Вы находитесь на странице: 1из 6

Big Data

Start Date: August 4, 2019


Apply by: May 27, 2019
Analytics
(Batch 4)
(Classes Conducted On-Campus)
collect data from different functions, and utilize versatile
forms of storage to query the data in real time. Finally, the
company must have professionals on board, who can
work with as well as enhance software that helps them
make sense of the results.

In an article1 based on a survey of nearly 3000 executives, Program Objectives


MIT Sloan Management Review has reported that there
This program is designed to equip its participants
is a striking correlation between a firm’s analytical
sophistication and its competitive performance. The with an in-depth knowledge of Big Data Analytics
biggest obstacle to adopting big data analytics is the (BDA). We will use real case studies and hands-
lack of knowhow among key personnel, whose decisions on demonstrations to illustrate the applications
will impact outcomes. The Big Data Analytics program of key concepts. At the end of the course, the
draws from a diverse mix of statistics and operations participants will be able to:
research, machine learning, deep learning, algorithm
1. Appreciate the emergence of business
design, and systems engineering. It provides an in-depth
knowledge of big data techniques, and their applications analytics and big data as a competitive
in improving business processes and decision-making. strategy.
2. Analyze datasets by applying techniques

Big Data Analytics from statistics, operations research,

Certificate Programme A triad of terms captures the essence of “big data”:


machine learning, deep learning, network
analysis and data mining.

on Big Data Analytics volume, velocity and variety. The volume and pace at
which data is created can challenge existing computing
3. Process unstructured data such as social
media messages and machine generated
BATCH 4 (2019-20) infrastructure. For example, every flight of a Boeing 777 clickstream logs.
can generate up to 1 terabyte (~1000 gigabytes) of data. 4. Have a working knowledge of languages,

The
Making sense of this data is imperative for decision-
discipline and practice of platforms and tools that support statistical
making and troubleshooting.
management is rapidly becoming analysis and visualization (R/Python),
evidence-based. Consider the Organizations large and small are forced to grapple distributed computing (Hadoop/Spark) and
area of marketing, which once idolized freewheeling with problems of big data, which challenge the existing network analysis (Gephi).
thinkers, elevating them to the status of visionaries tenets of data science and computing technologies. 5. Apply the theories, techniques and tools
for their wild, imaginative inputs. Although these Straightforward tasks such as interpreting descriptive to solve problems from industry sectors
inputs were helpful at the time, functions such as statistics have their share of issues. We begin to question such as manufacturing, services, retail,
advertising and sales today have become software- the utility of summary measures and diagrams. software, banking and finance, sports,
assisted. Cold, calculative reasoning is required to
Algorithms that work well on “small” datasets crumble pharmaceuticals, and aerospace.
justify any business decision. In this digital world,
there is little utility in seeking validation via sound when the size of the data extends into the terabytes.
bytes from the person on the street. Want to know Time series techniques must be revamped to handle
how well a disruptive ad campaign is faring? Gather streaming data in continuous time. Social media
what thousands of fans say about it on Twitter and messages are unstructured, and have data formats that
Facebook, and apply a host of analytical techniques. are unfit to be represented by traditional databases.
The role of analytics in solving business problems has While these may appear to be difficult problems, there
increased manifold in recent years, causing a spike has been tremendous progress in analyzing such data.
Columnar databases have significantly boosted query
In God We Trust, in the demand for trained professionals. In his book
titled, “Competing on Analytics: The new science of speeds. File systems can seamlessly distribute datasets
on multiple hard drives, and facilitate analytics on them
winning”, Thomas Davenport claims that a significant
All Others Must proportion of high-performance companies employ
personnel with high analytical skills. Another study
in real time. Finally, the free and open source nature
of big data platforms promotes their rapid adoption.

