied to current data to predict what will happen next.
◦Back in our hospital example, predictive analytics may forecast a surge in patients admitted to
the ER in the next several weeks.
◦Based on the predictions, the illness will be spreading at a rapid rate.
 Different Types of Business Analytics

◦Prescriptive Analytics
◦What is our best course of action ?
◦Prescriptive analytics takes predictive data to the next level. Now that you have an idea of
what will likely happen in the future, what should you do?
◦It suggests various courses of action and outlines what the potential implications would be for
each.
◦Back to our hospital example: now that you know the illness is spreading, the prescriptive
analytics tool may suggest that you increase the number of staff on hand to adequately treat the
influx of patients.
 Descriptive Analytics

Type Explanation

Central Tendency Central tendency represents the center point or typical value of a dataset.
EG: what are the average sales per quarter?

Dispersion Dispersion (also called variability, scatter, or spread) is the extent to which a distribution is stretched or
squeezed.
EG: How much do the sales vary ?

Shape and Form Shape and form of the data provide general features of the data, often very useful for data exploration and
diagnostics

Association Association is any relationship between two measured quantities that renders them statistically dependent
EG: Do sales depend on product prices?
 Predictive Analytics
Type Explanation
Causal modeling Causal modeling attempts to help us understand what events or actions actually influence others.
E.g.: “Will a particular drug help decrease blood pressure level in patients”.
Value estimation Value estimation attempts to estimate or predict, for each individual, the numerical value of some variable for that individual
E.g.: “How much will a given customer use the service?”

Classification Classification attempt to predict, for each individual in a population, which of a (small) set of classes this individual belongs to
E.g.: “Among all the customers, which are likely to respond to a given offer?”

Clustering Clustering attempts to group individuals in a population together by their similarity, but not driven by any specific purpose
E.g.: “Do our customers form natural groups or segments?”
Co‐occurrence Grouping Co‐occurrence Grouping (also known as frequent item‐set mining, association rule discovery, and market‐basket analysis)
attempts to find associations between entities based on transactions involving them.
E.g.: What items are purchased together?

Similarity Matching Similarity Matching attempts to identify similar individuals based on data known about them; it can be used directly to find similar
entities
E.g.: “Find people who are similar to you in terms of the products they have liked or have purchased”.
Network Link Prediction Network Link Prediction attempts to predict connections between data items, usually by suggesting that a link should exist, and
possibly also estimating
the strength of the link.
E.g.: “Since you and Karen share 10 friends, maybe you’d like to be Karen’s friend?”
Change Point Detection Change Detection in a time series data attempts to detect changes quickly that are significant for either action to be taken or as a
result of an action taken
E.g.: “Has there been an increase in global temperature over a period of time to alarm us about global warming?”
Forecasting Time Series Forecasting Time Series involves taking models fit on historical data and using them to predict future observations
E.g. “What will be the temperatures for the next few days in a city?”
 Prescriptive Analytics

Type Explanation
Simulation Simulation attempts to show the eventual real effects of alternative conditions and courses of action when the real system
cannot be engaged.
E.g.: “How can we minimize the response time of serving customers in a call centre ? ”.
Optimization
Optimization attempts to derive optimal solutions based on several objectives. The objective most often considered is
the operational cost to be
minimized.

E.g.: “How can we give the best balanced diet to workers at the minimum cost?”

Logic‐based Logic‐based prescriptions involve incorporating the expert knowledge into the prescriptive analytics models and is based on
prescription rules represented by
mathematical functions.

E.g.: “How can we do prescriptive maintenance planning for all our machines in our factory ?”
 Different Types of Data

Categorised into 3 broad types

◦Structured Data
◦Semi-Structured Data
◦Un-Structured Data
 Different Types of Data

Structured Data
• Data that fits neatly within fixed rows and columns in relational databases and spreadsheets.
• Examples of structured data include names, dates, addresses, credit card numbers, stock
information, geolocation, and more.
 Different Types of Data

Semi-Structured Data
• Data not in rows and columns but in some other sort of structure
• Based on Industry standards
• Examples : data as XML, JSON , NOSQL formats
 Different Types of Data

Un-Structured Data
• Unstructured data is essentially everything else.
• It may be textual or non-textual, and human- or machine-generated.
• It can be of any form like text, image audio, video etc..
• Examples : data from text files, emails, websites, twitter feeds, Facebook and Instagram
posts, movie clips, surveillance camera footage, sensors…
 Different Types of Data: BIG Data

Big Data
• Big data is a term that describes large, hard-to-manage volumes of data – both structured
and unstructured – that inundate businesses on a day-to-day basis.
• The 5 V's of big data (velocity, volume, value, variety, veracity, and variability)
Analytics Terminology
 Analytics Terminology

• Each column/field in a dataset is called a variable

• Categorical vs numerical

• Each row/record in a dataset is a case/observation/datapoint:

• Customer, tax return, applicant, day, application,…
 Supervised Vs Unsupervised Learning

• What Does “Learning” Refer to ?

• Machine Learning

• Machine learning is a branch of artificial intelligence (AI) and

computer science which focuses on the use of data and
algorithms to imitate the way that humans learn, gradually
improving its accuracy.

• Machine learning algorithms build a model based on sample

data, known as "training data", in order to make predictions or
decisions without being explicitly programmed to do so
 Supervised Learning

• Supervised learning, also known as supervised machine learning, is a

subcategory of machine learning
• It is defined by its use of labeled datasets to train algorithms to predict outcomes
accurately.
• In supervised learning, we need to find the mapping function to map the input
variable (X) with the output variable (Y).
• Here the input variable (X) is a dataset with n columnsx1, x2, x3…..xn, called
features/predictors
• The Output variable(Y) is a column called target/response/label. These are the
responses for the corresponding values of X
• Takes the form : Y= f(X)
• The goal is to find the function f
 Unsupervised Learning

• Unsupervised learning uses machine learning algorithms to analyze

and cluster unlabeled data sets.
• The goal of unsupervised learning is to find the structure/trends in data
• The entire dataset is the input variable X
• The output variable Y is the group/segment/cluster each
observation/data point can be categorized into. This is NOT GIVEN in
the dataset
• The goal is to find the output variable Y
Idea Exchange
Describe a scenario from your work/life in which it
would be useful to use business analytics .
What questions would you aim to answer using
analytics?

Copyright © SAS Institute Inc. All rights reserved.

Lecture Outline

1.1 Introduction
1.2 Basics of Business Analytics
1.3 Life Cycle of Business Analytics
Life Cycle of Business Analytics
 Life Cycle of Business Analytics

1. Business Understanding: stating your goal and defining your question(s),

2. Data Understanding : exploring the data, determining if the data that you have is appropriate

3. Data preparation : data cleansing and wrangling

4. Modeling: building models to summarize data, quantify the relationships, make predictions

and/or prescribe solutions.

5. Evaluation: Evaluating the model to see its fit

6. Deployment: Deploying the solution and prepare report for communicating the results to the

appropriate audience
Life Cycle of Business Analytics
 Life Cycle of Business Analytics : Business
Understanding
The purpose of Business Understanding is:
Defining what the objective of this business analytics initiative is
i.e. defining what question(s) do we want to answer ?

What is a good question:

• it has to be of interest to your audience
• It hasn't already been answered
• it's plausible ; you should be able to explain the mechanism for how things
work
• It has to be answerable
• It has to be very specific
Life Cycle of Business Analytics
 Life Cycle of Business Analytics : Data
Understanding
Data Understanding means:
• Acquiring the Dataset
• Importing the Dataset into the Analytics Platform/tool
• Exploring the Dataset
 Life Cycle of Business Analytics : Data
Understanding
Doing a basic sanity check:
• Look at the metadata, variable names and the number of rows.
• Check the edges of the data set : the top and the bottom few rows.
• Check to see is the data formatted correctly and if it ‘looks right’ ?
• Check the total number of observations, or your sample size.
• Try to validate it with at least one external data source
 Life Cycle of Business Analytics : Data
Understanding

Explore the data using Visualizations:

● Look at your data as a whole: make plots. Eg: a scatter plot to
see any obvious relationships, or a box plot, to compare 2
distributions
● Look at Central tendencies like mean, median,mode
● Look at variances like range, standard deviation, skewness,
kurtosis
● Look at the Quantile-Quantile (Q-Q) Plot to confirm normal
distribution
 0.2
0.18

Exploration 0.16
0.14
0.12
openvino-cpu

0.1 onnxruntime-
One picture = one thousand words! 0.08 CPUExecutionProvider

0.06 onnxruntime-
TensorrtExecutionProvider
0.04
0.02
0
1 2 4 8 16 32 64 128

700
600
500
400
300
200
100
0
1 2 4 8 16 32 64 128

fps avg lat med

 Correlation

Direction of the
association ?
Correlation

In this case, it is quite clear:

good weather increases
sales of ice-cream
Correlation may not be causation!

Is ice cream dangerous?

 Correlation is not causation

Shark attacks, bushfires and ice-cream sales are correlated because they are also correlated with hot weather!
 Survivor Bias

You study/analyze what you

can observe only!

https://en.wikipedia.org/wiki/Survivorship_bias#/media/File:Survivorship-bias.png
Life Cycle of Business Analytics
 Life Cycle of Business Analytics: Data Preparation

• Data must be accurate and complete! GIGO

rule:
Garbag
e In
Process
Garbag
e
Out
• Data may come from a variety of sources:
company reports, documents, employee interviews, direct measurement, statistical
sampling,…

• Data often needs to be cleaned!

Life Cycle of Business Analytics: Data Preparation

Preparing the Data : Cleansing and

Wrangling
• Formatting the Data
• Dealing with Missing Values
• Adjusting the Data
• Dealing with Outlier Data
• Dealing with Categorical Data
• Splitting the Data for Model Validation
Life Cycle of Business Analytics: Modeling
 Life Cycle of Business Analytics: Modeling

• Used for :
• Predicting the unknown(s)
• Recommending best solutions

All the COOL stuff we hear about!

 Life Cycle of Business Analytics: Modeling
• There are many, many, many modeling techniques available
• Sourced from statistics and computer science fields

• This course will cover some important modeling techniques

• Linear Regression
• Logistic Regression
• Decision Trees
• K-Means Clustering
• Others…
• Other MBAS9XXX courses cover more of these techniques in
more detail
 Life Cycle of Business Analytics: Modeling

• But the BIG question is :

• WHICH modeling technique do we choose for a problem
at hand ?
• Depends on:
• Nature of the problem
• Type of the problem
• Nature of the available data

• My research is based on this (more later)

Life Cycle of Business Analytics: Evaluation
 Life Cycle of Business Analytics: Evaluation

• Evaluation measures How “good” your model is:

• Different Measures for different models
• Common measures include:
• r square and adjusted r square
• T-statistic
• confusion matrix
• Receiver Operating Curve
• Akaike Information Criterion
• Bayesian Information Criterion
Life Cycle of Business Analytics: Deployment
 Life Cycle of Business Analytics: Deployment

Deployment involves moving the selected model into a production/

live environment
FOCUS of this subject
My Research : A Framework For Applying Analytics
to Business Practice
Link to Paper : https://aisel.aisnet.org/pacis2021/198/

Illustration

Vision:
To be the first choice car rental brand for business users in the
countries in which we operate

Goal:
To be a ‘premium brand’ car rental company, positioned alongside
companies such as Hertz and Avis

Objective:
Within six months, 10% increase in product sales

Contextual Questions:
(a)Given the current situation, what will be our predicted sales in next
six months ?
(b)how can we segment our customers to bring about an increase of
10% in our total sales in the next six months?
 A Framework For Applying Analytics to Business
Practice
• 1. Identify an ‘objective’ you wish to meet (determined by an overarching ‘goal’ that, in
turn, would realize the ‘vision’)
• 2. Identify one or more ‘contextual questions’ that you need to answer for your objective
• 3. For each contextual question:
• 3a. Determine the nature of the question (Descriptive, Predictive, Prescriptive)
• 3b.Determine the type of the question vis a vis its nature (Classification,
Regression etc.)
• 3c. Analyze the underlying dataset to identify which ‘constraints’ listed in the
matrix it satisfies
• 3d. In the matrix, look for row(s) that meets the ALL of the following conditions:
• -----matches the nature of the contextual question and
• -----matches the type of the contextual question and
• -----has a blank cell value for any constraint the dataset does NOT meet
• 3e. Mark the model(s) in the row(s) of the matrix thus obtained as the
recommended model(s) to answer your question.
• 3f. For each marked model:
• -----Use human expert judgment to assess the suitability of the model(s)
• -----Build custom solution(s) based on the recommended model(s)
• -----Assess the quality of the solution(s);
• 3g. Repeat step 3f to arrive at an optimal solution to answer the contextual
question; compare and/or combine if necessary
• 4. Repeat Step 3 to find optimal solution for all contextual questions determined in Step
2.
 Video Tutorials on SAS Viya

Basics of SAS Viya

https://video.sas.com/detail/videos/sas-visual-analytics_/video/4420977648001/overview-of-sas-vi
sual-analytics?autoStart=true&page=1

Getting Started with SAS Viya

https://video.sas.com/detail/videos/on-sas-viya/video/6106569850001/accessing-content-in-sas-d
rive?autoStart=true

https://video.sas.com/detail/videos/on-sas-viya/video/6106571457001/organizing-and-sharing-con
tent-in-sas-drive?autoStart=true
Questions?

Copyright © SAS Institute Inc. All rights reserved.