A Hands-On Introduction to Data Science 2020 Fall


We are living in an era in which the kind of problems that could be solved using data are driving a huge wave of innovations in various industry, from healthcare to education, and from finance to policy-making. Data science is one of the fastest-growing disciplines at the University level. Increasingly, data and data analysis are playing a large role in our day-to-day life, including in our democracy. Thus, knowing the basics of data and data analysis has become a fundamental skill that everyone needs. This course is not just for data science majors but also for those who want to develop their data literacy. It provides a very easy entry for almost anyone to become introduced to data science, but also has enough fuel to take one from that beginning stage to a place where they feel comfortable obtaining and processing data for deriving important insights. In addition to providing basics of data and data processing, this course teaches standard tools and techniques. It also examines implications of the use of data in areas such as privacy, ethics, and fairness.


A hands-on introduction to data science by Chirag Shah, Cambridge University Press, ISBN 978-1-108-47244-9.
Copies of books are made available in the library and by Prof. Hong Yu




  • Lecturing is based on a textbook or/and learning materials provided.
  • Introductory programming may be practiced in the lab.
  • Students will be expected to be prepared for class, and must complete the assignments by the due dates.


Class Attendance Policy

Students should attend the class in the classroom.

Cheating and Plagiarism Policy

All forms of academic dishonesty will result in an F for the course and notification of the Academic Dishonesty Committee. Academic dishonesty includes (but is not limited to) plagiarism, copying answers or work done by another student (either on an exam or assignment), allowing another student to copy from you, and using unauthorized materials during an exam.

Make-up Exams

  • Make-up exams will only be given in case of serious need and only when the instructor is notified prior to the exam time. Otherwise, the grade is automatically zero for that exam/quiz.
  • Written verification for the student s inability to take an exam will be required.
  • The make-up exams will be different from those given to the class.


  • Basic data science concepts and methods
  • Data science tools
  • Data analysis and evaluation


Components of Course Grade:

Assignments 60%
Midterm Exam 20%
Final Exam 20%

Grade Scale: A (4.00), A- (3.75), B+ (3.25), B (3.00), B- (2.75), C+ (2.25), C (2.00), C- (1.75), D+ (1.25), D (1.00), D- (0.75), F (0.00)

Homework Assignments

  • All assignments are to be turned in on or before the due date and time. If students try and cannot turn in an assignment electronically because the campus network is down, you will not be penalized.
  • An assignment turned in up to 24-hours late will be reduced by 10% of the assignment s worth, more than 24 hours late will be reduced 100%.
  • The due date and time for each assignment will be specified on assignment postings.
  • All assignments are expected to be individually and independently completed. Should two or more students turn in substantially the same solution or program, in the judgment of the instructor, the assignment will be given a grade of zero. A second such incident will result in an F grade for the course.


Exams are based on textbooks, supplementary materials, and assignments.



EARLY ALERT STATEMENT Academic Success Support As your professor, I am personally committed to supporting YOUR academic success in this course. For that reason, if you demonstrate any academic performance or behavioral problems which may impede your success, I will personally discuss and attempt to resolve the issue with you. If the situation persists, I will forward my concern to the Student Development Office and your academic advisor to seek their support and assistance in the matter. My goal is to make your learning experience in this course as meaningful and successful as possible.

UMass Data Science Seminar 2019 Fall

What is about

UMass Data Science Seminar is for everyone who is either interested or experienced in the methods and application of machine learning in data science. We want to build a community of data science researchers and practioners in Umass Lowell that can share knowledge, projects, and experience. Our events usually take place on North Campus of UMass LowellWednesday and include two parts: a lecture focusing on topics in deep learning at 3:30-4:40pm; and a seminar talk by invited speaker at 5:00-6:20pm. If you would like to present at the seminar, please email one of the organizers: Hong YuYonghao Jin.


Fall 2019 Speaker List

2019-09-04 Hong Yu

Deep learning is conquering human tasks. How far can it go in medicine?  Advances, Challenges, and future directions

UMass Lowell
2019-09-11 Kun Chen
Reduced rank stochastic regression with a sparse singular value decomposition   University of Connecticut
2019-09-18 Byron Wallace
A Sensitivity Analysis of (and Practitioners’ Guide to) ConvolutionalNeural Networks for Sentence Classification   Northeastern University
2019-09-25 Madalina Fiterau
Deep neural decision forests   UMass Amherst
2019-10-02 Mohit Iyyer
Deep contextualized word representations  UMass Amherst
2019-10-09 Sasha Rakhlin  Consistency of Interpolation with Laplace Kernels is a High-Dimensional Phenomenon MIT
2019-10-16 Stephen Bach
Hinge-loss Markov random fields and probabilistic soft logic  Brown University
2019-10-23 Rocco Anthony Servedio
Learning k-Modal Distributions via Testing  Columbia University
2019-10-30 Zak Kohane
The electronic medical records and genomics (eMERGE) network: past, present, and future   Harvard University
2019-11-06 Fei Wang
Deep learning for healthcare: review, opportunities and challenges   Cornell University
2019-11-13 Salil Vadhan
The complexity of differential privacy  Harvard University