Earlier this week, I posted an article on How to Create an Amazing Data Science Portfolio:
At the end of the article, I posted a link to an example portfolio that I liked by Tim Dettmers. Afterward, I had a few people ask me to compile a larger list of great data science portfolios and projects. So – here we go! My 5 favorite data science portfolios.
iSee
While not a portfolio, but rather a project, I think this is a great format to try and exemplify.
Melissa Runfeldt did a great job defining and motivating her problem, discussing how she gathered data and explaining her methods with images of results. All in a way that would be easy for a non-technical person to follow (at least at a high level). One thing I thought was missing, though, was a link to the code.
Fake News Classifier
This is a more advanced way to showcase a project, but I believe it is really well done.
The website provides a nice, simple overview of the approach and even a way to have your own news articles classified (though, it now appears to be broken). Also, the website links to a series of Medium articles that go into more depth on the approach and results. Again, though, I had trouble finding any links to the code the team wrote.
Stanford Machine Learning Projects
This example is not a portfolio of a person, but rather of a class. The Machine Learning class at Stanford consists of a project and the reports and posters can be seen online:
Here you will find a lot of really nice reports such as the one on Eluding Mass Surveillance: Adversarial Attacks on Facial Recognition Models. This report is more academic in its nature but does a great job explaining the problem, data, and experimental results. It even links to the code on GitHub!
FastML
FastML is a great website run by Zygmunt Zając.
In his own words, "FastML probably grew out of a frustration with papers you need a Ph.D. in math to understand and with either no code or half-baked Matlab implementation of homework-assignment quality." One of his most popular posts is Deep learning made easy and does a good job discussing deep unsupervised learning and links to code.
Lego Sorting
This project is amazing!
Not only is it a really cool idea that involved 200 hours spread across 6 months, but he also took the time to create an amazing explanation of the entire process on Medium. He even uploaded the datasets he created to Kaggle! If you are looking for the gold standard in data science projects, this is definitely it.
Conclusion
I hope my attempt to showcase some of the data science projects and portfolios that have inspired me will also help you go and create something awesome! If you have other examples that you love or even your own, please put them in the comments so everyone can enjoy them!
This article can also be found here.