Do Retailers Benet from Deploying Customer Analy cs?
Vishnu Charan Gudipalli
Journal of Retailing
Documented research on general posi ve rela onship between analy cs and rm
performance.
Using survey data from 418 top managers based in the Americas, Europe
R language application in Real world
Vishnu Charan Gudipalli
Origin of R language:
R is an implementation of the S programming language combined with lexical scoping
semantics inspired by Scheme. S was created by John Chambers while at Bell Labs. There ar
Global E-Business Strategies
Assignment Week 4
Gudipalli Vishnu Charan
University Strategies
1) Decide what types of strategy each pursues.
2) How does each university or college determine its customer groups and utilize its core
competencies to attract a
Week-4
What is the relationship between a firms customers and its
business-level strategy in terms of who, what, and how? Why is this
relationship important?
An organization's core competencies should be focused on satisfying customer
needs or preferences
R language applica on in Real world
Vishnu Charan Gudipalli
Origin of R language:
R is an implementa on of the S programming language combined with lexical scoping
seman cs inspired by Scheme. S was created by John Chambers while at Bell Labs. There are
s
15.097: Probabilistic Modeling and Bayesian Analysis
Ben Letham and Cynthia Rudin
Credits: Bayesian Data Analysis by Gelman, Carlin, Stern, and Rubin
1
Introduction and Notation
Up to this point, most of the machine learning tools we discussed (SVM,
Boost
Introduction to Statistical Learning Theory
MIT 15.097 Course Notes
Cynthia Rudin
Credit: A large part of this lecture was taken from an introduction to learning
theory of Bousquet, Boucheron, Lugosi
Now we are going to study, in a probabilistic framework
Kernels
MIT 15.097 Course Notes
Cynthia Rudin
Credits: Bartlett, Schlkopf and Smola, Cristianini and Shawe-Taylor
o
The kernel trick that Im going to show you applies much more broadly than
SVM, but well use it for SVMs.
Warning: This lecture is technical
Support Vector Machines
MIT 15.097 Course Notes
Cynthia Rudin
Credit: Ng, Hastie, Tibshirani, Friedman
Thanks: Seyda Ertekin
Lets start with some intuition about margins.
The margin of an example xi = distance from example to decision boundary
= yi f (xi
Convex Optimization Overview
MIT 15.097 Course Notes
Cynthia Rudin
Credit: Boyd, Ng and Knowles
Thanks: Ashia Wilson
We want to solve dierentiable convex optimization problems of this form, which
we call OPT:
minimize f (x)
n
xR
subject to gi (x) 0, i =
Decision Trees
MIT 15.097 Course Notes
Cynthia Rudin
Credit: Russell & Norvig, Mitchell, Kohavi & Quinlan, Carter, Vanden Berghen
Why trees?
interpretable/intuitive, popular in medical applications because they mimic
the way a doctor thinks
model discre
Boosting
MIT 15.097 Course Notes
Cynthia Rudin
Credit: Freund, Schapire, Daubechies
Boosting started with a question of Michael Kearns, about whether a weak
learning algorithm can be made into a strong learning algorithm. Suppose a
learning algorithm is o
Clustering
MIT 15.097 Course Notes
Cynthia Rudin and Seyda Ertekin
Credit: Dasgupta, Hastie, Tibshirani, Friedman
Clustering (a.k.a. data segmentation) Lets segment a collection of examples
into clusters so that objects within a cluster are more closely r
Fundamentals of Learning
MIT 15.097 Course Notes
Cynthia Rudin
Important Problems in Data Mining
1. Finding patterns (correlations) in large datasets
-e.g. (Diapers Beer). Use Apriori!
2. Clustering - grouping data into clusters that belong together - ob
Logistic Regression
MIT 15.097 Course Notes
Cynthia Rudin
Thanks to Ashia Wilson
Credit: J.S. Cramers The Origin of Logistic Regression
Origins: 19th Century.
