Download Syllabus - Stevens Institute of Technology

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
Revised: Jan 23, 2015
Stevens Institute of Technology
Howe School of Technology Management
Syllabus
BIA 656
Statistical Learning and Analytics
Fall 2014
Germán Creamer, Babbio 637
[email protected]
6.15-8.45PM, Babbio 320
Office Hours:
M, 1.30 PM – 3 PM
Also by appointment
Course Room/Web Address:
Babbio 320/
http://www.stevens.edu/moodle
Overview
The significant amount of corporate information available requires a systematic and
analytical approach to select the most important information and anticipate major events.
Machine learning algorithms facilitate this process understanding, modeling and
forecasting the behavior of major corporate variables.
This course introduces statistical and graphical (machine learning) models used for
inference and prediction. The emphasis of the course is in the learning capability of the
algorithms and their application to several business areas.
Prerequisites: Basic course in probability and statistics at the level of MGT 620 or BIA
652 Multivariate data analytics.
Prerequisites
Admission requirements for the BI&A program.
Course Objectives
Students will:



Learn the fundamental concepts of statistical learning algorithms.
Explore existent and new applications of statistical learning methods to business
problems, and to generic classification problems.
Learn to solve analytical problems in groups and effectively communicate its
results.
Relationship of Course to Rest of Curriculum
Students will have the opportunity to explore the main concepts of statistical learning that
will be used in the applied modules of this program.
List of Course Outcomes:
By the end of this course, the students will be able to:
1.
Understand the foundations of statistical learning algorithms
2.
Apply statistical models and analytical methods to several business domains using
a statistical language.
3.
Recognize the value and also the limits of statistical learning algorithms to solve
business problems.
Additional learning objectives include the development of:
1.
Written and oral communications skills: students are required to communicate
properly during the class discussions and project class presentations. Homeworks and
project report should be presented “as if” they were submitted to a senior manager of a
major corporation.
2.
Solve a major analytical problem using large and heterogeneous datasets in a
group project and communicate its results in a professional way.
Pedagogy
The class will combine class presentations, discussions, exercises and case analysis to
motivate students and train them in the appropriate use of statistical and econometric
techniques.
Readings
Required Text
Foster Provost and Tom Fawcett, Data Science for Business, O’Reilly, 2013.
(code to get a discount on oreilly.com: AUTHD)
Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.
(Amazon.com sells the paperback version (2013))
Case Pilgrim Bank A (602104), Harvard Business School
You must register in the following website, buy the case and download related
documents: https://cb.hbsp.harvard.edu/cbmp/access/28615189
Optional Texts
Trevor Hastie, Robert Tibshirani and Jerome Friedman, The Elements of Statistical
Learning. Springer-Verlag, New York, 2010 (selected sections)
(downloadable at http://www-stat.stanford.edu/~tibs/ElemStatLearn/).
Christopher D. Manning, Prabhakar Raghavan and Hinrich Schutze, Introduction to
Information Retrieval, Cambridge University Press. 2008 (downloadable at
http://nlp.stanford.edu/IR-book).
R.O. Duda, P.E. Hart and D.G. Stork, Pattern Classification, John Wiley & Sons, 2001.
Tom M. Mitchell, Machine Learning, McGraw-Hill Series in Computer Science, 1997.
Vasant Dhar and Roger Stein. Seven methods for transforming corporate data into
business intelligence. Upper Saddle River: Prentice Hall. 1997.
Additional Free Texts
A. Rajaraman, J. Ullman Mining of Massive Datasets Book (very useful for big data
problems)
Mohammed Zaki and Wagner Meira Jr. Mohammed Zaki and Wagner Meira Jr. Data
Mining and Analysis: Fundamental Concepts and Algorithms (draft)
StatSoft Electronic Statistics Textbook (statistics and data mining)
Roberto Battiti and Mauro Brunato LIONbook: Learning and Intelligent Optimization
(introductory)
Assignments
The course will have a main project and 4 assignments/cases of data analysis. The
assignments must be submitted electronically through the course website before the
beginning of the class of the assigned day. Each student must submit his/her own report.
You should also include the Readme, log and code files if you used a script or wrote a
program. E-mail submissions will not be accepted. Each assignment has a value of 5
points.
Project
The project requires that participants build a decision support system (DSS) based on one
of the methods explored in this course. Each project must be developed by groups of
three students and they should present a project proposal at the middle of the semester.
Grades
Assignment
Assignments/cases
Team project
Participation
Final exam
Total Grade
Software
Grade %
20%
30%
10%
40%
100%
Python is the main software packages that will be used. You should participate in the
Python bootcamp offered by the school at the beginning of the semester.
Class policy
Late Policy: 1 point lost for each day late. No assignments accepted after 3 days.
Cooperation: You are allowed to discuss lecture and textbook materials, and how to
approach assignments.
You cannot share ideas in any written form: code, pseudocode or solutions. You cannot
submit someone else's work found through internet or any other source, or a modification
of that work, with or without that person's knowledge, regardless of the circumstances
under which it was obtained, copied, or modified. Of course, no cooperation is allowed
during exams.
