Course Outline

RUTGERS UNIVERSITY

SCHOOL OF BUSINESS - CAMDEN

FALL 2022: MACHINE LEARNING APPLICATIONS FOR BUSINESS

Instructor: Ram Gopalan

E-mail: ram.gopalan@rutgers.edu

Course materials: Please access the CANVAS course site for power point materials and lecture videos.

LEARNING GOALS:

The focus of the course will be to introduce basic concepts in machine learning and data-analytic thinking to

students, with an applied business orientation. Students will understand how to use data to competitive

advantage and to build and evaluate models for decision-making. Companies today have access to vast

amounts of data from their business operations. Data Science is the craft of extracting patterns from this data

and using available information for competitive advantage. This course represents an introduction to data

science and data analytic thinking. Students will learn to leverage data to answer business questions relating

to classification tasks (e.g., will this credit card prospect default or not?), prediction (e.g., how much will this

customer spend/year?) and similarity profiling (what do my most profitable customers look like?).

TEXTBOOKS/COURSE PACK (HBS):

1. Data Science for Business, by Foster Provost and Tom Fawcett, O'Reilly publishing,

ISBN: 978-1-449-36132-7.

2. Introduction to Machine Learning with Python, A Guide for Data Scientists, by Andreas. C.

Muller and Sarah Guido, also O'Reilly publishing, ISBN: 978-1-449-36941-5.

3. Course pack to be obtained from this: https://hbsp.harvard.edu/import/961465

COMPUTER SOFTWARE:

The course is very hands-on. Students will be expected to download and use free, open-source packages

for machine learning and to use them for model-building. (e.g., the Anaconda package).

Office hours: Wednesdays 4 – 5 p.m., or by appointment.

COURSE PREREQUISITES: Students must be comfortable installing packages independently and

navigating in a computing environment. Specifically, you must be able to complete installation of

the Anaconda package on your own as demonstrated in the PREREQUISITES Module. Important:

The course assumes the student already has some basic familiarity with the Python programming

language as well as a working knowledge of Jupyter notebooks. See the PREREQS module for

greater detail.

COURSE REQUIREMENT AND GRADES:

Student grades will be based upon the following items:

(1) Exam I (15%)

(2) Exam II (20%)

(3) FINAL EXAM (25%)

(4) Team case analyses (10% = 2*5%)

(5) Team programming assignment (15%).

(6) Team data science project proposal (10%)

(7) Individual class participation and team work (5%).

More detail on the deliverables will be provided during the semester. Teams for these projects will be

assigned. Moreover, the team assignments must be uploaded on CANVAS by the deadline

stipulated, so that the system can verify that you submitted the assignment prior to the deadline.

Note: Class participation and team effort (5 %):

Class participation grades will be assigned upon the following criteria: a) displaying a positive attitude and

demonstrating a “solutions-focus” for any problems encountered during the course of the semester; b)

helping fellow students succeed; and c) displaying a strong commitment to your team’s success and discipline

in terms of showing up for team meetings. You will have an opportunity to submit peer evaluations for

all your team mates at the end of the semester!

Academic Integrity:

“Academic integrity requires that all academic work be wholly the product of an identified individual or

individuals. Joint efforts are only legitimate when the assistance of others is explicitly acknowledged… The

principles of academic integrity entail simple standards of honesty and truth. Each member of the university

has a responsibility to uphold the standards of the community and to take action when others violate

them…Students are responsible for knowing what the standards are and for adhering to them. Students

should also bring any violations of which they are aware to the attention of their instructors.”

Students are

expected to know, understand and adhere to the policies on academic integrity outlined above. Procedures

for violation of these policies outlined in the University Code of Academic Conduct will be followed. IN

THIS CLASS ALL EXAMS ARE INDIVIDUAL – YOU CANNOT COLLABORATE ON ANY

ONLINE TEST.

Rutgers University Code of Academic Conduct, taken from the Student Advising Handbook -

http://camden-sbc.rutgers.edu/CurrentStudents/students/advising.pdf.

