A common popular technique for learning data science is starting a project. Here are the 3 E’s for why building a data science project is a good idea.
A data science project will expose you to all the stages of the data science process. You will need to start with the identification of an interesting question or problem. Then you will have to find and collect the necessary data. After that, cleaning and modeling the data is important. Finally, the result needs to be presented (this is called deployment).
Employers always want experience, and a project can provide that.
Once a project is started, there is always something new to learn. For example, the project will have data. Where do you store that data? It might need a database. Should you use a cloud database or install one locally. You will have to learn how to do that. Another example, after you have collected your data, you might realize some rows are incomplete. Then you will have to look into methods for dealing with missing data. This will require more learning.
A project provides a better learning environment than a list of courses because each new thing you learn has a reason and a purpose. To follow the examples from above: You know why you are creating a database and you know why missing data is important.
Make sure to pick a project that is interesting to you. This decision will make the project more enjoyable. If you enjoy sports, maybe build a project around fantasy sports. Sports are filled with data. If you enjoy reading, build something around books, authors, or magazines. If exercise is your thing, build an app to predict your progress. Plus, if you find an answer to your question, it is always fun to solve something.