Pono Music Leveraging Co-occurence Recommendations on Datameer.

Title: 'Pono Music Leveraging Co-occurence Recommendations on Datameer'

Presenters:

Kevin Fielding, Full Stack Engineer at PonoMusic

Shiuh-Wuu (Victor) Liu, Solutions Engineer - Global Philanthropy at Datameer.org

Kevin is a Full Stack Software Engineer at PonoMusic. He graduated from the University of Western Ontario with a BS in Computer Science. He had previously worked with ITS eLearning which involved building a content delivery platform that required high levels of functionality and user interactivity and Zynga which is a notable gaming company. He is passionate about music and is a musician outside of his tech endeavors.

Shiuh-Wuu (Victor) has tried four other industries prior to finding his true calling in Big Data Analytics. In prior professional lives, he has worked in civil structural engineering, edtech, natural gas waste pressure recovery, and solar. He is a licensed Professional Engineer with a BS in Structural Engineering from the University of California, San Diego and an International Business MBA from National Sun Yat-sen University. He is currently serving on the Board of Directors for both SunSaluter and Solar Richmond. At Datameer, he is launching Datameer.org which is the philanthropic arm of the company serving both academia and nonprofits.

Abstract:

Kevin Fielding, Full Stack Engineer at PonoMusic, and Shiuh-Wuu (Victor) Liu, Solutions Engineer - Global Philanthropy at Datameer.org will demonstrate how Pono Music used Datameer, a self-service big data analytics solution, to create a custom music album recommendation system based on co-occurences to help customers discover music that people with similar tastes purchase in order to increase sales. In this talk, Kevin and Victor will cover the primary challenges of creating a recommendation system, including the need to plan for computing resources and data modeling necessary to address sparsity, scalability, and cold start. The two will share how Hadoop, an open source platform that leverages commodity hardware to parallelize process big data calculations, plays a key role in addressing scalability, and how creating another source of related data pulled in by deploying Google Analytics more extensively on each music page addresses the cold start challenge.

It should be noted that this type of recommendation is easy to deploy and is not restricted to only music recommendations. With minimal modifications, the basic workbook could be applied to many industries including ecommerce, gaming, academia, retail, etc.

Part I: Live Datameer Demonstration

Overview of Salesforce objects – Kevin

Data ingestion, preparation, analysis, visualization, and operationalization – Victor

PonoMusic Interface and Integration with Salesforce - Kevin

Part II: Next Steps? Ideation.

Google Analytics Data - Victor

Improving the Model – Group Discussion [img src="https://ci6.googleusercontent.com/proxy/RnNZfQn2o2xpggJQqefCOervMbPIci5mujDPJnvl43kv6Rtxjyh5gHN_JKVzeU-aaGz3pePFgxfoAAtZJZNx8mveVTc-11j98EfuAJVcumUenA=s0-d-e1-ft#https://ssl.gstatic.com/ui/v1/icons/mail/..." class="CToWUd">

PonoMusic is revolutionizing music listening by bringing back native high-resolution audio, the way the artists intended and recorded their music. Neil Young, the iconic artist and musician is the founder and CEO. PonoMusic is the only complete high-resolution music ecosystem that includes the PonoPlayer, PonoMusic Store and PonoMusic Community.

What is Datameer.org?

Datameer.org was created to serve the academic, research and non-profit communities who desperately need the ability to access and analyze vast amounts of data to find the insights that matter to them. By providing access to Datameer to these deserving communities for free, together we can accelerate the time it takes to make meaningful progress on making the world a better place. Datameer is a proven self-service big data analytics solution that makes it easy for anyone to get from raw data to insights, fast.