Abstract
A large volume of research in ubiquitous systems has been devoted to using data, that has been sensed from users’ smartphones, to infer their current high level context and activities. However, mining users’ diverse longitudinal behavioral patterns, which can enable exciting new context-aware applic
Highlights
While discrete observations of an individual’s behavior can appear almost random, typically there are repetitive and identifiable patterns or routines in every person’s life
We show that our algorithms and approaches can model user behavior with high accuracy and demonstrate improved performance over existing approaches
We evaluate the approaches that we have implemented currently on the entire dataset of 200 users from the DeviceAnalyzer data collection as well as the dataset collected from the field study with 10 fold cross validation
Summary
While discrete observations of an individual’s behavior can appear almost random, typically there are repetitive and identifiable patterns or routines in every person’s life. People carry them around everywhere and use them as their primary medium for many day to day activities These devices can collect a variety of data about users such as their locations (from GPS), sensory data (from various sensors), call and sms logs etc. As part of this vision, we plan to collect large-scale data from users’ smartphones and employ it to infer diverse frequent patterns that capture different aspects of their behavior. A contextaware middleware extracts the user’s context history (for instance, call or location history) from these logs It utilizes this history to learn and store the users’ behavioral models (behavioral patterns that users exhibit in similar context or situations over a period of time) for predicting future behavior. We show that our algorithms and approaches can model user behavior with high accuracy and demonstrate improved performance over existing approaches
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
