Data Science Weekly Newsletter

Issue

May 14, 2015

‍

Editor's Picks

‍

How I Became a Data Scientist Despite Having Been a Math Major
Caution: the following post is laden with qualitative extrapolation of anecdotes and impressions. Perhaps ironically (though perhaps not), it is not a data driven approach to measuring the efficacy of math majors as data scientists...

Software development skills for data scientists
Data scientists often come from diverse backgrounds and frequently don't have much, if any, in the way of formal training in computer science or software development. That being said, most data scientists at some point will find themselves in discussions with software engineers because of some code that already is or will be touching production code...

The Machine Vision Algorithm Beating Art Historians at Their Own Game
Classifying a painting by artist and style is tricky for humans; spotting the links between different artists and styles is harder still. So it should be impossible for machines, right?...

‍

‍

Analytics Of Optimal 2-for-1 Strategy In NBA Basketball
For those that don’t watch basketball, the two for one shot involves a strategic play at the end of the quarter... Jeff Van Gundy...said all the analytics guys push to take the 2 for 1 shot in every situation, but asked if it’s a tied fourth quarter Game 7 of the NBA finals, would it be smart to go for the 2 for 1 situation...It’s a hard data science questions as there aren’t that many past Game 7’s in the NBA even, much less ones that get decided by fewer than one or two possessions. But still a 2 for 1 situation just makes intuitive sense at the very least. Two chances to score are always better than the one. But how much better is it?...

The evolution of popular music: USA 1960–2010
Study of pop music evolution using LDA (topic analysis) on music samples...

Forbidden Data - Wyoming just criminalized citizen science
Wyoming doesn’t, of course, care about pictures of geysers or photo competitions. But photos are a type of data, and the new law makes it a crime to gather data about the condition of the environment across most of the state if you plan to share that data with the state or federal government...

Using The Right Tool For the Job - Prototyping in R, Productionalizing in Go
Slides from a talk given by Drew Lanenga, a Data Scientist at Lytics, about how we try to balance the best of both languages at Lytics. We'll walk through a changepoint detection package we recently made and open sourced, prototyped and validated in R, then ported to Go to fit in nicely with the rest of our stack...

Deep Learning for Image Understanding in Planetary Science
I stumbled upon a tweet by Leon Palafox, a Postdoctoral Fellow at the The University of Arizona Lunar and Planetary Laboratory, and reached out to him to discuss his success with GPUs and share it with other developers interested in using deep learning for image processing...

Problem with Data Science - Probabilistic Programing for Non-Statisticians
Data Science is about learning from data, often using Machine Learning and statistics...Ideally, we build the model that most accurately describes our data, makes the best predictions, and provides the answers of interest...

Statistics: P values are just the tip of the iceberg
Ridding science of shoddy statistics will require scrutiny of every step, not merely the last one, say Jeffrey T. Leek and Roger D. Peng...

Genetic Programming in Python, with a scikit-learn inspired API
While Genetic Programming (GP) can be used to perform a very wide variety of tasks, gplearn is purposefully constrained to solving symbolic regression problems. This is motivated by the scikit-learn ethos, of having powerful estimators that are straight-forward to implement...

Movie Recommendation with MLlib
In this chapter, we will use MLlib to make personalized movie recommendations tailored for you. We will work with 10 million ratings from 72,000 users on 10,000 movies, collected by MovieLens. This dataset is pre-loaded in the HDFS on your cluster in /movielens/large. For quick testing of your code, you may want to use a smaller dataset under /movielens/medium, which contains 1 million ratings from 6000 users on 4000 movies...

Logistic Regression and Gradient Descent
Logistic regression is an excellent tool to know for classification problems. Classification problems are problems where you are trying to classify observations into groups. To make our examples more concrete, we will consider the Iris dataset. The iris dataset contains 4 attributes for 3 types of iris plants. The purpose is to classify which plant you have just based on the attributes...

‍

‍

Data Scientist - GrubHub - NYC
GrubHub Holdings Inc. is the nation's leading online and mobile food-ordering company dedicated to connecting hungry diners with local takeout restaurants. The GrubHub Holding Inc. portfolio of brands includes GrubHub, Seamless, MenuPages and Allmenus. The company's online and mobile ordering platforms allow diners to order directly from thousands of takeout restaurants across the country and London, and every order is supported by the company's 24/7 customer service. GrubHub Holdings Inc. has offices in Chicago, New York City and London. With a career at GrubHub Holdings Inc., you can order your cake and eat it, too!...

‍

‍

How to get your very own RStudio Server and Shiny Server with DigitalOcean
If you've always wanted to have an RStudio Server of your own so that you can access R from anywhere, or your own Shiny Server to host your awesome shiny apps or Rmarkdown documents, DigitalOcean (DO) can help you get there easily...

New York R Conference Videos Are Up
2015 New York City R Conference - videos include talks by Chris Wiggins (Chief Data Scientist, The New York Times), Hilary Parker (Senior Data Analyst, Etsy, Inc.), Andrew Gelman (Statistics Professor, Columbia University), Saar Golde (Chief Data Scientist, Knowledgent), and many more...

PyNeural: A simple but fast Python library for training neural networks
PyNeural is a neural network library written in Cython which is powered by a simple but fast C library under the hood. PyNeural uses the cblas library to perform the backprogation algorithm efficiently on multicore processors. PyNeural exposes a simple Python API that plays nicely with NumPy, making it easy for you to munge your data sets as needed and quickly use them to train a neural network for classification...

‍

‍

Statistics Done Wrong: The Woefully Complete Guide
A pithy, essential guide to statistical blunders in modern science that will show you how to keep your research blunder-free...
"If you're interested in learning about the current debates/ problems/ challenges in statistics, this is a good primer. The book is written for scientists, but it can benefit anyone who deals with data and analysis on a regular basis..."...

‍