- Tuesday, 12 May: Tidewater Analytics – APIs: The Good, The Bad, and The Ugly
- Tuesday, 19 May: 757 R User Group – The apply() family of functions
MOOCs and other educational venues:
- Applied Logistic Regression – Ohio State is offering this eight-week course via Coursera starting Monday, 11 May. Logistic regression is an increasingly popular statistical model has become the standard method in many domains for regression analysis of binary response data. Here is the link: https://www.coursera.org/course/logisticregression. Note: Enrollment for this course will close on Wednesday, May 13, 2015.
- Machine Learning for (Smart) Dummies – Yahoo Labs (a new one on me) is offering a course called Machine Learning for (Smart) Dummies. Here is the link: http://yahoolabs.tumblr.com/post/97839313996/machine-learning-for-smart-dummies. I have no idea what this is about or how good it is. If someone wants to take one for the team and present on it that would be nice.
- Text Mining and Text Analytics – Natural language processing is rapidly entering the mainstream and gaining credibility. Statistics.com is offering a offering a four-part Text Analytics series that covers (1) Text Mining, (2) Natural Language Processing, (3) Python’s Natural Language Processing Toolkit, and (4) Sentiment Analysis. This seems like a decent entry point into text analytics.
Books and such:
- Statistics (the easier way) with R. This appears to be a very good book. Lots of good reviews on Amazon. Here’s a link to an article on r-bloggers: http://www.r-bloggers.com/my-new-favorite-statistics-data-analysis-book-using-r/
- Bayesian Statistics – For those who attended Luke Watson’s presentation on Bayesian statistics in April, you might find this blog post by Allen Downey interesting. It’s about using Bayesian methods to predict marathons: http://allendowney.blogspot.com/2015/04/two-hour-marathon-by-2041-probably.html
- Data Science IDE for Python – Apparently there is a new IDE for Python called Rodeo that’s specifically meant for data science/analytics. And apparently it’s modeled on RStudio. Here’s a link to the article announcing it: Introducing Rodeo. And here’s a link to some comments about it on Reddit: http://www.reddit.com/r/Python/comments/33ln2b/rodeo_a_data_science_ide_for_python/ (But keep in mind…it’s Reddit).