Applied Machine Learning and its use in Analytics
Last date for registration: 29 Jun, 2015
End Date : 11 Jul, 2015
Early Bird Discount Date : 18 Jun, 2015
Residential Early Bird Fee(excluding GST) : Rs. 61,750
Non-Residential Fee(excluding GST) : Rs. 57,500
Non-residential Early Bird Fee(excluding GST) : Rs. 54,625
Machine learning is a new data analysis domain which is increasing becoming popular to automate advanced analytical model building. Using new algorithms that iteratively updates from data, machine learning allows computers to find hidden patterns and insights from the data. Due to the growing volumes and varieties of the available data, computational processing is cheaper and more powerful and thus the increasing demand of knowledge of machine learning. Machine learning is getting used today in fraud detection, web search, text mining, reccommender’s system, network intrusion prediction etc etc.
This course is for those who are at the advanced stage of analytics profession across various industries.
Prof Pulak Ghosh: Prof. Ghosh is a professor in Quantitative Methods & Information Systems at IIM Bangalore. He is Ph D. in Statistics, Oakland University, Michigan (2003), and B.Sc. & M.Sc. in Statistics, University of Calcutta (1998). He works in the domain of Big data, Machine learning, Marketing analytics, Business analytics, Banking analytics, Econometrics and Bayesian Statistics.
Prof Pulak Ghosh is a member of the Data Privacy Advisory Group of Global Pulse- the UN Secretary General’s Big Data Initiative. Ghosh, the only expert from India, along with international experts, will advocate responsible use of Big Data for sustainable development and humanitarian value creation. Prof. Ghosh is also an academic fellow at the Center for Advanced Financial Research and Learning of Reserve Bank of India, Advisor of analytics to the State Bank of India. He is recently named as the top 10 most influential analytics leader.
A judicious mix of lectures, class-discussions, research, cases, videos, and exercises will be used.
Programme Charges*
Residential: Rs. 65,000/-(subject to availability of rooms on campus)
Non-residential: Rs. 57,500/-
Early bird cut-off date: 18-Jun-2015
Residential: Rs.61,750/-
Non-residential: Rs.54,625/-
Please Note * Please add service tax at prevailing rates to the programme fee. Group discount (10%) may be availed for a group of 5 or more participants from an organization for a programme, on upfront payment before the start of the programme.
All enrolments are subject to review and approval by the programme director. Joining Instructions will be shared with the organization if sponsored or to the participants on selection. Kindly do not make your travel plans unless you receive the letter from IIMB.
A certificate of participation will be awarded to the participants by IIMB.
Registration
Please logon to IIMB website www.iimb.ac.in/eep for registering online. Do feel free to get back to us if you should have any clarification.
Executive Education Programmes
Indian Institute of Management Bangalore
Bannerghatta Road, Bangalore 560 076
Phone: +91 - 80 - 2699 3264 / 3475
Fax: +91 - 80 - 2658 4004 / 4050
E-mail: openpro@iimb.ernet.in
Session 1: Introduction and Overview: What is Machine learning : supervised and Unsupervised learning
Will learn how machine learning transform the modeling and data understanding in analytics that transform organizational decisions, Making, pitfalls, and payoffs of the business and the three pillars of Analytics: descriptive, predictive, and prescriptive
Readings:
- HBR Case Study: The Big Idea: The Next Scientific Revolution
- Competing on Analytics – HBR article (2006)
- HBR article: 10 Insights: A First Look at The New Intelligent Enterprise Survey on Winning With Data & 10 Data Points: Information and Analytics at Work
Session 2, 3 & 4: Regression and Classification
- Simple Regression, Multiple Regression, Panel Regression
- Logistic Regression, Panel Logistic Regression, LDA, QDA, KNN
Will learn to answer questions such as: “what do data tell me?” The process involves “different kind of regression, classification” or unearthing useful patterns and variability in data in order to predict customer preferences. What is likely to occur based on what happened in the past? "What if?" scenario analysis to evaluate decisions and improve decision-making. Predictive analytics help in exploring all relevant data quickly and easily, uncover hidden opportunities, identify key relationships and make more precise decisions faster than ever before.
Readings :
- Case Study on “Harman Foods, Inc” –HBS
- Case Study on German credit applications
- Forecasting with Regression Analysis-- HBS
Session 4 & 5 : Resampling Methods and Dimension Reduction Methods
Resampling methods are an indispensable tool in modern Statistics and it involves repeatedly drawing samples from a training set and refitting a model of interest on each sample in order to obtain additional information. Dimension reduction techniques can help in yielding better prediction accuracy and model interpretability.
Session 5 & 6:Moving Beyond Linearity
Will learn methods beyond linear regression. Polynomial Regression, Splines, Generalized Additive Model etc.
Session 7 & 8: Tree-Based Methods and Support Vector Machines
We will learn various decision tree algorithm, Bagging, Random Forests and Boosting. Will also introduce the concept of Maximal marginal classifier through simple support vector machines and classifier.