Big Data Applications at AT&T Labs (featuring hands-on RCloud demo)

Event space and refreshments sponsored by: AT&T (http://stats.research.att.com/people.php)

Event agenda:

6:30 - 6:50 PM: Networking & Food

7:00 - 7:20 PM: “Hands-on RCloud Demo ( http://rcloud.social ) - Bring Your Laptops!” by Jo Frabetti (https://www.linkedin.com/in/jfrabetti)

7:20 - 7:40 PM: “Visualizing the Customer Experience: Shiny & RCleaflet in RCloud” by Emily Dodwell (https://www.linkedin.com/in/emdodwell)

7:40 - 8:00 PM: “Traffic Anomaly Detection” by Zhengyi Zhou (https://www.linkedin.com/in/zhengyizhou)

8:00 - 8:20 PM: “Deconstructing Domain Names to Reveal Latent Topics” by Cheryl Flynn (https://www.linkedin.com/in/cheryl-flynn-b7068256)

8:20 - 9:00 PM: Socializing

Preparation for Event:

Bring your laptop if you would like to participate in the hands-on RCloud demo. RCloud is a cloud application, no admin permission is required, but you will need to connect to an open Wi-Fi network.

What is RCloud?

RCloud is a social coding environment, designed to jumpstart your work. With RCloud you can rate and share code with other developers. You can form social circles around specific topic areas, and search for similar, relevant work, so you don’t have to recreate the wheel. In addition, RCloud is highly scalable. RCloud gives you superfast interactions with data in HDFS or other systems. This is possible because of a built-in chunk-wise compute + combine paradigm via customized functions for fast I/O.

Hands-on RCloud Demo ( http://rcloud.social ) - Bring Your Laptops! by Jo Frabetti

AT&T Labs will kick-off with a hands-on demo of RCloud presented by Jo Frabetti (https://www.linkedin.com/in/jfrabetti). We will start with the basics from logging-in and familiarizing the audience with the RCloud environment and progress to advanced big data applications. RCloud (on GitHub) (http://github.com/att/rcloud) is open-source software developed primarily by Simon Urbanek (http://urbanek.info/), Gordon Woodhull (http://gordon.woodhull.com/) (AT&T Research) and Carlos Scheidegger (http://cscheid.net) (University of Arizona) to facilitate collaboration for visualization and big data analytics.

Visualizing the Customer Experience: Shiny and RCleaflet in RCloud by Emily Dodwell

Leaflet is the leading open-source JavaScript library for mobile-friendly interactive maps. RCleaflet is an RCloud package used for interactive maps in RCloud. In this talk Emily Dodwell will illustrate how to use both Shiny and RCleaflet to visualize the customer experience. Emily is a Data Scientist in the Statistics Research Department at AT&T Labs. She is interested in issues of data quality and data integrity and works on problems related to customer experience modeling and AT&T network usage. Emily holds a M.A. in Statistics from Yale University.

Traffic Anomaly Detection by Zhengyi Zhou

Zhengyi will talk about how we detect and interpret anomalies in traffic flow in a robust, efficient, and unsupervised manner. Zhengyi works in the Statistics Research Department in AT&T Labs. She develops methods for analyzing trajectories and modeling customer experience. She holds a Ph.D. from Cornell University in Applied Mathematics, specializing in spatio-temporal modeling and time series.

Deconstructing Domain Names to Reveal Latent Topics by Cheryl Flynn

This talk on lightweight web mining analyses will compare inferred topics across data sets and across time periods, and show how these topics can be used as features to a supervised learning model for the detection of malicious domain names. Cheryl Flynn is a member of the Statistics Research Department at AT&T Labs. She is interested in problems related to text mining, statistical privacy, model selection, and unsupervised learning. She holds a Ph.D. in Statistics from New York University.

Twitter: @RClouddotsocial (https://twitter.com/RClouddotSocial)