MicroMasters in Data Science
Being part of an Artificial intelligence (AI) Association, I arrange mentor sessions for people who want to pursue the field of Artificial Intelligence, Data Science and Machine Learning. Often times I come across the same question “where to start from”. Sometimes this question is from an experienced professional who wants to switch the field, other times it is from a fresh graduate who wants to pursue the field of AI. In this blog, I will share my journey to a MicroMasters in Data Science.
Every time I get to hear this question “where to start from”, it reminds me of the time when a few years ago I wanted to study something related to Artificial Intelligence/ Data science but wasn’t able to find anything in the universities/colleges in my city. I spent a lot of time finding the right program. Now that I see others struggling, I want to share the program I completed recently and give feedback on this program. Hopefully, it will help someone out there to make the right decision.
Recently, I completed a MicroMasters in Data Science from the University of California, San Diego. It contains four graduate-level courses (Please bear in mind that this program is not suitable for people who don’t have a background in Math, Statistics and programming). It is a program offered through edX (if you don’t know about edX, check out this link it has a ton of amazing free courses offered by universities all over the world).
The first course in this series is Python for Data Science, it is an informative and well-prepared course to learn about the libraries of Python that are used for Data Science. It teaches you Jupyter notebook, Python libraries like Numpy and Pandas, Matplotlib for data visualization, machine learning and scikit-learn, natural language processing using NLTK, and how to work with databases using Python.
The second course in this series is Probability and Statistics in Data Science using Python. It will introduce you to probability and statistics, sets, counting, combinatorics, conditional probability, random variable, expectation, variance, discrete and continuous distribution families, inequalities and limit theorems, parameter estimation and confidence interval, regression and PCA, and hypothesis testing. This was the hardest course of this program. Statistics was my minor in my undergrad program even then I found it difficult. Many concepts needed to answer problem sets and programming assignments were not covered in the videos/material provided. The course introduction says you need to spend 10–12 hours/week but you will need to spend way more time than that.
The third course is Machine Learning Fundamentals. This course provides a thorough grounding in a wide range of machine learning methods, for classification, regression, conditional probability estimation, clustering, and dimensionality reduction. If you want to learn Machine Learning, you not only need to know computer programming but some Math as well. The math needed for this course is Probability, Linear Algebra and Calculus. Trust me, the final exam was like a Math paper.
The last course in this series is Big Data Analytics Using Spark. This course teaches you how to perform statistical analysis of very large datasets that do not fit on a single computer. You will learn some of the most popular tools for performing this type of analysis: apache spark, XGBoost and TensorFlow.
You have to finish all these courses within 15 months in order to get the Micromasters certificate. As I was working, I took one course/semester. If you are not passionate about Micromatsers you can only take the course you are interested in. This was my learning journey and although it was not easy I am glad that I learnt ML and DS from knowledgeable professors from the comfort of my home. Let me know in the comments below if you have more questions.