Re-grades: If you dispute the grade received for an assignment, you must submit, in
writing, your detailed and clearly stated argument for what you believe is incorrect and
why. This must be submitted by the beginning of the next class after the assignment was
returned. Requests for re-grade after the beginning of class will not be accepted. A
written response will be provided by the next class indicating your final score. Be aware
that requests of re-grade of a specific problem can result in a regrade of the entire
assignment. This re-grade and written response is final; no additional re-grades or debate
for that assignment.
Ethical Conduct
The following statement is printed in the Stevens Graduate Catalog and applies to all
students taking Stevens courses, on and off campus.
“Cheating during in-class tests or take-home examinations or homework is, of course,
illegal and immoral. A Graduate Academic Evaluation Board exists to investigate
academic improprieties, conduct hearings, and determine any necessary actions. The
term ‘academic impropriety’ is meant to include, but is not limited to, cheating on
homework, during in-class or take home examinations and plagiarism.“
Consequences of academic impropriety are severe, ranging from receiving an “F” in a
course, to a warning from the Dean of the Graduate School, which becomes a part of the
permanent student record, to expulsion.
Reference:
The Graduate Student Handbook, Academic Year 2003-2004 Stevens
Institute of Technology, page 10.
Consistent with the above statements, all homework exercises, tests and exams that are
designated as individual assignments MUST contain the following signed statement
before they can be accepted for grading.
____________________________________________________________________
I pledge on my honor that I have not given or received any unauthorized assistance on
this assignment/examination. I further pledge that I have not copied any material from a
book, article, the Internet or any other source except where I have expressly cited the
source.
Signature _________________________
Date: _____________
Please note that assignments in this class may be submitted to www.turnitin.com, a webbased anti-plagiarism system, for an evaluation of their originality.
Course/Teacher Evaluation
Continuous improvement can only occur with feedback based on comprehensive and
appropriate surveys. Your feedback is an important contributor to decisions to modify
course content/pedagogy which is why we strive for 100% class participation in the
survey.
All course teacher evaluations are conducted on-line. You will receive an e-mail one
week prior to the end of the course informing you that the survey site
(https://www.stevens.edu/assess) is open along with instructions for accessing the
site. Login using your Campus (email) username and password. This is the same
username and password you use for access to Moodle. Simply click on the course that
you wish to evaluate and enter the information. All responses are strictly
anonymous. We especially encourage you to clarify your position on any of the
questions and give explicit feedbacks on your overall evaluations in the section at the end
of the formal survey which allows for written comments. We ask that you submit your
survey prior to end of the examination period.
COURSE SCHEDULE
8/25
9/8
Topic(s)
Introduction to data science and
data analytic thinking
Predictive modeling
9/15
From correlation to supervised
segmentation
9/22
Linear models
Reading(s)
PF, ch. 1 and 2
PF, ch. 3
B., 1.3, 1.4, 1.5
B, 1.6, 14.4
Optional reference:
HTF, ch. 9.2
PF, ch. 4
Hwks
9/29
Support vector machines
10/6
Model performance analysis
10/14
Graphical models
(Tuesday
)
10/20
10/27
11/3
11/10
11/17
Graphical Models
Relational learning: Bayesian
models
Application to marketing:
Targeting consumers
Sequential data (time series):
Markov decision processes:
-Reinforcement learning
-Time series
-Application to trading
Sequential data (time series):
Hidden Markov models
Mean variance decomposition
Combining models:
Ensemble methods
B. 3.1, 4.1.1-4.1.3, 4.3.2
B, 6.1, 6.2, 7.1
Optional references:
HTF, ch. 12
MRS, ch. 15
PF, ch. 5, 7 and 8
PF, ch. 9
B. 1.2
Optional references:
HTF, ch. 8.3-8.4
MRS, ch. 11, 13
B, Ch. 8
PF, ch. 11
B, 11.1, 11.2, 11.3
Case Pilgrim Bank 1st part
Hwk 1: Python
Project proposal
Hwk 2: classification
Hwk 3:
Case Pilgrim Bank I
discussion
B, 13.1
http://www1.icsi.berkeley.ed
u/~moody/MoodySaffellTN
N01.pdf
B, 13.2
Case Pilgrim Bank 2nd part
Hwk 4:
B, 3.2, 14.2-14.3
Case Pilgrim Bank II
PF, ch. 12
discussion
Optional references (click on
each):
ADTrees, Bagging, Random
Forests
11/24
Combining models.
12/1
Application to finance:
Mixed trading strategies
algorithmic trading
Final presentations
HTF, 8.7, 10.1, 15.1-15.3, 16
B, 14.1, 14.4, 14.5
Creamer, Model calibration…,
Quantitative.
Final project report
(12/5/2014)
PF: Provost and Fawcett, Data Science for Business
B: C. Bishop, Pattern Recognition and Machine Learning
Optional readings:
HTF: Hastie, Tibshirani and Friedman, The Elements of Statistical Learning. 2010
MRS: Christopher D. Manning, Prabhakar Raghavan and Hinrich Schutze, Introduction
to Information Retrieval, 2008.