INF 428/528: Analysis, Visualization, and Prediction in Analytics

Semester: Fall 2022 
Day/Time: Mondays Wednesdays 18:00 - 19:20	
Location: Physics 224
Instructor: Chen Zhao
Contact: czhao4@albany.edu
Office Location and Hours: By appointment
Course Website: https://charliezhaoyinpeng.github.io/ualbany-inf-428-528/ 

Course Description:

Principles of data analysis, emphasizing modern statistical and machine-learning based approaches. Also, the important role of simple analyses and visualization to gain an overall understanding of data sets, regardless of size. The role of analytics in creating predictive models of phenomena. The importance of understanding the nature of the data and other methodological considerations. Prerequisites: Some statistics and database experience.

Student Learning Objectives:

• Understand and be able to demonstrate principles of data analysis.
• Be able to do simple data and visualization analyses to understand a new and complex data set.
• Demonstrate an understanding of and be able to use statistical and machine learning methods for data analysis.
• Demonstrate an understanding of and be able to use visualization analyses.
• Demonstrate an understanding of and be able to use predictive models.

Course Outline:

Lecture	Topic	Notes
1	Introduction and Overview
2	Intro/Review of Python I (Variables, If/Else, Loop )
3	Intro/Review of Python II (List, Dictionary)
4	Intro/Review of Python III (Class)	Quiz1
5	Machine Learning and Predictive Models I (Linear/ Logistic regression, K-mean)	HW1
6	Machine Learning and Predictive Models II (Decision tree, SVM)
7	Machine Learning and Predictive Models III (Neural Networks)
8	Data Wrangling with Pandas I (Missing values, Outliers, Feature selection)	Quiz2,HW2
9	Data Wrangling with Pandas II (Imbalance, Standardization)
10	Machine Learning with Python I (Loss functions, Optimizers, Model Training)
11	Machine Learning with Python II (Implementation, Evaluation, Interpretation)
12	Visualization with Matplotlib I
13	Visualization with Matplotlib II
14	Introduction to Dash Plotly
15	Invited Talk
16	Special topic and projects (TBD)

Reading Materials (recommended but not required):

• Python Data Analysis, Ivan Idris, 2014 Packt Publishig, ISBN-13: 978-1783553358.
• Pattern Recognition and Machine Learning (PRML) by Christopher M. Bishop
• Dash Plotly Official, https://plotly.com/

Examinations and Grading:

• Midterm Exam: 30%
• Quizzes: 10%
• Homework Assignment: 30%
• Final Project: 30%
• Bonus: 10%

If a student feels they will miss a course obligation for any reason, they must reach out to the course instructor for guidance.

Homework:

Homework Assignments

• Homework assignments will be given every two weeks
• All assignments must have your name, student ID, course name/number, and section number
• Submit HW on the blackboard
• File Name: HW#_Name.pdf (writeup) Or HW#_Name.zip (code)

Late Homework

• -10 penalty for the first 24-hour
• -20 penalty for >= 24 hour
• Zero score for >= 3 days

Disability Policy:

Reasonable accommodations will be provided for students with documented physical, sensory, systemic, medical, cognitive, learning, and mental health (psychiatric) disabilities. If you believe you have a disability requiring accommodation in this class, please notify the Disability Resource Center (518-442-5490; drc@albany.edu). Upon verification and after the registration process is complete, the DRC will provide you with a letter that informs the course instructor that you are a student with a disability registered with the DRC and list the recommended reasonable accommodations. You can review the Equity and Compliance website as well for additional information.

Academic Integrity:

It is every student’s responsibility to become familiar with the standards of academic integrity at the University. Claims of ignorance, of unintentional error, or of academic or personal pressures are not sufficient reasons for violations of academic integrity. See http://www.albany.edu/undergraduate_bulletin/regulations.html

Course work and examinations are considered individual exercises. Copying the work of others is a violation of university rules on academic integrity. Individual course work is also key to your being prepared and performing well on tests and exams. Forming study groups and discussing assignments and techniques in general terms is encouraged, but the final work must be your own work. For example, two or more people may not create an assignment together and submit it for credit. If you have specific questions about this or any other policy, please ask.

The following is a list of the types of behaviors that are defined as examples of academic dishonesty and are therefore unacceptable. Attempts to commit such acts also fall under the term academic dishonesty and are subject to penalty. No set of guideline scan, of course, define all possible types or degrees of academic dishonesty; thus, the following descriptions should be understood as examples of infractions rather than an exhaustive list.

• Plagiarism
• Allowing other students to see or copy your assignments or exams
• Examining or copying another student’s assignments or exams
• Lying to the professor about issues of academic integrity
• Submitting the same work for multiple assignments/classes without prior consent from the instructor(s)
• Getting answers or help from people, or other sources (e.g. research papers, web sites) without acknowledging them.
• Forgery
• Sabotage
• Unauthorized Collaboration (just check first!)
• Falsification
• Bribery
• Theft, Damage, or Misuse of Library or Computer Resources

Any incident of academic dishonesty in this course, no matter how “minor” will result in

• No credit for the affected assignment.
• A written report will be sent to the appropriate University authorities (e.g.the Dean of Undergraduate Studies)
• One of (1) A final mark reduction by atleastone-half letter grade (e.g. B, B-, C-, D+), (2) A Failing mark (E) in the course, and referral of the matter to the University Judicial System for disposition.