TENTATIVE COURSE OUTLINE (NOTE: Instructor may make modifications to this schedule as necessary!):

Module

Topic(s) for Module

Assigned Reading

Module 1

Data analytic thinking and data science solutions for business

Provost-Fawcett (PF): Chapters 1,2

Module 2

Supervised learning and classification via Decision Trees and the

Nearest Neighbors algorithm

Ch3: PF

Ch1: Muller-Guido (MG)

Case: Stitch Fix (5%)

Module 3

Decision Trees (continued)

Ch3: PF, Ch2: MG

Module 4

Regression/Logistic Regression

Ch4: PF, Ch2: MG

Module 5

Regression/Logistic Regression

Ch4: PF, Ch2: MG

TEST 1: 15%

Module 6

Understanding overfitting and cross-validation for models

Ch5: PF, Ch5: MG,

Case: Byte-Dance (5%)

Module 7

Unsupervised learning, similarity and clustering

Ch6: PF, Ch3: MG

Module 8

Unsupervised learning, similarity and clustering

Ch6: PF, Ch3: MG

Module 9

Decision analytic thinking: what is a good model?

Ch7: PF, Ch5: MG

Module 10

Visualizing model performance: ROC curves

Ch8:PF, Ch5: MG

Programming Assignment due.

(15%)

Module 11

Evidence and probabilities, Naive Bayes

Ch9: PF

TEST 2: 20%

Module 12

Representing and mining text

Ch10: PF, Ch7: MG

Module 13

Representing and mining text

Ch10: PF, Ch7: MG

Data science project proposal

PPT due(10%).

Module 14, Final Exam

TBD (25%)

NOTE: Final exam (25%)

Ch 13,14: PF

Rutgers email - ALWAYS USE YOUR RUTGERS EMAIL ADDRESS:

All communications to students will be done using the Rutgers email address provided to you. Please forward your

Rutgers email to your personal email if necessary. Not checking your Rutgers email is not an excuse for missing

any communications.

Disability Services/Accommodations

Rutgers University welcomes students with disabilities into all of the University's educational programs. In order to

receive consideration for reasonable accommodations, a student with a disability must contact the approp riate disability

services office at the campus where you are officially enrolled, participate in an intake interview, and provide

documentation: https://ods.rutgers.edu/students/documentation-guidelines.

If the documentation supports your request for reasonable accommodations, your campus’s disability services office

will provide you with a Letter of Accommodations. Please share this letter with your instructors and discuss the

accommodations with them as early in your courses as possible. To begin this process, please complete the Registration

form (https://webapps.rutgers.edu/student-ods/forms/registration).

Erin G. Leuthold, MS Ed

(856) 225-2717

Rutgers-Camden Disability Services:

311 North Fifth Street, Camden, NJ 08102-1405

Web page: https://ods.rutgers.edu/contact-ods/rutgers-university-camden

E-mail: disability-services@camen.rutgers.edu

Exam Make-up Policy/Late Policy

If, for a university approved reason, you cannot take an exam at the scheduled time you must give the professor written

notice at least one week in advance so that other arrangements can be made. If the situation does not allow for advance

notification (for example, emergency hospitalization), contact the professor as soon as possible after a missed exam.

Make-up exams for non-university approved reasons are not guaranteed. The professor reserves the right to request

written documentation to support your absence (such as a doctor’s note, an obituary, or military orders).

Student Code of Conduct http://studentconduct.rutgers.edu/university-code-of-student-conduct

Violations of the Student Code of Conduct are considered serious infractions of student behavior and students who

violate the code are subject to penalties relative to the level of the matter. In general, students may not disturb normal

classroom procedures by distracting or disruptive behavior. Examples of disruptive behavior include, but are not

limited to, the following (please adapt for online environments):

• Repeatedly leaving and entering the classroom without authorization

• Answering cellular phone or allowing pager to beep

• Making loud or distracting noises

• Repeatedly speaking without being recognized, interrupting the instructor or other students, or otherwise acting

in disregard of the instructor’s requests

• Threats or violence

Violations of the code should be reported to the Dean of Students office deanofstudents@camden.rutgers.edu or 856-

225-6050. If the violation is immediate and a potential threat is a concern, call the Rutgers-Camden police at 856-225-

6111.