Participation Selection in Mobile Crowd Sensing with Diversity Ensures

Mobile Crowd Sensing (MCS) systems are reshaping the data collection process by utilizing the capabilities of smart devices. Traditional MCS models employ cost-effective algorithms, treating their participants as unit workers. However, this approach overlooks the nuance of the MCS system, which is not limited to collecting quantitative data such as weather or traffic conditions. Instead, MCS also has the potential to perform opinionative tasks. By selecting diverse participants to offer opinionative data, we can gain unique but easily overlooked perspectives. This paper proposes a model to improve diversity in participant selection to ensure that diverse social identities and voices are considered, thereby promoting equality. We first divided participants into distinct similarity groups, each characterized by shared behavior patterns, responses, interests, etc., and then selected participants by prioritizing diversity. The simulations presented in this paper demonstrated the effective-ness of our proposed model.