— In the News —
Companies are amassing an enormous database of human emotions using technology that relies on psychology and data mining to analyze people’s faces. Paul Ekman, the pioneer of this technology, fears he has created a monster.
Microsoft agreed this week to acquire Revolution Analytics. For the data science community, that's a big deal. Here's why it matters.
— Tools and Techniques —
Interested in learning how to do data analysis with Python but not sure where to start? Here's a comprehensive learning guide to get you going.
Just because two things appear to be related to each other doesn’t mean that one causes the other. Trends in the data are particularly problematic. Here's why, with suggestions on how to handle the issue.
— Resources —
Nice collection of Artificial Intelligence courses, books, video lectures and papers. If you're interested in AI, you definitely don't want to miss this.
The author of spaCy, Matthew Honnibal, claims that this is the fastest Natural Language Processing (NLP) software ever released. It's written in Python and is available under an open-source license for research.
Nice collection of data sources on the web. Most are free and many are formatted for easy access via R.
— Data Viz —
An image kernel is a matrix that's used to apply effects such as blurring, sharpening, outlining or embossing. Image kernels are also used in machine learning for feature extraction, which is a technique to determine the most important portions of an image. Here's how they work.
This site puts one year's State of the Union in context with any other. Select a word or two, and a frequency graph of those words is displayed according to their use by other presidents. This is built on the Bookworm project, which is available on GitHub and allows for easy analysis of text repositories.
Plotting data in Python is a good news/bad news story. The good news is that there are a lot of options. And the bad news is that there are a lot of options. Here's an overview of what those options are and when to use them.