The future of social science research: A data science perspective

The future of social science research: A data science perspective

Last week, the Urban Institute hosted a discussion on the evolving landscape of data and the potential impact on social science. “Machine Learning in a Data-Driven World” covered a wide range of important issues around data science in academic research and their real policy applications. Above all else, one critical narrative emerged:

Data are changing, and we can use these data for social good, but only if we are willing to adapt to new tools and emerging methods.

Book Review: R for Everyone by Jared P. Lander

This is a guest post by Vadim Y. Bichutskiy, a Lead Data Scientist at Echelon Insights, a Republican analytics firm. His background spans analytical/engineering positions in Silicon Valley, academia, and the US Government. He holds MS/BS Computer Science from University of California, Irvine, MS Statistics from California State University, East Bay, and is a PhD Candidate in Data Sciences at George Mason University. Follow him on Twitter @vybstat.

Recently I got a hold of Jared Lander's book R for Everyone. It is one of the best books on R that I have seen. I first started learning R in 2007 when I was a CS graduate student at UC Irvine. Bored with my research, I decided to venture into statistics and machine learning. I enrolled in several PhD-level statistics courses--the Statistics Department at UC Irvine is in the same school as the CS Dept.--where I was introduced to R. Coming from a C/C++/Java background, R was different, exciting, and powerful.

Learning R is challenging because documentation is scattered all over the place. There is no comprehensive book that covers many important use cases. To get the fundamentals, one has to look at multiple books as well as many online resources and tutorials. Jared has written an excellent book that covers the fundamentals (and more!). It is easy-to-understand, concise and well-written. The title "R for everyone" is accurate because, while it is great for R novices, it is also quite useful for experienced R hackers. It truly lives up to its title.

Chapters 1-4 cover the basics: installation, RStudio, the R package system, and basic language constructs. Chapter 5 discusses fundamental data structures: data frames, lists, matrices, and arrays. Importing data into R is covered in Chapter 6: reading data from CSV files, Excel spreadsheets, relational databases, and from other statistical packages such as SAS and SPSS. This chapter also illustrates saving objects to disk and scraping data from the Web. Statistical graphics is the subject of Chapter 7 including Hadley Wickham's irreplaceable ggplot2 package. Chapters 8-10 are about writing R functions, control structures, and loops. Altogether Chapters 1-10 cover lots of ground. But we're not even halfway through the book!

Chapters 11-12 introduce tools for data munging: base R's apply family of functions and aggregation, Hadley Wickham's packages plyr and reshape2, and various ways to do joins. A section on speeding up data frames with the indispensable data.table package is also included. Chapter 13 is all about working with string (character) data including regular expressions and Hadley Wickham's stringr package. Important probability distributions are the subject of Chapter 14. Chapter 15 discusses basic descriptive and inferential statistics including the t-test and the analysis of variance. Statistical modeling with linear and generalized linear models is the topic of Chapters 16-18. Topics here also include survival analysis, cross-validation, and the bootstrap. The last part of the book covers hugely important topics. Chapter 19 discusses regularization and shrinkage including Lasso and Ridge regression, their generalization the Elastic Net, and Bayesian shrinkage. Nonlinear and nonparametric methods are the focus of Chapter 20: nonlinear least squares, splines, generalized additive models, decision trees, and random forests. Chapter 21 covers time series analysis with autoregressive moving average (ARIMA), vector autoregressive (VAR), and generalized autoregressive conditional heteroskedasticity (GARCH) models. Clustering is the the topic of Chapter 22: K-means, partitioning around medoids (PAM), and hierarchical.

The final two chapters cover topics that are often omitted from other books and resources, making the book especially useful to seasoned programmers. Chapter 23 is about creating reproducible reports and slide shows with the Yihui Xie’s knitr package, LaTeX and Markdown. Developing R packages is the subject of Chapter 24.

A useful appendix on the R ecosystem puts icing on the cake with valuable resources including Meetups, conferences, Web sites and online documentation, other books, and folks to follow on Twitter.

Whether you are a beginner or an experienced R hacker looking to pick up new tricks, Jared's book will be good to have in your library. It covers a multitude of important topics, is concise and easy-to-read, and is as good as advertised.

DC NLP November 2014 Meetup Announcement: Introduction to Topic Modeling with LDA

Curious about techniques and methods for applying data science to unstructured text? Join us at the DC NLP November Meetup!

This month's event features an overview of Latent Dirichlet Allocation and probabilistic topic modeling.

Topic models are a family of models to estimate the distribution of abstract concepts (topics) that make up a collection of documents. Over the last several years, the popularity of topic modeling has swelled. One model, Latent Dirichlet Allocation (LDA), is especially popular.

Notes on a Meetup

This is a guest post by Catherine Madden (@catmule), a lifelong doodler who realized a few years ago that doodling, sketching, and visual facilitation can be immensely useful in a professional environment. The post consists of her notes from the most recent Data Visualization DC Meetup. Catherine works as the lead designer for the Analytics Visualization Studio at Deloitte Consulting, designing user experiences and visual interfaces for visual analytics prototypes. She prefers Paper by Fifty Three and their Pencil stylus for digital note taking. (Click on the image to open full size.)

October Starts Off Right: A Full Month of Events for Women in Tech

This is a guest post by Shannon Turner, a software developer and founder of Hear Me Code, offering free, beginner-friendly coding classes for women in the DC area. In her spare time she creates projects like Shut That Down and serves as a mentor with Code for Progress. 