Studying growth of populations and the course of chemical reactions using
d
W (t) = W (t)
dt
W
Rule Mining and the Apriori Algorithm
MIT 15.097 Course Notes
Cynthia Rudin
The Apriori algorithm - often called the rst thing data miners try, but somehow doesnt appear in most data mining textbooks or courses!
Start with market basket data:
Some importa
R for Machine Learning
Allison Chang
1
Introduction
It is common for todays scientic and business industries to collect large amounts of data, and the ability to
analyze the data and learn from it is critical to making informed decisions. Familiarity with
Na Bayes
ve
MIT 15.097 Course Notes
Cynthia Rudin
Thanks to Seyda Ertekin
Credit: Ng, Mitchell
The Na Bayes algorithm comes from a generative model. There is an imporve
tant distinction between generative and discriminative models. In all cases, we
want t
Bias/Variance Tradeo
A parameter is some quantity about a distribution that we would like to know.
Well estimate the parameter using an estimator . The bias of estimator for
parameter is dened as:
Bias(,) := E() , where the expectation is with respect to
2/17/12
K-NN
15.097 MIT, Spring 2012, Cynthia Rudin
Credit: Seyda Ertekin
K-Nearest Neighbors
Amongst the simplest of all machine learning
algorithms. No eXplicit training or model.
Can be used both for classifcaton and
regression.
Use XIs K-Neare
Problem Set 5: RSS Feed Filter
Handed out: Lecture 10.
Due: 11:59pm, Lecture 12.
Introduction
In problem set 5, you will build a program to monitor news feeds over the Internet. Your
program will filter the news, alerting the user when it notices a news s
6.00: Introduction to Computer Science and Programming
Problem Set 8: RSS Feed Reader, Part II
Handed out: Thursday, November 8, 2007
Due:
Part I Due: 11:00am Friday, November 16, 2007.
Part II Due: 11:00am Tuesday, November 20, 2007.
Introduction
In Part
Transforming Regional, National,
& International Institutions
Managing Transformations in Work, Organizations,
and Society
Todays Guests:
Annette Dixon, The World Bank
John Grierson, The Kennedy School of Government
Todays Objectives
Place Public Service
Managing Diversity for Business and Personal Success
Managing Transformations in Work, Organizations, and Society
Key Issues for Today
How do we gain competitive advantage from the diverse backgrounds, knowledge bases, and cultural experiences present in
Leadership Skills for the
21st Century
Managing Transformations in Work,
Organizations, and Society
Todays Objectives
Discuss Your Favorite Leaderswhat do they
have in common?
Examine the Sloan Leadership Model
Learn from leading-edge research and
prac
15.394
Designing and Leading the
Entrepreneurial Organization
MIT Sloan School of Management
Agenda
From Entrepreneurial to Professional
Management
Day 1: Founding and Building
SCORE! Educational Centers
Day 2: Expanding and Professionalizing
SCORE!
NAME _
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
Sloan School of Management
15.565 INTEGRATING INFORMATION SYSTEMS:
TECHNOLOGY, STRATEGY, AND ORGANIZATIONAL FACTORS
15.578 GLOBAL INFORMATION SYSTEMS:
COMMUNICATIONS & CONNECTIVITY AMONG INFORMATION
SYSTEMS
Spr
-1-
Name _
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
Sloan School of Management
15.565 INTEGRATING INFORMATION SYSTEMS:
TECHNOLOGY, STRATEGY, AND ORGANIZATIONAL FACTORS
15.578 GLOBAL INFORMATION SYSTEMS:
COMMUNICATIONS & CONNECTIVITY AMONG INFORMATION SYST
Managing Strategic Partnerships
Managing
Module 3
Managing Transformations in Work,
Organizations, and Society
Todays Participants
Todays
Students on campus and at a distance
Kaiser Permanente Partnership Leaders:
Leslie Margolin, Kaiser Permanente
Pe