Key points are not available for this paper at this time.
• The study identifies the limitations of existing computational models in crisis informatics and proposition using VOSviewer for text network analysis to enhance the analysis of social media crisis datasets. • The crisis datasets were obtained from the literature, preprocessed to comply with VOSviewer requirements, and then analysed to identify common terms and generate visual maps. • The findings illustrate the interconnectedness of needs during a crisis and emphasise the importance of understanding relationships between terms. • The study provides insights into public discourse, helping crisis managers prioritise resources and improve response strategies. • The study advances the field of data science research methodologies by identifying and applying contemporary data analysis techniques to the domain of crisis response and management. Analysing social media data is crucial for crisis management organisations to make timely decisions. Researchers in crisis informatics have devised various methods and systems to process and classify large volumes of crisis-related social media data for effective crisis response and recovery. However, the complexity of previous solutions hampers the timely processing of this data, its visualisation, and its interpretation, which is necessary for effective crisis response. Hence, this study addresses this challenge by employing visualisation of similarities to analyse and visualise crisis datasets to aid crisis management and decision-making. The results reveal a "nine-cluster community” of relevant keywords comprising “Green, Brown, Red, Blue, Pink, Purple, Yellow, Orange, and Cyan” colours, in both binary and full count. Specifically, the findings reveal various keywords such as the needs for food, water, shelter, medicine, and electricity. Thereafter, the study discusses the implications of VOSviewer for analysing crisis data theoretically and practically.
Bukar et al. (Mon,) studied this question.