Trust me, my neighbors say it's raining outside: ensuring data trustworthiness for crowdsourced weather stations

摘要: Decision making in utilities, municipal, and energy companies depends on accurate trustworthy weather information predictions. Recently, crowdsourced personal stations (PWS) are being increasingly used to provide a higher spatial temporal resolution of measurements. However, tools methods ensure the trustworthiness data real-time lacking. In this paper, we present Reputation System for Crowdsourced Rainfall Networks (RSCRN) assign trust scores region. Using real PWS from Weather Underground service high flood risk region Norfolk, Virginia, evaluate performance proposed RSCRN. The method is able converge confident score within 10--20 observations after installation. Collectively, results indicate that derived RSCRN can reflect collective measure PWS, ensuring both useful modeling decision-making future.