Course Outline Template

Indian Institute of Management Calcutta

Programme: MBA
Name of the Course: Business Analytics for Managerial Decisions (Earlier – Business Data Mining)
Core / Elective: Elective Cap (for Elective): 75
Credit: 3
Academic Year & Term: 2023-24, Term IV
Course Coordinator: Prof. Vimal Kumar M.
Email: IIMC id: [email protected]

Instructor: Prof. Vimal Kumar M.

Affiliation: IIM Calcutta
Email: IIMC id: [email protected]

Introduction to Course /Course Description

The increased digitization of business processes has resulted in explosive growth in the amount of structured and
unstructured data collected by firms. What can be learned from these mountains of data to improve decisions?
Business analytics is also a process for increasing the competitive advantage of a firm through mining and intelligent
use of available data in decision making. This course covers the concepts and fundamentals of business analytics and
machine learning with various business decisions as a backdrop. The emphasis of the course is on the application of
machine learning and associated techniques in real-world business problems.
Course Objective and Key-takeaways from the course
This course aims to provide students with
 an introduction to theoretical and practical aspects of core business analytics
 understand a wide variety of machine learning algorithms and applications
 hands-on experience in applying these techniques to practical real-word business problems using open-
source software as well as commercial software.
Given the applied nature of the course, time spent on the algorithms involved in each technique will be minimal.
Instead, the emphasis of the course is more on gaining a holistic understanding of the different techniques - how they
work, when to use the technique and how to interpret the results. While the same techniques are applicable to a
variety of business contexts, we will largely focus on datasets available in public domain.
Students who complete this course would be able to:
 understand the importance of data as a strategic resource
 articulate how a specific technique can solve a class of business problems
 translate a business problem into a analytics problem, identify the relevant data and select one or more
appropriate techniques to solve the specific problem
 use existing analytics software to mine a prepared data set, interpret and evaluate the results of data analytics
 interact competently on the topic of business analytics.

Programme Level Learning Goals (1 and 5) to be achieved:

1. Possess state-of-the-art knowledge of theory and practice in all functional fields of management and
the ability to think critically, and apply them to diagnose and find solutions to organisational
problems, even in unfamiliar or uncertain situations
 a. Attain capability to apply learning from experience and learning from observation aspects of data
analytics in decision making
5. Be able to appreciate the dynamics of Information Technology acquisition and absorption by present
day business organizations and develop ability to use modern IT enabled decision support tools for
improved understanding of business
a. Attain capability to appreciate the power of analytics and artificial intelligence-based approaches to
support business decisions
b. Attain capability to delve into potentials and pitfalls in data analytics based on IT enabled tools and
techniques

International Components in your course (if any): NA

Pre-requisites: None
Required Textbook(s):
1. Machine Learning, Springer, Zhi-Hua Zhou, 2021
2. Pandas for everyone: Python data analysis. Addison-Wesley Professional. Chen, D. Y. , 2017.

Technology enabled learning component

The pedagogy of the course is a mix of lectures, readings and hands-on training exercises for student teams. At least
25% of the sessions will be devoted to lab sessions where students will use analytics software on prepared data sets to
understand the scope and application of core techniques.
Session Plan
Session Topic Cases and Readings
1 Introduction to Required Reading:
business analytics  Keeping Up with the Quants 1. Why Everyone Needs
Analytical Skills.
 Chapter 1(Textbook)
Recommended Reading:
 Competing on Analytics. (HBR)
 Know What Your Customers Want Before They Do

2 Business Analytics and Required Reading:

Machine Learning  Keeping Up with the Quants 2. Framing the Problem.
Framework Recommended Reading:
 How to Spot a Machine learning Opportunity, even if You
Aren’t a Data Scientist? (HBR)
 What Every Manager Should Know about Machine Learning
(HBR)

