Meta & resources
Probability, statistics and data analysis
Machine Learning: concepts & procedures
Machine Learning: fundamental algorithms
Machine Learning: model assessment
Artificial neural networks
Natural language processing
The computer science appendix
The mathematics appendix
Beautiful web of data science
The web is a beautiful place, you can learn so much on it, for free. This document puts together some resources on (or related to) data science which are absolutely brilliant. We also list books, which might not be freely available, but that are definitely worth owning.
Topic-specific resources are already reported in every page for details, more insight, citation purposes. Here, we list material which is comprehensively good: online courses, books, blogs, and which span several of the topics we touch on in the book. All the material listed here can be of any format.
Statistics and Probability
- C Bergstrom and J West, Calling Bullshit in the age of Big Data, a course on spotting manipulative use of data, wrong conclusions and overall misunderstandings
- D Huff, How to lie with Statistics, a lovely little book on the common mistakes and misunderstandings around the use of numbers for reaching conclusions. Old (1954), but very valuable and entertaining
- T Vigen, Spurious Correlations, a project which visually displays improbable correlations between sets of data to demonstrate that, well, correlation is not causation
Machine Learning - general material
Neural Networks
- F Chollet, Deep Learning with Python, Manning, 2017, a book by the creator of Keras
- The TensorFlow Neural Network playground, an interactive tool to visualise the inner workings of ANNs
Computer Vision
- ​The Hypermedia Image Processing Reference, a website built by the University of Edinburgh, School of Informatics
- ​Pyimagesearch, a website curated by A Rosebrock on Computer Vision and Machine/Deep Learning on images, with tutorials for OpenCV and lots of good material
Coding and Computer Science
- ​Practical Business Python, a website By C Moffitt devoted to best practices on using Python for practical reasons, it's very good
- Gayle Laakmann McDowell, Cracking the Coding Interview, CareerCup
Python references relevant to Data Science
- ​Scipy lecture notes - they're pretty brilliant and obviously focussed on Python, but you can learn lots of concepts in data
- ​scikit-learn has tutorials and extensive explanations for every algorithm they support, as well as general notes on Machine Learning
- ​The Hitchhiker's guide to Python (not particularly targeted at data science, but a very useful reference)
Meta & resources - Previous
The meta on all this
Next - Probability, statistics and data analysis
Probability, its interpretation, and statistics
Last modified 8mo ago