Time series sentiment analysis (SA) of relief operations using social media (SM) platform for efficient resource management

Abstract

The ease of access to the internet has sparked a worldwide interest in SM in recent years. The possibilities for the use of SM as a potential source in improving the management of disasters are increasing. Although there is an existing research ongoing with the analysis of the usage of SM during disasters, but none of the works have exploited the use of explainable artificial intelligence (XAI) for the validation of the transparency of the ML models. The contribution of this research is two-fold: Firstly, ML based time series analysis for relief operations using social media information with situational information gathered using Twitter from the users and resource providers and secondly, XAI has been used to increase the transparency and understand-ability of model decisions. For the case study, public dataset from Nepal, Italy earthquakes, COVID-19 dataset along with originally collected Twitter dataset has been considered. The performance of the extreme gradient boosting (xgboost) model is relatively superior than other techniques with 10-fold training mean accuracy of 87.17%. Thus, the experiments conclude the possibility of automation for the time series analysis for optimal relief operation management to serve the victims in the most efficient way and to control legal and administration implications.