CX4240: Introduction to Computational Data Analysis (2023 Spring)

Table of Contents

Logistics

Course Content

Q: What will be covered in this course? A: This course introduces techniques for computational data analysis, with an emphasis on machine learning algorithms and their applications to real-world data. On the technique side, we will cover key supervised machine learning methods (linear regression, logistic regression, neural networks, tree-based models) and unsupervised method (k-means, Gaussian mixture models, expectation-maximization, dimension reduction). On the application side, it will introduce various applications of these techniques, particularly on text data analysis and natural language processing. It will introduce how to formulate real-world tasks as data analysis problems, key methods for solving these problems, and their advantages and disadvantages.

Q: Who will benefit from this course? A: The learning objective is that by the end of this course, the students are able to formulate their data analysis problems at hand, choose appropriate computational models to acquire insights from data automatically, and even come up with innovative solutions for solving open problems in this field. The course will be helpful for students who want to solve practical problems using machine learning and data science techniques. The course will provide useful techniques for students who want to do edge-cutting research in data mining, machine learning, natural language processing, and others.

Q: What are the prerequisites? A: Prerequisites for this course include 1) solid knowledge of probability, statistics, calculus, and linear algebra; 2) basic knowledge of machine learning; 3) solid programming skills, preferably in Python.

Schedule

Date Topic Due
01/09/2023 Course Overview  
01/11/2023 Probability and MLE Piazza Signup
01/16/2023 No Class (Martin Luther King Day)  
01/18/2023 Data Analysis Toolbox  
01/23/2023 Linear Regression  
01/25/2023 Linear Regression HW1 Out
01/30/2023 Example Projects  
02/01/2023 Naïve Bayes Classifier  
02/06/2023 Logistic Regression  
02/08/2023 Feature Design and Learning for Text  
02/13/2023 Feature Design and Learning for Text HW1 Due
02/15/2023 Project checkpoint & discussion  
02/20/2023 Neural Networks HW2 Out
02/22/2023 Neural Networks  
02/27/2023 CNNs and RNNs  
03/01/2023 Transformers  
03/06/2023 Large Language Models HW2 Due
03/08/2023 Decision Trees  
03/13/2023 Random Forest HW3 Out
03/15/2023 Project checkpoint & discussion  
03/20/2023 No Class (spring break)  
03/22/2023 No Class (spring break)  
03/27/2023 Clustering (K-Means & Hierarchical)  
03/29/2023 Guest Lecture by Prof. Celine Lin  
04/03/2023 Dimension Reduction HW3 Due
04/05/2023 Review Class project presentation signup
04/10/2023 No Class (Exam Preparation)  
04/12/2023 Exam  
04/17/2023 Project presentation  
04/19/2023 Project Presentation  
04/24/2023 Project Presentation  

Grading

Homework (30%)

There will be three assignments, each account for 10% towards your final score. Each assignment includes written analysis and/or programming for testing your understanding of the taught content.

  • Late policy: Assignments are due at 11:59PM of the due date. You will be allowed 2 total late days (48 hours) without penalty for the entire semester (for homework only, not applicable to exams or projects). Once those days are used, you will be penalized according to the following policy:
    • Homework is worth full credit before the due time.
    • It is worth 75% credit for the next 24 hours.
    • It is worth 50% credit for the second next 24 hours.
    • It is worth zero credit after that.
  • Follow the Georgia Tech Academic Honor Code.

Project (30%)

You need to complete a project on using computational data analysis techniques to tackle a real-life data analysis problem. Each project needs to be completed in a team of 2-4 people. Here are some guidelines and resources for doing your project smoothly.

Exam (40%)

One exam will be held on April 12 in lieu of the regular class:

  • The exam will be closed-book. No peer communication is allowed—you may not message or collaborate with others.
  • There will be no make-up exams. You will get zero credit for your missed exam.

Resources

Other resources, such as machine learning toolboxes and datasets, will be provided throughout the course.