CISE Research Resources: Being There: Mobile Devices for Community and Commerce

University Of Minnesota-Twin Cities, Minneapolis MN

Investigators

Abstract

EIA 0224392 Terveen, Loren Konstan, Joseph A. Riedl, John Shekhar, Shashi University of Minnesota Twin Cities Title: CISE RR: Being There: Mobile Devices for Community and Commerce This project, building a mobile computing system, provides a set of mobile, connected, location-aware computing units consisting of Personal Digital Assistants (PDAs) and wearable displays with some input ability to allow experimentation to affect deployment on the types of interfaces, applications, and technologies needed as these devices become commonplace. Devices will carry out a wide range of experimentation, from tightly controlled experiments to broad field studies, from previously planned extensions of existing research to last-minute ideas created by student projects, and from small pilot studies to large deployment studies. The infrastructure supports research projects in human computer interaction (HCI), recommender systems, and spatial data management for environments where users are mobile; specifically, Ubiquitous Peripheral Social Interaction (UPSI), Location-Aware Recommender Systems, and Spatial Data Management for Location Aware Mobile Computing. The first project presumes availability of several wireless communication channels. Information, at the periphery of users' attention, is constantly streamed to the device and displayed as a scrolling "ticker." A user carries a pocket computer and wears a Networked Wrist Device linked by Bluetooth. This approach presents challenges in populating and managing the information model driving the interaction, designing the interface for the networked wrist display, coupling the wrist computer and PDA, and identifying the types of tasks for which this interaction paradigm is suitable. Using algorithmic and interface research, pilot tests, real-world experiments, and metrics, the second project looks into suitable applications among wired, disconnected, and wireless connected recommenders. Given local and temporal patterns of users, this project examines when it is appropriate to interrupt its user with a recommendation, how to present recommendations most effectively across multi-device interfaces, and how to compute recommendations when the user is disconnected from the network. The last project looks into the requirements of spatial data management systems for applications like mobile location based recommendation systems. The project gathers a trace-based workload from location aware recommendation systems to evaluate solutions for critical design decisions such as algorithms for map compression and predictive pre-fetching strategies to cache relevant segments of maps based on past spatial queries, scheduled meetings locations, and related information. The educational plan includes integration of the platform into courses and student involvement in summer internships.