Enhancing User Experience of Task Assignment in Spatial Crowdsourcing: A Self-Adaptive Batching Approach

Lai Qian,An Liu,Guanfeng Liu,Zhixu Li,Yu Wang,Fei Zhu

doi:10.1109/access.2019.2940028

Abstract

Faced with the explosive demand of real-world applications, spatial crowdsourcing has attracted much attention, in which task assignment algorithms take the dominant role in the past few years. On the one hand, most recent studies concentrate on maximizing the overall benefits of the platform, ignoring the fact that user experience also plays an essential role in task allocation. On the other hand, they focus on matching, that is, how to assign tasks, rather than batching, that is, when to make assignment. In fact, user experience also depends on batching, but this is largely overlooked by current studies. In this paper, we propose a self-adaptive batching mechanism to enhance user experience in spatial crowdsourcing. With appropriate start-up timestamps, previous matching methods can perform better. Multi-armed bandit algorithm in reinforcement learning is adopted to split the batch dynamically according to historical current states. Extensive experimental results on both real and synthetic datasets demonstrate the effectiveness and efficiency of the proposed approach.

Highlights

With the rapid development of mobile Internet and sharing economy, a novel framework called spatial crowdsourcing (SC) is proposed to meet with the growing demand for suitable solutions to the problem of allocating spatial tasks
Compared with similar work [8], our work focuses on the user experience instead of the overall utility of the platform, and we will focus on the timing of allocation and the impact of current real-time user supply and demand on batch size settings
We present a Multi-armed bandit (MAB)-based self-adaptive batching approach to optimize SC task assignment, which can adapt to conventional assignment algorithms and the changing of real-time supply-demand relationship

Summary

INTRODUCTION

With the rapid development of mobile Internet and sharing economy, a novel framework called spatial crowdsourcing (SC) is proposed to meet with the growing demand for suitable solutions to the problem of allocating spatial tasks. Existing solutions to this problem is either real-time or fixed batch Both of them sometimes have unsatisfied performance in terms of user experience. If the batch size is set to 2, the platform will wait for 2 units of time after each allocation and perform the allocation In this example, we do not consider the worker waiting time, and if the worker is not assigned to the task, he will be treated as an idle worker to join the batch for further allocations. Compared with similar work [8], our work focuses on the user experience instead of the overall utility of the platform, and we will focus on the timing of allocation and the impact of current real-time user supply and demand on batch size settings.

RELATED WORK

PROBLEM DEFINITIONS

EXPERIMENTS SETUP We use both synthetic and real-world datasets in our experiments

Findings

CONCLUSION