Big data through cross-platform interest-based interactivity

Given the ubiquity of social media, we capitalized on interest-based relevance as a key component to enhance user experience. Interest-based relevance modeling was extracted from user interaction in a cross-platform social media Big Data repository. The goal of this study was twofold: first we addressed theoretical dilemmas of a cross-platform user experience; second, we implemented an android-based mobile application and designed a cloud architecture to account for theoretical parameters of Big Data User-centric approach and interactivity. To address cross-platform Big Data challenges, we relied on cloud computing to perform computationally intensive operations such as searching, data mining, and data processing at large scale. Our use case was based on a cross-platform interest-based navigation and content filtering across multiple radio content streams. The streams consisted of tags from radio stations´ programming and social media content through a discovery process. User interaction was geared to enable preferred topic filtering, flexibly shifting participation roles, notifications, and navigation through external data sources. We tested our application on a list of popular radio stations and their social media content streams (including Facebook, Twitter, Google+) to generate a Big Data scenario. Even if diverse and nature, data stream traffic generated through a mobile application KASU has proven to be a robust Big Data source with a list of ad hoc user-centric applications.