My plan to learn at least the basics of Data Science as a 18 year old.
Learning data science can be daunting especially when there are thousands of courses to pick from. Initially, I thought Data Science was all about coding. I would follow some tutorials on YouTube, and then BOOM….I am a data scientist. However, Data Science is not just coding. It also consists of math, stats, some business knowledge, and much more. Therefore, you really need time to learn the basics of data science.
Even though it’s overwhelming to learn Data Science, it can be made easy if you take everything step by step and learn efficiently. In order to learn Data Science efficiently, I have designed a plan to help me learn at least the basics of data science. You can use the same plan to assist you in your data science career and see if it works for you.
Step 1: Learn math
For math, you need to know statistics, linear algebra, and calculus. All the basics of these subjects can be learned from Khan Academy and for free! Khan does a great job in explaining all the major math concepts to a beginner. However, you don’t need to learn every math subject right away since it’s going to be overwhelming for you. Initially, I would recommend learning statistics since it’s vital for a data scientist to analyze graphs and trends. After learning statistics, you can move onto learning Linear Algebra and then Calculus. However, you can pick up any math subject you like and get started. Here the links to every math subject :-
Statistics: https://www.khanacademy.org/math/statistics-probability
Linear Algebra: https://www.khanacademy.org/math/linear-algebra
Calculus: https://www.khanacademy.org/math/differential-calculus
Statistics is going to be easy if you know the basics of Algebra 1. For linear algebra, you should know the basics of Algebra 2. And for Calculus, you should know the basics of Pre-calculus and Algebra 2.
If you’re unfamiliar with Algebra 1, Algebra 2 and Pre-calculus, I highly recommend studying these first. Again, you can study these subjects from Khan Academy.
However, is it worth the time to learn all these math subjects? Of course! The following explain how these math subjects will apply to your data science career.
Statistics: As a data scientist, you should know your stats. Statistics is a vital tool used to analyze graphs and trends. For example, you might collect some data about a company’s XYZ product, graph the data, and see whether the product is popular among customers by analyzing the graph using standard deviation, mean, median, and/or mode.
Linear Algebra: NumPy is a popular module used by data scientists that requires you to have an understanding of vectors and matrices. However, in data science, linear algebra is more than vectors and matrices.
Calculus: Calculus is important for calculating derivatives, slope of a curve, and/or gradient descent. It’s mainly used in key Machine Learning algorithms, therefore, it’s good to know calculus if you want to transition your career from data science to ML or be a successful data science.
Step 2: Learn how to code
The most common programming languages used in data science are Python, SQL, and R. Almost all developers recommend starting with Python since it’s a general purpose language. However, I personally recommend learning SQL and Python together since these two languages go hand in hand as you progress in your data science careers. You will use Python libraries, for example, NumPy, Pandas, and/or MatplotLib to analyze and visualize data. Learning SQL will be vital for knowing how to interact with databases and create tables, which is one of the main jobs of a data scientist. In order to become familiar with Python, Automate the Boring Stuff is a great free online book to follow. For SQL, this SQL course by Mike Dane is great to get you started with the basics of SQL and databases.
Don’t worry about learning R right now and instead focus on Python and R. However, it will be a good idea to learn R once you get comfortable with Python and SQL.
Step 3: How to learn
Given that you have so much on your table, how can you squeeze all of it in your free time? To learn data science, you don’t need to learn everything in one day. In order to learn bits of everything, pick a different subject each day and learn the subject for 30–60 minutes. For example, today you can learn a math subject, tomorrow you can learn Python, and the day after that you can learn some SQL. In order to commit everything to memory, be sure to revise everything once in a week. For example, if you learn Python concepts one day, then the next time you sit down to learn Python, revise everything you learnt in Python so far.
CONCLUSION
Learning data science is a marathon not a sprint, therefore, take your time. Nobody can be a data scientist in a month or 6 months unless you have a math or computer science degree. However, learning data science with a solid plan will make it easy for you to learn at least the basics of data science. Even though I am not good at making step-by-step plans, you can try my plan and see if it works for you.
With everything said, if you are familiar with SQL and Python and are worried about how to start your first project, be sure to check out my blog post on how you can store employee information using a voice bot in Python.
Was this article helpful? Be sure to give it a clap and leave a feedback :)
