You are on page 1of 5

COURSE 5: QUANTITATIVE TECHNIQUES FOR DECISION MAKING – II

(Machine Learning Techniques)

Credits: 3
Sessions: 20

OVERVIEW OF COURSE

Broadly Analytics rely on techniques based on Statistics and Machine Learning for developing
solutions. The purpose of this course is to cover Machine Learning techniques that are
commonly applied in the Industry. Currently industry has experienced tremendous growth in
Machine Learning applications. These are suited for challenging problems involving
nonlinearity, messy data and data with huge dimensionality. It will cover topics from
supervised learning, unsupervised learning and text mining.

Specific Objectives
 To educate students about supervised and unsupervised machine learning techniques.
 Help students to choose the right technique given a business problem.
 Evaluate the performance of these techniques in terms of accuracy and resource
requirements.
 Provide hands on exposure to R, which is one of the industry leading software for
analytics.

LEARNING OUTCOMES

The student should be able to understand and appreciate advanced techniques in Machine
learning and be able to apply such techniques to solve business problems. She will be able to
identify business situations where use of these techniques will provide the best solution. The
participant will be comfortable in using the R software and should be able to interpret R
output and communicate using business language.

SCOPE OF THE COURSE

 The study plan covering readings/handouts, classroom sessions, Group Work sessions
and Assignments corresponding to each session are outlined below.
 Quizzes or surprise tests would be conducted on the same. For post contact reading,
please refer to the handouts that will be provided at the end of each session.
 In addition, there will be regular analysis sessions using SAS - several sessions will be
dedicated for these hands-on exercises.
 The course is spread over 20 sessions. Each session will be of 90 minutes.
Following broad topics will be covered in the course:
 Unsupervised Machine Learning techniques like Association Analysis and
Cluster Analysis
 Supervised Machine Learning Techniques like Support Vector Machines,
Random Forest, Neural Net etc.
 Text Mining Techniques

STATISTICAL SOFTWARE
We will use R Studio for class exercises and exams. This software is available free of cost.
Follow the manual for installation of R and R Studio. Note that it requires many libraries to be
loaded as we cover different techniques in the class.

PEDAGOGIC TOOLS
The pedagogic tools of the courses are:
 Study Hand-outs
 Textbook
 Study Plan Folder
 SAS for hands-on exercises

The study encompasses the following:


 Classroom Lectures and Interactions
 Reading of Handouts being provided
 Practical Analysis session using R (bring laptop to class)

PREREQUISITES
 Quantitative Techniques (Statistics – 1)
 Business Analytics for Decision Making

TEXT BOOKS

Raschka S. Python Machine Learning, Packt Publishing, BIRMINGHAM.

Lantz B. 2016) Machine Learning with R, Packt Publishing, BIRMINGHAM.

Zhao Y. , R and Data Mining: Examples and Case Studies


(www2.rdatamining.com/uploads/5/7/1/3/57136767/rdatamining-book.pdf)
REFERENCE BOOKS

Berry , M . J . A . and Linoff , G . S . ( 2004 ) Data Mining Techniques for Marketing, Sales and
Customer Relationship Management , Wiley, Indianapolis, IN

John E. Hanke and Dean W. Wichern, Business Forecasting, Pearson, Prentice Hall.

Hand , D . J . , Mannila , H . and Smyth , P . ( 2001 ) Principles of Data Mining , MIT Press,
Cambridge, MA .

Chiu , S . and Tavella , D . ( 2008 ) Data Mining and Market Intelligence for Optimal
Marketing Returns , Butterworth-Heinemann, Burlington, MA.

Saxena, R and A Srinivasan, Business Analytics: A Practitioner's Guide, Springer.

Matthew Drake , The Applied Business Analytics Casebook: Applications in Supply Chain
Management, Operations Management, and Operations Research, FT Press.

Phillip E. Pfeifer, Ichiro Shiraki, Paul W. Farris, Identifying the VIPs at eTOTS, Source: Darden
School of Business.

Session Topic Reading


Overview of the Course/ Readings:- Chapter 1: Introducing Machine
Introduction to Machine Learning (BL)
Learning Algorithm
Video:- What is Machine Learning
Development, Growth, (https://www.youtube.com/watch?v=WXHM
comparison with Statistics _i-fgGo)
(Analytics), Supervised and
Unsupervised learning
Association Analysis Readings:- Association Analysis
Simplified.docx/ Chapter 8: Finding Patterns –
Market Basket Analysis Using
Association Rules

Data:- groceries.csv/ lastfm.csv/


transactions.csv
Cluster Analysis Readings:- Cluster Analysis for Identifying
Bank Customers Profile.pdf/ Cluster Analysis
Hierarchical and k-means Practical Example.pdf/ Chapter 9: Finding
clustering Groups of Data – Clustering with k-means

Data:- diabetes.csv/ cars.csv/ stocks.csv/


Assignment – 1 Business problem and dataset will be
provided.
k-NN Algorithm Readings:- Chapter 3: Lazy Learning –
Classification Using Nearest Neighbors (BL)/
Best way to learn kNN Algorithm using R
Programming

Data:- Prostate_Cancer.csv/ stocks.csv/


Support Vector Machines Readings:- Chapter 7: Black Box Methods –
Neural Networks and
Support Vector Machines.pdf/
svmbasic_notes.pdf/ SVM Tutorial.docx

Video: 16. Learning: Support Vector Machines


(https://www.youtube.com/watch?v=_Pwhi
WxHK8o&index=14&list=WL)

Data:- SVM using IRIS data.R/ SVM example


with data.R/ SVM bank marketing R.R
Random Forest Readings:- A Complete Tutorial on Tree Based
Modeling from Scratch.pdf/ Introduction to
Random forest.docx/

Data:- bank-additional-full.csv/ college rf.R/


Titanic.csv/ winequality-white.csv
Neural Networks Readings:- Chapter 7: Black Box Methods –
Neural Networks and
Support Vector Machines(BL)/
neuralnet.pdf/
Video:- Intro to Neural Network
(https://www.youtube.com/watch?v=DG5-
UyRBQD4)

Data:- boston R.R/ creditset.csv/ neural


network example.R
XG BOOST Readings:- 01_Introduction.pdf/ boost.pdf/
gradient_boosting.pdf/ mboost_tutorial.pdf

Data:- otto product xgboost.R/


overfitting.csv/ complete data xgboost.R/ IRIS
gbm R.R
Assignment – 2 Business problem and dataset will be
provided.
Text Mining Readings:- TextMining using R.pdf/ tm.pdf/
Basic Text Mining in R.docx/
Word Cloud/ Sentiment 1tmexercise.docx
Analysis/ Twitter Analysis
Manual:- setting up R for twitter analysis
(https://www.r-bloggers.com/setting-up-the-
twitter-r-package-for-text-analytics/ )
Data:- gs.txt, gs2.txt, gs3.txt/ 1tm exercise.R/
2tm exercise.R

Assignment – 3 Business problem and dataset will be


provided.
Assignment Presentation/
Review

You might also like