Bring Data reveals that close to 60% of organizations do not


have the requisite information for decision-making.
1 M S Hopkins, S LaValle, F Balboni, N Kruschwitz and R
Shockley, “10 Insights: A First look at The New Intelligence En-
W Edwards Deming Tackling this issue warrants a company-wide effort to terprise Survey on Winning with Data”, MIT Sloan Management
Review, Vol. 52, No. 1, 21–31, 2010.
Module 1: Module 2: Module 3:
Prior Preparation Foundations of Data Predictive Analytics Machine Learning
Make no mistake: the program will be relatively fast
Science (6 days) (6 days) (Module 1)
paced to help us achieve the goals that we have set.
To help BDA participants understand and assimilate The process of fact-based decision-making requires Predictive analytics models predict the occurrence of future events (3 days)
pre-requisite material, we recommend they go through managers to know how to summarize, analyze, and such as customer churn, default in loan repayment etc. based on
Contents
IIMB’s MOOCs on edX.org titled Statistics for Business interpret data, as well as to communicate the results historical data. In many business problems, we deal with data on
several variables, sometime more than the number of observations. • Overview of supervised and unsupervised
– I & II.2 These MOOCs cover topics in single variable using data visualization. Some of the techniques that we methods
statistics having to do with data analysis, visualization, introduce here apply to “small” datasets: they will have Regression models help us understand the relationships among
these variables, and how these relationships can be exploited • Evaluating models with RMSE, precision,
probability, Bayes rule, decision making with conditional to be suitably modified to handle large data volumes.
to make decisions. The primary objective of this module is to recall, sensitivity and specificity.
probability, random variables, Binomial, Poisson and Along the way, we shall introduce the participants to two
understand how regression and causal forecasting models can • Validating models, problem of overfitting,
Normal distributions, and Monte Carlo simulation. The platforms for machine learning: R and Python.
be used to analyse real-life business problems such as prediction, regularization
courses supply abundant examples from industry, and
Contents classification and discrete choice problems. • Decision trees, k-nearest neighbor (KNN)
also provide illustrative tutorials to some of the tools we
method, Naïve Bayes classifier
shall be using: spreadsheets and R. • Foundations of Data Science: Probability and The focus will be on case-based practical problem-solving using • Unsupervised methods: cluster analysis,
We shall skip covering the above material in the Random variables predictive analytics techniques to interpret model outputs. The association rule mining
classroom. Instead, the forums on the MOOCs will help • Exploratory data analysis with R and Python. participants will be exposed to software tools such as MS Excel,
address any concern you may have. • Data visualization – techniques and principles. The R, SPSS, and SAS and how to use these software tools to perform
grammar of ggplot2 regression, logistic regression and forecasting. Module 4:
Program Design • Handling geospatial datasets Contents Prescriptive
The course consists of ten modules and a project. All
• Normality. Sampling and central limit theorem.
Estimation and hypothesis testing., maximum
• Regression model building framework: Problem definition, Data Analytics –
Pre-Processing; Model Building; Diagnostics and Validation
modules are application-oriented, without compromising
on theoretical aspects. The following sections go over
likelihood estimation
• Simple linear regression: Coefficient of determination,
Optimization
the individual modules and their contents.
• Matrix algebra. Eigenvalues and diagonalisation.
Significance tests, Residual analysis, Confidence and
(3 days)
Singular value decomposition.
Prediction intervals Optimization models are core tools used in
Case-based teaching shall be employed wherever • Bayes theorem
possible, across all modules, with cases studies sourced • Multiple linear regression: Coefficient of multiple coefficient prescriptive analytics and are used in arriving at
• Concepts of multivariate calculus of determination, Interpretation of regression coefficients, optimal or near optimal decisions for a given set of
from IIMB, Harvard Business School (HBS), Darden,
Ivey, and Kellogg. Categorical variables, heteroscedasticity, Multi-collinearity, managerial objectives under various constraints.
Case Studies:
outliers, Autoregression and Transformation of variables, Optimization techniques such as gradient decent
The program shall have a special focus on business 1. Central Parking Solutions Private Limited (IIMB Case); Regression Model Building plays an important role in many machine learning
analytics as practiced in India. A significant proportion 2. A Dean’s Dilemma: To Admit or Not to Admit (IIMB algorithms. The objective of the module is to
• Logistic and Multinomial Regression: Logistic function,
of the cases that we use in the program have been Case) 3. Analytics in HR – Predicting Job Acceptance acquaint participants with the construction of
Estimation of probability using logistic regression, Deviance,
published by IIMB faculty on the Harvard Business (IIMB Case) mathematical models for managerial decision
Wald Test, Hosmer Lemshow Test, Classification table, Gini
Publishing site. A few of them are published by the alumni situations and use freely available Excel Solver
co-efficient.
of other business analytics programs at IIMB, based on and OPL to obtain solutions and interpret the
• Forecasting: Moving average, Exponential smoothing, Casual
their project work. results.
Models, ARIMA
Contents
• Application of predictive analytics in retail, direct marketing,
health care, financial services, insurance, supply chain, etc. • Introduction to Operations Research
(OR), linear programming (LP), formulating
Case Studies: decision problems using linear programming,
Pricing of players in the Indian Premier League (IIMB Case), Package interpreting the results and sensitivity analysis.
Pricing at Mission Hospital (IIMB Case), Colonial Broadcasting Concepts of shadow price and reduced cost.
Company (HBS Case), Pedigree vs Grit: Predicting Mutual Fund • Multi-period LP models. Applications of linear
Manager Performance (Kellogg Case), Breaking Barriers – Micro- programming in marketing and financial
Mortgage Analytics (IIMB Case), A game of Two Halves: In-Play planning
