Characterizing User Behavior and Items in Multimedia Streaming Services

Abstract

Recently, Multimedia Web content has gotten so much popular worldwide (e.g., YouTube and Netflix). Traditional corporations are also contracting online multimedia platforms to provide content to their users. These services are called non-UGC (non User Generated Content), where these companies produce and publish content for client consumption. In this work we present a methodology for characterizing and analyzing multimedia streaming services. We validate it using actual data from one of the most popular online video streaming provider of Latin American, called Samba Tech. Our methodology is divided in four modules that allow evaluating and understanding different views and data characteristics.