Image from Unsplash

 

Key Takeways

 

• Data science is a field that is constantly evolving
• In the field of data science, learning is lifelong
• A data science professional must continue to improve their knowledge in the field to keep up with new technological developments and software applications

 

Introduction

 

I can remember the joy and excitement that I had when I began my data science journey some 6 years ago. For me, the transition to data science was pretty smooth because of my strong background in advanced mathematics and computational physics. 

However, as I got further and further into my data science journey, I realized that I was not making a lot of progress in terms of learning advanced concepts. I got caught up with learning just the basic concepts. Instead of applying the basic knowledge I already had to real-world data science projects, I kept taking all these different data science courses and data science specializations on platforms such as DataCamp, Udemy, YouTube, edX, and Coursera. 

At one point, it almost became like an addiction to me, as I was constantly searching for data science courses to enroll in, especially the ones that were free of charge. Most of the courses taught on these platforms covered fundamental concepts only, as advanced concepts are introduced, but most often superficially.

Reflecting on my data science journey, if I were to do it again, I would place more emphasis on project-based learning. In my opinion, project-based learning is the most reliable way of learning data science, because it gives you the opportunity to learn as you go. It also helps you to apply your knowledge to real-world data science projects.

While it’s exciting to acquire as much fundamental knowledge as possible, the focus should be to make gradual progress from fundamental concepts to more advanced concepts. Beginner in the field of data science must continue to make quantum leaps in their knowledge as they transition from beginner-level to advanced-level data science professionals.

In what follows, we discuss some of the essential levels of data science.

 

Level I Data Science

 

Level I data science could also be referred to as the Basic Level. At the level I, the data science aspirant should be able to acquire the following skills:

• Be able to work with data presented in a CSV (comma-separated value) file format
• Be able to clean and organize unstructured data
• Be able to work with data frames
• Be able to visualize data using different types of visualizations such as line graphs, scatter plots, qq plots, density plots, histograms, pie charts, scatter pair plots, heatmap plots, etc.
• Be able to perform simple and multiple regression analysis
• Gain competency in essential python libraries for data science such as numpy, pandas, scikit-learn, seaborn, and matplotlib

 

Level II Data Science

 

Level II data science could also be referred to as the Intermediate Level. At level II, the data science learner should master the following:

• Be able to use machine learning classification algorithms such as logistic regression, KNN (K-nearest neighbors), SVM (support vector machine), decision tree, etc.
• Be able to build, test, and evaluate machine learning models
• Be able to perform hyperparameter optimization
• Be familiar with advanced concepts such as k-fold cross validation, grid search, and ensemble methods
• Should be an expert in the use of the scikit-learn library for machine learning applications

 

Level III Data Science

 

Level III data science could be referred to as the Advanced Level. At level III, the data science student should gain the following competencies:

• Be able to work with data presented in advanced formats such as text, image, voice, or video
• Familiar with advanced machine learning techniques such as clustering
• Familiar with deep learning and neural networks
• Familiar with deep learning libraries such as TensorFlow and PyTorch
• Familiar with cloud-based platforms for machine learning deployment such as AWS and Azure

 

Conclusion

 

The three levels of data science discussed above could be summarized in the image below.

 

Three levels of data science | Image by Author.
 

While Level I and Level II competencies could be acquired from online courses, a lot of self-study is essential for learning Level III (Advanced) concepts. An important resource that could help data science aspirants to dive deep into advanced concepts is the following textbook: Machine Learning with PyTorch and Scikit-Learn.

 

Cover of the book
 
The GitHub repository for this textbook can be found here. 

In summary, we’ve discussed the three levels of data science. As data science is a field that is constantly evolving, every data science aspirant should continue to work hard to make the quantum leap to the next level.
 
 
Benjamin O. Tayo is a Physicist, Data Science Educator, and Writer, as well as the Owner of DataScienceHub. Previously, Benjamin was teaching Engineering and Physics at U. of Central Oklahoma, Grand Canyon U., and Pittsburgh State U.