betting in Football (IIMB Case); HR Analytics – Predicting Probability • Non-linear programming – Gradient descent
2 https://www.edx.org/course/statistics-business-i-iimbx- of Renege (IIMB Case), Predicting Demand for Food at Apollo method
qm101-1x Hospital (IIMB Case);
regression, Dynamic regression, Multinomial logistic
Module 5: Module 7: regression, quantile regression
Big Data Eco-System Deep Learning and AI • Dimension Reduction – Ridge, LASSO, Elastic Net Why IIMB?
(3 days) (3 days) IIMB is one of the first Institutes in the world to offer
Module 10:
We are now ready to scale up the solution techniques This model focuses on Introduction to Artificial Intelligence Business Analytics certificate course since 2010. IIMB
with the help of big data platforms. In this module, we Contents
Advanced Big Data faculty members have published more than 20 analytics
shall use Spark extensively to set up and solve problems
• Introduction to neural networks, rule based expert Analytics-II (2 days) case studies at the Harvard Business Publishing a
record for any Indian Institute. Students admitted to this
involving large datasets.
systems In this module, we introduce the participant to advanced group come from different organizations and different
Contents • Introduction to artificial neural networks (ANN), big data technologies and algorithms and their geographical locations across the world making it a great
neuron as computing element, perceptron, applications in business domains.  learning experience for students. Students get a chance
• Big data: changes in approach and analysis
McCullogh-Pitts model, back propagation algorithm, Contents to work on real world problems as part of the course.
• The Hadoop ecosystem: Storing and managing
multi layer neural networks
large scale structured and unstructured data • Bayesian regression; Bayesian optimization IIM Bangalore has invested heavily in building big data
• Deep learning algorithms, Convolutional networks,
• Large scale data ingestion using Sqoop, Flume, • Text Mining, Topic Modelling processing capabilities. Participants shall be working
Recurrent nets, auto encoders
NoSQL and Kafka • Deep learning platform: H20.ai, GraphLab, Tensor • Model averaging with a powerful cluster of 12 nodes on the campus. While
• Apache Spark with interfaces in Python and R Flow • Blockchains and Fintech we shall explore the cloud route to big data analysis, we
• Computing platforms for big data, which include • Introduction to Big data for Policy shall be able to explain to the participants aspects of the
technologies required to run big data computations.
Google Cloud, BigQuery, Amazon AWS, Microsoft Module 8:
Azure and MongoDB
Advanced Machine Course Evaluation:
Learning (3 days) Eligibility Criteria
Module 6: The participants will be evaluated through take-home
This module introduces the participant to machine
Machine Learning learning algorithms such ensemble methods and
assignments and a project work. At the end of each
module, the participants will be given a take-home
and Selection Process:
penalised regression, clustering, text analytics, spatio-
(Module 2) temporal analysis, association rule mining and Monte
assignment that should be completed and submitted The participants should have a Bachelor degree in
within 4 weeks. engineering/science/commerce or arts with mathematics
(3 days) Carlo simulation.
as one of the subjects during their Bachelor’s program.
This module introduces the participant to machine Contents Preferable work experience is 3 years, in exceptional
learning algorithms such as bagging and boosting, • Tree-based methods—Decision tree, Bagging, cases applicants with less than 3 years are admitted
recommender systems, clustering, text analytics, spatio- Random forest, boosting, stochastic gradient into the program. It is essential that the applicants have
temporal analysis, association rule mining and Neural boosting
Course Project: programming knowledge.
Networks. • Support vector machine Each participant should carry out an individual
• Neural Network project for 4 months based on a real-life problem/
Contents
• Deep Learning dataset. IIMB encourages students to publish
• Introduction to machine learning, different types of
machine learning algorithms.
• Survival Analysis cases studies based on their course project. Selection Process:
• Recommender Systems, Collaborative Filtering: Module 9: After submitting their applications online, candidates shall
Cosine Similarity, Jaccard Coefficient. Advanced be short-listed for an online test. Questions on the test
recommender system.
Advanced Big Data will examine the candidate’s grasp of basic quantitative
• Bootstrap Aggregating (Bagging), Random forest, Analytics-I (3 days) Who should attend? concepts. As prior preparation, the candidates are
Adaptive boosting, gradient boosting This certificate program will equip the participants suggested to enrol in a beginner edX course dedicated
Big-data is defined using the volume of the data, the
with a large suite of analytical tools, as well as prepare to Statistics and complete the exercises:
• Support vector machine and Neural Network velocity at which the data is created, and the variety in
the data. Sources of big-data include social networks, them for corporate roles in analytics based consulting https://www.edx.org/course/statistics-business-i-iimbx-
telecom and mobile services, healthcare and public in marketing, operations, supply chain management, qm101-1x-0
systems and machine generated data. In this module, finance, insurance and general management in various
we introduce the participant to big data technologies, Based on a combination of test score, past academic
industries. The course is suitable for those who are
and explore the challenges.  performance, quality of work experience and fit for an
already working in analytics, and wish to enhance their
analytics career, candidates will be called in for a face-
Contents knowledge. We also welcome participants with a strong
to-face interview.  The online test and interviews will be
analytical aptitude, who would like to start their career in
• Advanced regression methods, Zero-inflated conducted in succession during June 2019.
analytics.
Tentative Course schedule: Project:
Students are expected to do a live project as part of this
Program Schedule course. The project report should be submitted by 15
May 2020. The participants have to submit the project
Module Dates Venue proposal by 31 January 2020. The projects will be
supervised by an IIMB faculty member.
1 Foundations of Data Science August 4 - 9 IIM Bangalore

