In science consensus is irrelevant. What is relevant is reproducible results. The greatest scientists in history are great precisely because they broke with the consensus.
– Michael Crichton
Yihui Xie intended to make it easier to do his homework, but instead found himself tackling one of the greatest problems in modern science: the reproducibility of results.
I wanted to do my homework more conveniently
I am excited to announce that we are starting a new meetup group - Python Data Science LA - to serve the needs of the LA data science community using Python!
At an earlier Data Science/Machine Learning meetup we conducted an informal survey on tools for Data Science which showed that a large majority of data scientists use R or Python, with R taking a slight lead. (I know, those attending the meetups bias the results towards open source tools.)
Matt Dowle holds a wealth of knowledge and is worth listening to for a variety of reasons. He has worked for some of the world’s largest financial organizations, holding positions at the focal point between money and data. He has been programming in R for well over a decade. He has lived the open source mantra, by not only consuming open source software but also releasing his own community contributions to CRAN. On top of those accomplishments, he’s also quite funny and spins a heck of a good yarn.
(Note: This is my second installment on grad school, you can read the first part on coursework, the difference between a Master’s and a Ph.D., and proving your worth, here.)
Advancement to candidacy # After finishing coursework, the next step in our PhD program is to “advance to candidacy.” This is the formal process of proposing your dissertation research. Our department has a diverse set of research interests, both theoretical and applied, so dissertations can take a variety of forms. This is yet another difference between Master’s and PhD projects– typically, Master’s projects are more directed by faculty members and PhD projects are chosen primarily by the students themselves. Either way, the process begins with a conversation between the student and his/her advisor.
“At one point, the only source for daily R news was David Smith’s blog.” – Szilard Pafka
David Smith is an integral part of the R community. His background in computational statistics goes back to the early 90s . David has worked at Revolution R, one of the leading R companies, for nearly a decade. His name has been included in lists such as:
R is really cool and people that use it can do a lot…
(Please click here if the video above doesn’t load.)
The video above is from an event at useR! 2014 consisting of Yasmin Lucero’s presentation plus an incredible panel discussion, which is included here. The title of the panel was “R’s Place in the Production Environment”, and was moderated by Tareef Kawaf, President at RStudio. Five high-profile members of the R community were on this panel, spanning the gamut of the enterprise:
(Please click here if the video above doesn’t load.)
During useR! 2014, the LA R Meetup (run by Szilard Pafka) organized an event focused on R’s status in industry. This event took the form of a presentation by Yasmin Lucero, Senior Statistician at AOL-Gravity, followed by a panel discussion. The video above captures the presentation, where Yasmin provides an excellent overview of the current status of R in the enterprise and breaks down the taxonomy of solutions for which R is commonly chosen as a development platform. From batch reporting and dashboards for internal deployment to full blown public facing analytics data products, R has certainly found a home in business. Yasmin’s talk concludes with specific, operational aspects of managing R in the enterprise, such as library and dependency management. Enjoy the video and don’t forget to tune in for Part 2 of this session where the panel discussed “R’s Place in the Production Environment.”
Another exciting development in data science coming from our department at UCLA is a high school class called Introduction to Data Science (IDS). This project has been made possible by a National Science Foundation grant to support Mobilize, for which Rob Gould (mentioned in my previous post on DataFest) is the Principal Investigator.
Data Science @ IRIS.TV
by Dr. Thomas Sullivan, Chief Data Scientist and Joel Spitalnik, VP Engineering
While video recommendations best serve the viewer experience when presented in real time, machine learning is a computationally intensive process that is often done on the back end. IRIS.TV’s utilizes a hybrid process to synchronize computational tasks for learning so that engineering and data science teams can shrink the latency between analyzing history and presentation of recommended videos to improve the experience.
You’ve asked and we’ve listened. The audio content from our DataScience.LA interviews will now be released as a podcast! Check us out on iTunes or point your podcast app to here!
We’ve already released the audio from the userR! conference interviews of Hadley Wickham, Max Kuhn, and Tal Galili on iTunes and hope to release upcoming interviews in podcast form shortly after our video blog posts. Thanks for being a part of the DataScience.LA community and we look forward to hearing more from you!
What I can say is that I am joyful at looking at what people are doing!
“One can acquire everything in solitude except character.”
― Stendhal
The Interview # Tal Galili is, in many ways, a central spoke of the R community. Both gregarious and thoughtful, he has grown his website R-bloggers into the definitive aggregation of the R community’s voice through his genuine, passion-driven intensity. Tal had a simple desire as a young programmer - to learn more about his chosen tools - and looked to the internet to find other voices like his. When googling for “R blogs”, Tal found numerous blogs about pirates, but only a handful about R. This interview details how Tal started R-bloggers and decided to challenge the status quo, as well as giving us a peek behind the curtain for his new projects.
Five LA R meetup members will present short talks of what they learned at useR! 2014
(plus we have 5 more volunteers for the following meetup!)
1. Timothy J. Triche, Jr., PhD
Fellow, Jane Anne Nohl Division of Hematology, USC/Norris Comprehensive Cancer Center
Winston Chang’s “Interactive Graphics with ggvis” talk at useR! 2014 was in many ways a watershed moment for the the data science community, as the future directions of rich data visualization for analytics, communication, and the web were revealed. In this talk, Winston introduces the viewer to ggvis, lays out the fundamental architecture, and provides an overview of reactive architectures.
Eyeo festival highlights
Amelia McNamara
From the festival website, “Eyeo assembles an incredible set of creative coders, data designers and artists, and attendees – expect enthralling talks, unique workshops and interactions with open source instigators and super fascinating practitioners.” I’ll talk about some of the most inspiring talks and projects I saw at Eyeo this year, including work by Santiago Ortiz, Lillian Schwartz, Cod.Act, Kate Crawford and Nicholas Felton. Most of this work can be described best by the term “art,” so come expecting provocations, not prototypes. To learn more about the festival, visit eyeofestival.com
We’ve made a lot of the same mistakes, and we want to protect people from doing them
The Interview # In the video above, Max provides some amazing insights into the why and how of caret, an R package he created. He also discusses his book on Applied Predictive Modeling which he co-authored with Kjell Johnson, including details on how he set out to write the book he wished he would have had. As a special bonus, Max also describes Quinlan’s C5.0, an alternate “forest of decision trees” algorithm, the secrets of which were hidden behind commercial licensing for years - and which has recently been ported and made available to the R ecosystem. Whether you are a beginner just getting your feet wet with R and predictive modeling, or a seasoned data scientist, this interview has something for everyone.
At useR! 2014, John Chambers was generous enough to provide us with insight into the very early stages of user-centric interactive data exploration. He explains, step by step, how his insight to provide an interface into algorithms, putting the user first has placed us on the fruitful path which analysts, statisticians, and data scientists enjoy to this day. In his talk, John Chambers also does a fantastic job of highlighting a number of active projects, new and vibrant in the R ecosystem, which are helping to continue this legacy of “a software interface into the best algorithms.” The future is bright, and new and dynamic ideas are building off these thoughtful, well measured, solid foundations of the past.