3 Introduction to S/w Required Reading:

environment and Basic  Data Scientist’s Analysis Toolbox: Comparison
programming of Python, R, and SAS Performance (article)
 Chapter 1: Pandas data frame Basics [Textbook 2]
 Chapter 2: Pandas Data Structures [Textbook 2]
4 Data Cleaning, Required Reading:
Manipulation and  Chapter 3: Introduction to plotting [Textbook 2]
Visualization  Section II: Data manipulation [Textbook 2]
5 Dimensionality Required Reading:
reduction – Principal  Chapter 10: Dimension Reduction Methods [Textbook]
component analysis

6 Machine Learning Required Reading:

using Python - Logistic  A Refresher on Regression Analysis
Regression  Logistic Regression
 An introduction to machine learning with scikit-learn (article)
https://scikit-learn.org/stable/tutorial/basic/tutorial.html

7 Predictive Analytics- How to evaluate Predictive Models?

Evaluation Required Reading:
Chapter 2 (Textbook)

8 Predictive Analytics - Required Reading:

Decision Trees  Chapter 4 (Textbook)
Recommended Reading:
 Predictive Analytics: A Primer (HBR Web Article, Thomas H.
Davenport, 2014)

9 Predictive Analytics- Lab Session (Python)
hands on (Case handout and data will be provided in the Class)
(Decision Trees)

10 Predictive Analytics- Bagging, Boosting and Random Forest

Ensemble Methods Required Reading:
• Chapter 8 (Textbook)

11 Predictive Analytics – Required Reading:

Neural Networks  Chapter 5 (Textbook)
Recommended Reading:
Neural Networks in Business: Advantages and Perspectives-
InformationWeek 2017
https://www.informationweek.com/ai-or-machine-learning/neural-
networks-in-business-advantages-and-perspectives
12 Predictive Analytics- Lab Session (Python)
hands on (Neural  (Case handout and data will be provided in the Class)
Networks)

13 Clustering Required Reading:

 Chapter 9 (Textbook)
Recommended Reading:
 10 Interesting Use Cases for the K-Means Algorithm (Kaushik
Raghupati, 2018, dzone.com)
https://dzone.com/articles/10-interesting-use-cases-for-the-k-
means-algorithm
 14 Clustering – hands on Lab Session (Python)
(Case handout and data will be provided in the Class)

15 Associations/ Market Required Reading:

Basket analysis Amazon's recommendation secret (Fortune, 2012)

16 Associations – hands Lab Session (Python)

on  (Case handout and data will be provided in the Class)

17 Text Analytics Required Reading:

 Readings will be provided before the class
Recommended Reading:
 Six Ways to Find Value in Twitter's Noise.
 The Power of Social Media Analytics (CACM, 2014)

18 Text analytics-hands Lab Session (Python)

on (Case handout and data will be provided in the Class)

19 Practitioner insights Details for this session will be provided later.

20 Project presentation Student Presentation

Recommended Reading:
 Keeping Up with the Quants 5. Quantitative Analysis and
Creativity.

Evaluation Components

Students will be evaluated on individual and joint work throughout the course. The main components are as follows:

1. Project: Each student team of five members will need to complete a data analytics project. This will involve
selecting a business context, problem and accessing a relevant data set. The team will be responsible for applying
analytics to solve the business problem. Each team will submit a written report and will make a presentation to the
class for discussion.

2. Assignments: There will be case based group assignments based on the techniques covered in the class.

3. Quiz: Quizzes will be conducted to assess the theoretical and practical aspects of core business analytics
techniques. All quizzes will be surprise quizzes. No makeup quizzes will be provided. Best three quizzes out of
four will be considered for grading.

4. Final Exam: At the end of the term there will be a final comprehensive exam.
Evaluation Component Type (Individual/Group) Weightage Remarks
C1 Assignment Group 15 Case submissions
C2 Project Group 25
C3 Quiz Individual 10 * 3 = 30 4 quizzes (Best of 3)
C4 Final Exam Individual 30