This project will design and implement "Myrmex", a novel crowdsourcing system for physical-world tasks. For example, an elderly community member could use Myrmex for help getting mail from a remote post office, as the system would assign the task to a participant who routinely passes that way. Myrmex leverages the capacity of computers (monitoring, matching participants to tasks) to facilitate human activity (assigning and performing tasks). The project will: 1) develop new privacy-preserving methods for location and context tracking; 2) design and implement incentives for continuous use of a new social system; and 3) create and implement models of best times and places to notify and offer tasks for participants.
Broader impacts: The Myrmex system will help strengthen bonds and connectivity within a community and provide assistance for community members for running their daily lives. Matching tasks and people based on locations and paths will also help conserve resources and benefit the environment due to the reduced travel miles needed to perform a task. The project will provide multidisciplinary training for undergraduate and graduate students. It will also provide the research community with an anonymized dataset of human mobility patterns and task activities, and tools to analyze human behavior.