2 Predictive Analytics September 8 - 13 IIM Bangalore


Program Directors:
3 Machine Learning (Module 1) October 4 - 6 IIM Bangalore
Professors U Dinesh Kumar,
4 Prescriptive Analytics - Optimization November 15 - 17 IIM Bangalore Shankar Venkatagiri and
Pulak Ghosh
5 Big Data Ecosystem November 29 - December 1 IIM Bangalore
For any queries contact: Professor Shankar Venkatagiri (Email: shankar@iimb.ernet.in)
6 Machine Learning (Module 2) December 13 - 15 IIM Bangalore

7 Deep Learning and AI January 17 - 19 IIM Bangalore Program Delivery:


8 Advanced Machine Learning February 7 - 9 IIM Bangalore The program will be conducted live in the
classroom at IIMB.
9 Advanced Big Data Analytics - I February 28 - March 1 IIM Bangalore

10 Advanced Big Data Analytics - II March 14- 15 IIM Bangalore


Program Fee:
Class timings: 9 am to 5.15 pm (IST) The programme fee is Rs. 6,25,000/-+ GST
(applicable rates) per participant, payable in three
installments as per the following schedule.

Rs. 2,50,000/- + Applicable GST : I installment on


admission

Rs. 2,50,000/- + Applicable GST : II installment on


or before 04 November 2019

Rs. 1,25,000/- + Applicable GST : III installment


on or before 02 January 2020

Award of Certificate:
A certificate of completion will be awarded by IIMB
to the participants at the end of the program, upon
successful completion of the programme satisfying IMPORTANT DATES
the program requirements.
Application Deadline: 27 May 2019
Online Test: 15 - 16 June 2019
Alumni: Face-to-Face Interview: 29 - 30 June 2019
Successful completion of the programme also entitles Announcement of Decision: First week of July 2019
participants to be admitted to IIM Bangalore Alumni
Association Course Commencement: 4 August 2019
The Indian Institute of Management Bangalore (IIMB) is a leading graduate school of management in Asia.
Established in 1973, IIMB today offers a range of post-graduate and doctoral level courses as well as executive education
programmes. With a faculty body from amongst the best universities worldwide, IIMB has emerged as a leader in the
area of management research, education and consulting. IIMB’s distinctive feature is its strong focus
on leadership and entrepreneurial skills that are necessary to succeed in today’s dynamic business
environment. VISION
The Post Graduate and Doctoral Programmes offered by IIMB: To be a global,
renowned academic
• 2-year Post Graduate Programme in Management (PGP) institution fostering
• 1-year Executive Post Graduate Programme in Management (EPGP) excellence in management,
• 2-year weekend Post Graduate Programme in Enterprise Management (PGPEM) innovation and
• 1-year Post Graduate Programme in Public Policy & Management (PGPPM) entrepreneurship for
• Fellow Programme in Management (FPM, doctoral programme) business, government
and society
IIMB has obtained the European Quality Improvement System (EQUIS) accreditation awarded
by the European Foundation for Management Development (EFMD). IIMB has been ranked No.
2 in the India Rankings 2018 in the Management Education category under the National Institutional
Ranking Framework (NIRF) by the MHRD. IIMB has been ranked among the Top-60 global schools by
the Financial Times Executive Education Rankings 2018.

REGISTRATION
The organizations interested in nominating their
employees and individuals interested in the
programme may apply online.

Arun KR
Executive Education Programmes
Indian Institute of Management Bangalore
Bannerghatta Road, Bengaluru 560 076
Phone: +91 - 80 - 26993381, 26993660
Fax: +91 - 80 - 2658 4004
Email: arun.eep@iimb.ac.in
Web: www.iimb.ac.in/eep

Facebook: http://on.fb.me/1zWioPp
YouTube: https://bit.ly/1zWi8Qk
LinkedIn: http://linkd.in/1G31q38
Twitter: https://bit.ly/2LuODNn
Instagram: https://bit.ly/2koNKK3
Blog: http://blog.iimb.ac.in/
INDIAN INSTITUTE OF MANAGEMENT BANGALORE
Bannerghatta Road, Bengaluru 560 076

Вам также может понравиться