Home // IMMM 2015, The Fifth International Conference on Advances in Information Mining and Management // View article
Authors:
Kalpana Algotar
Sanjay Addicam
Keywords: Text-Rank Algorithm; Multimap; Adjacency List; Internal Prediction; External Prediction
Abstract:
The technological advances in mobile phone and their widespread use has resulted in the big volume and varied types of mobile data we have today. Researchers have begun to mine mobile data in order to predict a variety of social, economic, personal, location and health related events. Mobile data directly reflects individual’s life without disclosing personal information, and therefore it is an important source to analyze and understand the underlying dynamics of human behaviors or activities. In this paper, we describe an innovative and challenging process to predict user’s activity using mobile based data. We propose a graph-based framework that uses the user’s activities, social network, and product-keywords in order to provide recommendations which are also delivered through mobile phones. This paper summarizes the different types of prediction logic algorithms by constructing graphs from different data sources. Our graph-based approach is highly scalable and can be used to predict individual’s next activity, as well as prediction towards products purchase. The mobile recommendation engine incorporates three types of data to generate the graph and to predict activity and product. First, we collect product-keywords using text-rank algorithm. Second, we collect individual mobile’s past data, such as accelerometer, call log, battery status, app usage, browsing history, Facebook data, and Twitter data. Third, we collect user’s mobile phone activity 8 times during the day. By using multimap, we get fast prediction in real-time mobile.
Pages: 61 to 64
Copyright: Copyright (c) IARIA, 2015
Publication date: June 21, 2015
Published in: conference
ISSN: 2326-9332
ISBN: 978-1-61208-415-2
Location: Brussels, Belgium
Dates: from June 21, 2015 to June 26, 2015