Discovering App Relationships in Smart Phones through Large Scale Mining of User Journey Data
Submitted by Hanu Susarla (@susarlah) on Friday, 29 April 2016
User experience while navigating through home screen and apps is a key differentiator for any smart phone. Building a user interface giving ease of use and personalized and contextualized home screen requires deep understanding of how different users are using their phones. Mobile OEMs periodically collect application usage data from millions of smart phone users. Analyzing this massive amount of data can give insights about multi-dimensional relationships between apps in mobile phone usage scenarios. We used sequence mining and graph mining techniques to discover affinity between apps in a Terabyte scale mobile application log data. In this talk I will present the challenges in large scale mining of mobile application logs for discovering the app relationships, some of the approaches we used for solving these and the appropriate metrics that should be used to evaluate the quality of the relationships.
‒Quick intro to data collection from smart phones & data cleansing ‒Intro to sequence mining and its application in discovering user’s journey through apps ‒Discussion on applying lift and confidence for ensuring app correlation ‒Prediction of next potential app usage based on discovered app relationships ‒App clustering by discovering multi-level connected components ‒Potential opportunities for improving user experiences
Dr. Hanumantha Rao is Director and Team Head of Service Platform Team at Samsung R&D Institute, Bangalore. He has worked extensively on large scale data processing, content understanding, information retrieval, application servers and language processors in a career spanning over 24 years in Yahoo, Sun Mircosystems, BEA and TCS. Hanu has a Phd in computer science from IITB.