Designing A Trust and Reputation Framework for Spatial Tasks Allocations Using Fuzzy Method
Md Mujibur Rahman, Nor Aniza Abdullah · 2020
Unlike crowdsourcing, spatial crowdsourcing requires workers to travel to a specific physical location to complete a task. As a spatial crowdsourcing system can deploy any interested workers, the system is prone to unreliable and untrustworthy workers. Therefore, finding and allocating tasks to trustworthy workers is paramount to ensure the quality and sustainability of the system. Trust and reputation are key factors for evaluating workers’ trustworthiness. Unfortunately, current approaches to the evaluation of workers’ trustworthiness mostly rely on a single trust or reputation factor, and the trustworthiness is mainly based on a binary decision. To address these limitations, we propose a spatial task allocation framework that allocates every task in accordance with a workers’ degree of perceived trustworthiness, which is computed based on multi-criteria trust and reputation factors and a fuzzy method. Our approach evaluates workers’ reputations based on the ratings of workers’ past completed tasks, and workers’ trust is evaluated based on the sentiment analysis of their past tasks’ reviews. The attained reputation and trust values are then combined using a fuzzy inference system to obtain the degree of trustworthiness of each worker for every spatial task. Our experimental findings show that the proposed framework is able to find and allocate every spatial task to the most trustworthy workers from a pool of available workers. When evaluated against other baseline approaches, our proposed approach demonstrates the highest accuracy in terms of allocating every task to workers in accordance with their degree of perceived trustworthiness.