Review: Duke University’s “Data Science Math Skills” Course on Coursera
While on my self-taught Marketing Analytics and artificial intelligence engineer journey — I knew the first big challenge I’d face would be developing a strong foundation in the fundamentals of the math required for this particular field.
My Why
What I wanted was to gain a high-level understanding of the math I’d come across in the fields of data science, further the little knowledge I already had regarding how and what certain math would be used in this field, and truly grasp some of the underlying mathematical principles before going into more advanced courses.
The Cost
This course is free without a certificate and $49 if you want a certificate after successful completion — I opted for the certificate. This choice granted me the opportunity to take the practice quizzes after every section and the ability to solidify my knowledge with a graded quiz at the end of each week.
Time Commitment
Duke split the content into four weeks. They suggest a time investment of 3–5 hours per week, which would take approximately 16 hours to complete. I was able to complete this in under a week, spending about 10 hours total.
The Format
Before each section begins, they offer print-out resources of the entire content they’ll discuss in each video, which you can use to follow along. They call these video companions.
Most of the course content is executed in the form of tutorial-style lectures, where the professors use the entire video screen to write with colored digital pens on a black background — similar to watching a teacher write on a projector, all while doing voice commentary to teach the content.
The Material
Here’s the material covered over the course of four weeks:
Week 1: Building blocks for Problem Solving
- The basic notions of set theory (union, intersections, and cardinality) using a real-world application to medical testing
- Explains the notation we use to discuss intervals on the real number line
- The Jagged S Symbol and how to express a long series of additions and use this skill to define statistical quantities, like mean and variance
Week 2: Functions and graphs
- Covers the Cartesian plane, measures the distance in it using the distance formula, and finds the equations of lines (point-slope, slope-intercept)
- Introduces the idea of a function as an input-output machine
- How to graph functions in the Cartesian plane
Week 3: Measuring rates of change
- A very gentle introduction to the calculus concept of the derivative and how to apply these concepts to the real-world problem of optimization
- Exponents and logarithms — explains the rules and notation for these math tools.
- The change of the base formula
- The rate of change of continuous growth and the special constant known as e
Week 4: Introduction to probability theory
- The vocabulary and notation of probability theory
- The basic definitions and rules of probability — the probability of two or more events both occurring, the sum rule, and the product rule.
- The Binomial theorem and Bayes’ theorem and how they’re used in practical problems
Level of Difficulty
I chose this course based on these criteria:
- No prerequisites
- Beginner-friendly
- Quick overview
- Not too complex
Though the first two weeks are much less complex than the last two weeks, they never got too complicated. Everything is simplified, which makes it more understandable and easier to grasp each concept.
Your basic grade school arithmetic knowledge is required, but they teach you everything you need to know to answer problems you’ll see in the practice and graded quizzes if you choose that route.
Learning Tips
My biggest tip for this course, or any course you take, is to practice discipline and be engaged in the content. It can be easy to passively watch the videos without really solidifying your knowledge.
Spend 30 minutes a day on some practice math problems you can find online. Always look back at your notes, give yourself a refresher before jumping into the next topic, and take some time to study to really master what you learn.
Key Takeaways
The course was an action-packed few weeks during which a lot of topics were covered, but here are the key takeaways from my experience:
- This is a true introductory course, covering some of the math you might see in data science. While this course by no means prepared me to jump into a data science career, what it did was prepare me to further my education in this field and left me with confidence to continue down my educational path.
- Overall, this course definitely gave me what I was looking for. It was designed to give you a refresher and to get you up to speed with the basic math concepts needed to be successful to further your data science education.
Final Review
The professors do a great job of explaining all of the material — you can really feel the passion from them. They covered so many different topics in a short amount of time, oftentimes unrelated to one another. So the flow of the course felt rushed in certain moments, but they didn’t get bogged down by tackling complex topics that’d require an entire dedicated course to cover in proper detail.
The lectures don’t use prepared slides as you might see in a lot of online courses, but working through problems in real time was more illuminating than just flipping through a slide. I found this aspect to be very helpful because visual and verbal happen to my preferred learning styles.
What I liked the most about the way they taught this course was their ability to use real-world problems. They explain where you might see the math and touch on how it can be used in the real-world as a data scientist, which not only excited me but gave it that full-circle feeling.
I recommend it to anyone looking to get started in the field who might not be confident with their math skills. It is very high-level, however, so further learning would be required.
My Next Steps
While this certificate may seem significant, it can’t be the only math training I do to become a self-taught machine learning and artificial intelligence engineer.
For my next steps, I’ll be tackling the Mathematics for Machine Learning Specialization course from Imperial College London on Coursera.
