Summary and Winners for BMD Spring Competition 2014

Competition information.

Big Mountain Data had its first Big Data competition over the last month.  The goal of the competition was to provide a Big Data solution to the companies that provided data sets.  This website (https://bigmountaindata.wordpress.com/competition/) has detailed information on the competition.  We will provide a summary here and a summary of what each competitor did.

Skullcandy

They were looking to understand the customer better and find the customers that Skullcandy was not aware of.  The primary goal was to use Social Media to find these customers.  One such question from Skullcandy “Who are the non-customers who follow us, and visit the same websites as our customers?”  They challenged the contestants to acquire data on social media sites and find what the sentiment of those customers were.

Health Equity

They focused on predicting when customers would need to add more funds into the HSA account they maintained with Health Equity.  The competitors were given data from Health Equity and then needed to predict spending habits, how accounts are funded and take into account spending patterns for the customers.

Both data sets were sanitized of the Personal Information for the customers.

We have already started planning our next competition and will be releasing more information as we start to finalize the competition.

 

Team Details and Solutions

 

The PowderQuants (Ben Taylor, Ben Peters) Team of Ben Taylor and Ben Peterson worked on both the Skullcandy Data Set and Health Equity Data Set.

PowderQuants team For Skullcandy, They did an awesome grouping/analysis/bucketing of tweets around Skullcandy customers and allowing them to gather a rich data set of tweets for analysis. PowderQuants_skullCandySlideDeck

For Health Equity the PowderQuants Team focused on Predicting the spending of customers so that they could properly analyze when they would need to fund an HSA account.  Combining this with data from Health Equity and using algorithms they were able to not only predict when people should fund but they could add more criteria into the model and have it change the predictions based on richer data sets.  PowderQuants_HEQ_presentation

David Gonzalez, Alton Alexander and Brett Denney make up the next team that focused on the Skullcandy data Set.  They created several key components.  A program to pull data from Amazon to get ratings on products, a dashboard to review the data retrieved from amazon and all the backend functions/scripts needed so that Skullcandy could not only find the sentiment of customers but they could also compare that to the sentiment of other competitor products.  (Click here for Code/Solution)

The Flash Team (Gregg Cowley, Meitetsu Todaka, Landon Jensen,Jason Andrus) focused on the Health Equity data set and created a Health Score concept that allowed the company to mix industry standard data with Health Equity data to create a score for the customer.  (Click here for Code/Solution)

Ryan Tarver , Tyler Hawkes, Brandon Hawkes focused on the Health Equity data sets and created a system that would predict when someone should add money into the HSA account based off a simple true or false flag to let the customer know they should fund. Click link for PPT presentation of solution Big Mountain Data Competition Hawkes_Tarver

Mike Roylance & team focused on Health Equity Predicted funding trends based on past history of data and used that information to tell when people should fund the HSA account. (Click here for Code/Solution)

Think Big Analytics (Miquel Rosas,Chitra Ramanathan) Team Focused on Health Equity data sets and showed a very detailed full architectural plan of how the solution they created could be implemented into the customers system. Think Big Analytics Health Equity presentation

Seba Jean-Baptiste and Richard Miner focused on the Health Equity data set created predictive analytics on when to fund the HSA account.  (Did not finish formal solution)

 

What is Big Mountain Data?

Big Mountain Data Spring will be held April 12th In downtown SLC.  at 222 S Main Street the Goldman Sachs offices. 

Big Mountain Data was an event held on Sept 7th 2013 in Utah.

Visit http://www.uhug.org for more great Big Data information.
This page is simply a placeholder for bigger things to come if you want future information about events coming up sign up at http://www.utahgeekevents.com and you will be notified about future events.

The schedule for the event can be found here

http://www.sqlsaturday.com/246/schedule.aspx
What is Big Mountain Data Conference?
Big Mountain Data Conference is all about bringing together the community to discuss data and the role it will play in our future. The market has seen a huge surge in the need for Big Data solutions and the need for Data Architects, Data Analyst and Database Professionals. We wanted to put on a conference that didn’t just focus on one technology. So we decided to plan a SQL Saturday with more to it than just the SQL Focus. We partnered with the local Utah Hadoop Users group (www.uhug.org) and Utah Geek Events (a 501c3 company) to create a conference all about DATA! We will be inviting speakers from all technologies to come and share their knowledge and experience of working with data. We will be holding a traditional SQL Saturday along with additional sessions/tracks involving non SQL Server technologies. We will also have a mixing track showing attendees how to integrate multiple data technologies.
The goal of the conference is to help attendees learn to build data solutions, manage the data they have and socialize and meet other people in the industry to build the community. We will have tracks/sessions that focus on SQL Server, NoSQL, Hadoop, MySQL and everything in between. Our goal is to show that regardless of the technology you use, if you have a Data problem large or small you can find a tool/method to solve that problem.