Poster: On Cost-Sensitive Task Allocation in Social Sensing: An Online Learning Approach

Abstract

Listen

Translate article

Social sensing has emerged as a new sensing paradigm where human sensors collectively report measurements about the physical world. This paper focuses on the cost-sensitive task allocation problem in social sensing where the goal is to effectively allocate sensing tasks to the human sensors to meet the desirable data quality requirement of the applications while minimizing the sensing cost. While recent progress has been made to tackle the cost-sensitive task allocation problem, an important challenge has not been well addressed, namely real time task allocation, the task allocation schemes need to respond quickly to the potential large dynamics of the measured variables in social sensing. To address this challenge, this paper presents a Cost-Sensitive Task Allocation (CSTA) scheme inspired by techniques from online learning. The preliminary results show that our new scheme significantly outperforms the-state-of-the-art baselines.