- Table of contents
- MSc Data Science - AUEB
- Introduction
- Navigation
- About the program
- Disclaimer
- Contact Information
- University: Athens University of Economics and Business
- Program: Master of Science in Data Science
- Class: Part Time 2021-2023
- Required ECTS points: 75
- Structure: 6 quarters, only courses
- Website: datascience.aueb.gr
- Final Grade: 9.12/10
Welcome to the repository for the projects completed during my Master in Data Science program at AUEB. This repository contains a collection of projects and assignments from various courses throughout my academic journey. Each project is organized within its respective folder, making it easy to explore and understand the work I've done.
Each project contains the solution that was submitted, usually consisting of a report along with the code. To respect the work of the professors, the handouts and data are not included in the repository. In cases where the data is publicly available, corresponding links for data download are provided.
To explore the projects, you can navigate through the folders corresponding to each course. Each folder is typically structured as follows:
{Project_Name or Assessment_id}/: Contains the project files, including code, documentation, and any additional resources.README.md: Provides a brief overview of the project, its objectives, and instructions for running or replicating the analysis (if applicable).
The README.md also contains information regarding the course (number of projects, final exams, individual/group
projects, etc.). Moreover, the final grade of each course is mentioned. However, please note that this grade represents
the aggregated grade, including the final exam. Most projects received a grade of 10/10, with some exceptions where the
grade was 9.5/10.
This section is aimed to provided my honest opinion on the program, which could help future candidates to decide
whether this master's degree is suited for them.
-
Part-Time option:
- The biggest benefit of the Program is the fact that it offers a Part-Time Option. This is targeted to professionals who want to combine studies with work. Although the load is very high, it is manageable and well-structured. Provided that there isn't another Master in Data Science in Greece that is truly part-time, pursuing the MSc in Data Science in case you are working is a no-brainer.
-
Course quality:
- The program exhibits a noticeable course quality. For instance, the "Advanced Customer Analytics" course provided a fantastic learning experience, showcasing real-world applications of machine learning.
- Most courses will introduce new concepts and go into detail. The only downside, is that oftentimes the introductory material is repeated. For example, the concepts of classification/regression etc. were covered by more than 4 courses.
- On the downide, some courses fell short of expectations, having an "undergraduate" structure. These courses either had hideous exams or bad course material, which was not suited to level of the MSc.
-
Communication with administration is sub-par:
- Timely and effective communication with the administration is crucial for a smooth academic experience. Unfortunately, we've encountered difficulties in getting our questions answered and obtaining clear information on basic information. For example, the academic calendar with available (elective) courses per quarter was never announced, thus we could not make a plan for the year.
-
Courses list was announced one week prior to starting the quarter:
- Late course announcements can be challenging for students trying to plan their schedules and prepare for coursework. In most cases, the quarter schedule (which included the courses list) wasn't announced until a couple of days before starting.
-
Overall too expensive:
- The cost of education is a critical factor for students, and it's important to ensure that the program offers commensurate value. In my opinion, the program was too expensive, especially when comparing the price to other public universities across Greece.
In this section, I've highlighted some of the best assessments and projects, according to my opinion.
-
Advanced Customer Analytics
- Description: One of the best Courses of the Program, where applications of machine learning in the industry were demonstrated. The first project was a review summarization where we had the opportunity to scrape, aggregate, using various techniques (LDA, openai API etc) and plot the results in a PDF
- Folder: Advanced Customer Analytics
-
Recommender Systems
- Description: Again one of the best Courses of the Program. The second project could be a real-life project. A (Beer) Recommender system was used, which produced excellent results, using various techniques.
- Folder: Recommender Systems
-
Deep Learning
- Description: Another excellent course, where we had the chance to dive deep into modern Deep Learning Algorithms and techniques. The assessments focus on computer vision starting from a simple dataset (Fashion MNIST) and the expanding to a more complex dataset containing X-RAYS (MURA Dataset)
- Folder: Deep Learning
-
Text Analytics
- Description: The 4 assessments contain a nice introduction to Deep Learning for Text. Complex models were built using the building blocks of LMs (GRUSs, (bi)LSTMs, self-attentions etc.). Models were trained to perfoms various classification tasks.
- Folder: Text Analytics
-
Machine Learning and Computational Techniques
- Description: Nice course with a LOT of assessments. The math behind the basic concepts of machine learning was explored during the assessments.
- Folder: Machine Learning and Computational Techniques
You may find the order I took the courses in. I have added one or two asterisks next to elective courses as follows:
*Recommended
**Not Recommended
I would suggest to avoid stacking so many courses early on, and take advantage of the final semester, to make the journey more enjoyable.
- Y1Q1
| Course | Course type |
|---|---|
| Practical Data Science | Core |
| Probability and statistics for data analysis | Core |
| Introduction to Data Management and Engineering ** | Elective |
- Y1Q2
| Course | Course type |
|---|---|
| Large Scale Data Management | Core |
| Machine Learning and Computational Techniques | Core |
- Y1Q3
| Course | Course type |
|---|---|
| Numerical Optimization and Large Scale Linear Algebra | Core |
| Text Analytics | Core |
- Y2Q1
| Course | Course type |
|---|---|
| Data visualization and communication | Core |
| Legal, Ethical and Policy Issues of Open Data | Core |
| Advanced Customer Analytics * | Elective |
| Social Network Analysis * | Elective |
- Y2Q2
| Course | Course type |
|---|---|
| Data Mining ** | Elective |
| Deep Learning * | Elective |
| Recommender Systems * | Elective |
- Y2Q3
At this point I had gathered 78 ECTS points thus didn't have to take any courses.
Please note that this repository is intended for educational and portfolio purposes only. It represents my progress and development as a data scientist during my studies. Some projects may include code or solutions provided as part of the course materials, while others may involve collaborative work with fellow students or guidance from instructors.
The author of this repository is not responsible for any academic issues that may arise from the use of the materials contained herein. It's important to adhere to the academic integrity policies and guidelines of your own institution when using this repository as a reference. Copying or submitting any part of this work as your own without proper attribution may violate academic integrity rules and regulations.
If you have any questions, want to discuss any of the projects in more detail, or simply want to connect, feel free to reach out to me:
- Name: George Chalkiopoulos
- Email: geo.chalkiopoulos [at] gmail [dot] com