Trustworthiness of Data in IoT Crowd Sensing Environments

In the context of mobile crowd sensing, an input data that is collected from multiple people may result to be incorrect or malicious and ultimately alter predictions, data reliability is one of the crucial factors taken into account. In this research, the reliability of the data is assessed using crowdsensing data from the transportation and traffic domain, where numerous people contribute data on the flow of traffic that may be used to calculate the shortest route and other traffic-related metrics. This project attempts to build a voting-based trustworthiness model. Game theory is utilized to encourage users to contribute to crowdsensing data, and metrics including voting capacity, user reputation, user pay-off, and trustworthiness are determined depending on the user’s votes. Another approach used to assess user data trust is based on the concepts of experience and reputation, with experience being determined by how frequently a user interacts with other users. Based on their reputation scores, each user receives rewards in the form of badges. By repeatedly running both algorithms, the stability of each is examined.