OPTIMIZED TASK ASSIGNMENT IN SPATIAL CROWDSOURCING WITH WORKERS PRIVACY PROTECTION

Abstract

Crowdsourcing has opened not just opportunities, but also options for people to work together without boundaries, and it is the future of collaboration. Spatial crowdsourcing (SC) is a form of location-based crowdsourcing. With the spread use of mobile phones and smart devices that can share location, spatial crowdsourcing gained a lot of attention, especially in ride-hailing services. Task assignments are the main function of any crowdsourcing platform. Many recent studies focused on the workers privacy without taking into consideration the overall task matching utility score. This research, adapts the batch matching method to maximize the overall task assignment and minimize the crowd workers total travel distance in spatial crowdsourcing while maintaining the crowd workers privacy using the (DCentroid) scheme. This research introduces (TASC) a task assignment approach with crowd worker location privacy protection in SC. The research theoretically analyzes TASC framework and guarantees the crowd workers privacy while minimizing the crowd worker travel distance, and system overhead. This research evaluates the performance of TASC a spatial crowdsourcing task assignment approach to increase the task assignment rate while preserving the location privacy of the crowd workers. The overall experiments on real-world data sets show that TASC framework results in minimize the total travel distance of assigned tasks without the disclosure of crowd workers' locations.