CSCI145 / MATH166: Data Mining

About the Instructor

Name	Mike Izbicki (call me Mike)
Email	[email protected] You should probably contact me through Github Issues
Office	Adams 216
Office Hours	See Issue #150
Zoom	See Issue #148
Webpage	https://izbicki.me
Research	Machine Learning (see izbicki.me/research.html for some past projects)

Fun facts:

1. grew up in San Clemente (~1 hr south of Claremont)
2. 7 years in the navy
   1. nuclear submarine officer, personally converted >10g of uranium into pure energy
   2. worked at National Security Agency (NSA)
   3. left Navy as a conscientious objector
3. phd/postdoc at UC Riverside
4. taught in DPRK

About the Course

General Information:

1. This is the theory course for CMC's Data Science major
2. Prepare you for industry or graduate school
   1. Especially for machine learning technical interviews
   2. No SQL in this course => that's CSCI143 Big Data

Course Content:

See the introduction Jupyter notebook.

This course is divided into three sections:

1. Pagerank - We will use the paper "Deeper Inside Pagerank".

   

   You will learn:

   1. How to use pagerank to find the most important items in a "network"
   2. Trade-offs in runtime and quality of results

2. Machine Learning - We will use the textbook Learning from Data by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, and Hsuan-Tien Lin. (I will provide you a free copy to keep.)

   

   You will learn:

   1. Techniques
      1. Logistic Regression
      2. Decision Trees
      3. Support Vector Machines
      4. Kernel Methods
      5. K-Nearest Neighbor
      6. Neural Networks
   2. Math
      1. Bias/variance trade-off
      2. VC Dimension theorem (fundamental theorem of statistical learning)
      3. Regularization (L1, L2, elastic net, weight decay, early stopping, etc.)
      4. Optimization algorithms (gradient descent, stochastic gradient descent, ADAM, etc.)

3. Applications - We will use a variety of papers/blog posts. Focus on text/web/social media examples.

Other Topics:

1. Ethical implications of data mining

   Pet peeve: You can't understand the ethics if you don't understand the technical details.

2. Apply data mining libraries (PyTorch, scikit-learn, GenSim, spaCy, etc.)

   1. Teaching you how to use these libraries is NOT the primary goal of the course.
   2. In-person class time will focus on the math, and I'm expecting you can figure out how to use the libraries on your own.

Prerequisite knowledge:

1. computation
   1. big-o analysis
   2. git
   3. download/use python libraries
2. linear algebra
   1. eigenvectors
   2. matrix rank
   3. linear independence of vectors
3. statistics
   1. super basic probability
   2. exposure to linear/logistic regression helpful but not required

Grades:

Category	Points
Projects	2**2-2**3	between 4-7 projects
Quizzes	2**2-2**3	as needed
Midterm 1 (Pagerank)	2**5	(approximately) 23 Sep
Midterm 2 (Machine Learning)	2**6	(approximately) 01 Nov
Final	2**7

Projects:

1. You lose 2**(i-1) points on every assignment, where i is the number of days late.

2. Do not expect partial credit for incomplete assignments. It is much better to submit a correct assignment late than an incorrect one on time.

3. I expect most people to get full credit on the assignments.

Exams:

1. Majority of your grade
2. No programming, only math
3. Oral format, 1-1 in my office
4. Final exam can replace all previous exams if you do well

This is a Hard Class:

1. The material is intrinsically hard.

   1. Very few people find linear algebra, statistics and programming to ALL be easy subjects, and this class combines them all.

   2. There's a reason people who understand this material get paid big salaries at FAANG.

      What do data scientists get paid?

      1. Generic data scientist (at FAANG): https://www.levels.fyi/t/data-scientist?countryId=254
      2. Generic software engineer (at FAANG): https://www.levels.fyi/t/software-engineer?countryId=254
      3. Machine learning engineer (this class): https://www.levels.fyi/t/software-engineer/title/machine-learning-engineer?country=254
      4. Generic data scientist (at non-FAANG): https://www.bls.gov/ooh/math/data-scientists.htm

2. You will have to read the required references.

   Not all the material will be covered in lectures, and that's intentional to force you to get practice reading research-level data mining texts.

   NOTE: In all of my other courses, I include required reading/watching tasks to learn about CS/DS culture. This course doesn't have these tasks because there is already a LOT of textbook reading that you will have to complete.

3. Comments from previous students:

   1. Holy fucking shit this was a hard class. I had no idea there was so much god damned fucking math involved in a CS class. You should warn students about that.

   2. I got my job because of the data mining course. Technical interviews were super easy because of this class.

   Unfortunately, I can't remove the math from this class, and I can't make the class easier. Otherwise, you wouldn't be learning the material needed to pass a technical interview / get a good job / go to grad school.

Collaboration Policy:

You are encouraged to discuss all projects with other students, subject to the following constraints:

1. you must be the person typing in all code for your assignments, and
2. you must not copy another student's code.

You may use any online resources you like (including ChatGPT) without restriction.

Basically, I'm trusting you all to be adults. You are ultimately responsible for ensuring you learn the material! So do what will help you learn best.

Accommodations for Disabilities

I've tried to design the course to be as accessible as possible for all students. If you need any further accommodations---even if you don't have an officially recognized disability---please ask.

I want you to succeed and I'll make every effort to ensure that you can.