The development of trust matrix for recognizing reliable content in social media

Abstract

Social media has emerged as a popular platform for users to share information about real-world events, particularly during disaster emergencies. However, disaster managers often having problem to gather an accurate information of the current situation since not all observations are made by reliable users. In this project, we address this problem by developing a trust matrix framework to identify the trustworthiness of the information shared on social media. Specifically, this project is focusing on flood disaster management and information shared on Twitter platform. The first objective of this project is to create text corpus for statistical and keyword analysis of a flood event shared on social media. Second objective is to develop a trust matrix for the flood events from social media. Third objective is to evaluate trust matrix for flood event using crowd-sourced data. The evaluation of trust matrix is done using real flood event dataset which are gathered between June and July 2018. To establishing a ground-truth for trust matrix framework, each data is mapped with actual flood event from news portal.