Congratulations to Team TwitterTV (Jony F, Justin S, Todd C, Chris K, and Jie Z) for taking top prize at the Hopper ElasticSearch Hackathon for their utilization of ElasticSearch to create interactive visualizations of the top 150 TV shows trending on Twitter. Summaries of the other five team (or individual) presentations are below, and viewable on ChallengePost. Thank you to everyone who participated, and to our generous sponsors ElasticSearch, Traackr, and SoftLayer!
- Exploring the Special Relativity of Politics (Aki B, Siva C, Vaidehi V) came in 2nd place winning a seat in ElasticSearch Training for one lucky team member! They filtered tweets by topic (Obama, McCain, Obamacare, Rowhani) and measured sentiment to "analyze the overall vibe around each topic by gathering aggregates of the positive and negative words"
- Enron Connections (Marty J and Peak X) queried the Enron data set (~500,000 emails) to quantify the connections between all the email participants and determine the degrees of separation between people. They created drop-down lists to facilitate the comparison of any two participants in their exploration of the "6 degrees of Kenneth Lay."
- Cluster in Azure - Boom on Load (Chris M) built an ElasticSearch cluster in Azure running on six machines and devised to create a map that showed "influencers near a given city", but he unfortunately encountered a few problems and ended up learning a lot about troubleshooting the ES cluster!
- Pathfinder (Sarah L) was inspired by the original form of the Wikipedia game 'Clicks to Hitler' and used ElasticSearch to generate the shortest optimal path between two Wikipedia articles.
- Test Run Kibana (Xiaodan Z, Jeff D, Xin C, Creence L) used the Kibana interface to visualize ElasticSearch queries for Justin Bieber from the Twitter river (~3 million tweets).