Over 200 women were in attendance for the DC Fem Tech Tour de Code Kickoff party held at Google Thursday night.  DC Fem Tech, a collective of over 25 women in tech organizations, collaborates to run events and support the women in DC's tech community.

DC NLP October 2014 Meetup Announcement: Automated Query Parsing, and Fact Checking with Truth Teller

Curious about techniques and methods for applying data science to unstructured text? Join us at the DC NLP October Meetup!

This month features an introduction to the art of automated query parsing, and a discussion of a WaPo app that automates some of the tedium of fact checking.

Tony Maull is a Senior Director of Enterprise at DataRPM. He will discuss the differences between computational search and content search. His primary focus is how the computation can be relied upon when a natural language question can be asked any number of ways but still needs to drive a consistently accurate answer.

DC NLP September 2014 Meetup Announcement: Natural Language Processing for Assistive Technologies

Curious about techniques and methods for applying data science to unstructured text? Join us at the DC NLP September Meetup!

This month, we're joined by Kathy McCoy, Professor of Computer & Information Science and Linguistics at the University of Delaware. Kathy is also a consultant for the National Institute on Disability and Rehabilitation Research (NIDRR) at the U.S. Department of Education. Her research focuses on natural language generation and understanding, particularly for assistive technologies, and she'll be giving a presentation on Replicating Semantic Connections Made by Visual Readers for a Scanning System for Nonvisual Readers.

Simulation and Predictive Analytics

This is a guest post by Lawrence Leemis, a professor in the Department of Mathematics at The College of William & Mary.  A front-page article over the weekend in the Wall Street Journal indicated that the number one profession of interest to tech firms is a data scientist, someone whose analytic skills, computing skills, and domain skills are able to detect signals from data and use them to advantage. Although the terms are squishy, the push today is for "big data" skills and "predictive analytics" skills which allow firms to leverage the deluge of data that is now accessible.

I attended the Joint Statistical Meetings last week in Boston and I was impressed by the number of talks that referred to big data sets and also the number that used the R language. Over half of the technical talks that I attended included a simulation study of one type or another.

The two traditional aspects of the scientific method, namely theory and experimentation, have been enhanced with computation being added as a third leg. Sitting at the center of computation is simulation, which is the topic of this post. Simulation is a useful tool when analytic methods fail because of mathematical intractability.

The questions that I will address here are how Monte Carlo simulation and discrete-event simulation differ and how they fit into the general framework of predictive analytics.

First, how do how Monte Carlo and discrete-event simulation differ? Monte Carlo simulation is appropriate when the passage of time does not play a significant role. Probability calculations involving problems associated with playing cards, dice, and coins, for example, can be solved by Monte Carlo.

Discrete-event simulation, on the other hand, has the passage of time as an integral part of the model. The classic application areas in which discrete-event simulation has been applied are queuing, inventory, and reliability. As an illustration, a mathematical model for a queue with a single server might consist of (a) a probability distribution for the time between arrivals to the queue, (b) a probability distribution for the service time at the queue, and (c) an algorithm for placing entities in the queue (first-come-first served is the usual default). Discrete-event simulation can be coded into any algorithmic language, although the coding is tedious. Because of the complexities of coding a discrete-event simulation, dozens of languages have been developed to ease implementation of a model. 

The field of predictive analytics leans heavily on the tools from data mining in order to identify patterns and trends in a data set. Once an appropriate question has been posed, these patterns and trends in explanatory variables (often called covariates) are used to predict future behavior of variables of interest. There is both an art and a science in predictive analytics. The science side includes the standard tools of associated with mathematics computation, probability, and statistics. The art side consists mainly of making appropriate assumptions about the mathematical model constructed for predicting future outcomes. Simulation is used primarily for verification and validation of the mathematical models associated with a predictive analytics model. It can be used to determine whether the probabilistic models are reasonable and appropriate for a particular problem.

Two sources for further training in simulation are a workshop in Catonsville, Maryland on September 12-13 by Barry Lawson (University of Richmond) and me or the Winter Simulation Conference (December 7-10, 2014) in Savannah.

DC NLP August 2014 Meetup Announcement: Automatic Segmentation

Curious about techniques and methods for applying data science to unstructured text? Join us at the DC NLP August Meetup! dcnlp_july

This August, we're joined by Tony Davis, technical manager in the NLP and machine learning group at 3M Health Information Systems and adjunct professor in the Georgetown University Linguistics Department, where he's taught courses including information retrieval and extraction, and lexical semantics.

Tony will be introducing us to automatic segmentation. Automatic segmentation deals with breaking up unstructured documents into units - words, sentences, topics, etc. Search and retrieval, document categorization, and analysis of dialog and discourse all benefit from segmentation.

Tony's talk will cover some of the techniques, linguistic and otherwise, that have been applied to segmentation, with particular reference to two use cases: multimedia information retrieval and medical coding.

Following July's joint meetup with Statistical Programming DC and Data Wranglers, where Tommy Jones and Charlie Greenbacker performed a showdown between tools in R and Python, we're back at our usual U-Street location.

DC NLP August Meetup

Wednesday, August 13, 2014 6:30 PM to 8:30 PM Stetsons Famous Bar & Grill 1610 U Street Northwest, Washington, DC

The DC NLP meetup group is for anyone in the Washington, D.C. area working in (or interested in) Natural Language Processing. Our meetings will be an opportunity for folks to network, give presentations about their work or research projects, learn about the latest advancements in our field, and exchange ideas or brainstorm. Topics may include computational linguistics, machine learning, text analytics, data mining, information extraction, speech processing, sentiment analysis, and much more.

For more information and to RSVP, please visit: