Exploiting Twitter for Informativeness Classification in Disaster Situations

Natural disasters cause significant damage, casualties and economical losses. Twitter has been used to support prompt disaster response and management because people tend to communicate and spread information on public social media platforms during disaster events. To retrieve real-time situational awareness (SA) information from tweets, the most effective way to mine text is using natural language processing (NLP). Among the advanced NLP models, the supervised approach can classify tweets into different categories to gain insight and leverage useful SA information from social media data. However, high-performing supervised models require domain knowledge to specify categories and involve costly labelling tasks. This research proposes a guided latent Dirichlet allocation (LDA) workflow to investigate temporal latent topics from tweets during a recent disaster event, the 2020 Hurricane Laura. With integration of prior knowledge, a coherence model, LDA topics visualisation and validation from official reports, our guided approach reveals that most tweets contain several latent topics during the 10-day period of Hurricane Laura. This result indicates that state-of-the-art supervised models have not fully utilised tweet information because they only assign each tweet a single label. In contrast, our model can not only identify emerging topics during different disaster events but also provides multilabel references to the classification schema. In addition, our results can help to quickly identify and extract SA information to responders, stakeholders and the general public so that they can adopt timely responsive strategies and wisely allocate resource during Hurricane events.

Social Media (SM) data are growing, and SM is becoming an acceptable part of daily life for billions of people around the world. Extracting information from Social Networking Sites (SNS) can provide great challenges as well as opportunities. Using SM data beyond day-to-day communication can provide additional values. There is much research and many products that are dedicated to take SNS beyond communication channels. In our research, we are going beyond specific tools inherent to the SM tools, such as Hashtag mentions and Like counts. Instead it will use text-based modeling, data mining techniques, natural process language, machine language, etc. to understand SM content to produce numeric ratings. The final contribution of this research is building a SM users' rating model for an event using SM data. At this point of our research, we are laying out a road map.

Social media provide valuable information for disaster risk reduction as well as a disaster emergency response. Previous studies claim that social media data can be used in the emergency response effort, which is worthwhile for the Government, BNPB, as well as the community, to get a direct and credible information about the ongoing disaster situations. However, the social media data has not been widely used for emergency response and situational awareness in Indonesia. Therefore, in this paper we implement BERT method applied to a set of tweets related to Jakarta Flood in early 2020. We have crawled the tweet dataset during the Jakarta flood disaster in early 2020 which had become a trending topic on Twitter. We aim to identify the relevance tweets that may provide useful information related to emergency responses in disaster management. The experimental results indicate promising outcomes. However, the quality of the dataset greatly influences the system performance.

Social media data have emerged as a new source for detecting and monitoring disaster events. A number of recent studies have suggested that social media data streams can be used to mine actionable data for emergency response and relief operation. However, no effort has been made to classify social media data into stages of disaster management (mitigation, preparedness, emergency response, and recovery), which has been used as a common reference for disaster researchers and emergency managers for decades to organize information and streamline priorities and activities during the course of a disaster. This paper makes an initial effort in coding social media messages into different themes within different disaster phases during a time-critical crisis by manually examining more than 10,000 tweets generated during a natural disaster and referencing the findings from the relevant literature and official government procedures involving different disaster stages. Moreover, a classifier based on logistic regression is trained and used for automatically mining and classifying the social media messages into various topic categories during various disaster phases. The classification results are necessary and useful for emergency managers to identify the transition between phases of disaster management, the timing of which is usually unknown and varies across disaster events, so that they can take action quickly and efficiently in the impacted communities. Information generated from the classification can also be used by the social science research communities to study various aspects of preparedness, response, impact and recovery.

Flooding management requires collecting real-time onsite information widely and rapidly. As an emerging data source, social media demonstrates an advantage of providing in-time, rich data in the format of texts and photos and can be used to improve flooding situation awareness. The present study shows that social media data, with additional information processed by Artificial Intelligence (AI) techniques, can be effectively used to track flooding phase transition and locate emergency incidents. To track phase transition, we train a computer vision model that can classify images embedded in social media data into four categories - preparedness, impact, response, and recovery - that can reflect the phases of disaster event development. To locate emergency incidents, we use a deep learning based natural language processing (NLP) model to recognize locations from textual content of tweets. The geographic coordinates of the recognized locations are assigned by searching through a dedicated local gazetteer rapidly compiled for the disaster affected region based on the GeoNames gazetteer and the US Census data. By combining image and text analysis, we filter the tweets that contain images of the “Impact” category and high-resolution locations to gain the most valuable situation information. We carry out a manual examination step to complement the automatic data processing and find that it can further strengthen the AI-processed results to support comprehensive situation awareness and to establish a passive hotline to inform rescue and search activities. The developed framework is applied to the flood of Hurricane Harvey in the Houston area.

ABSTRACTAs the world's climate changes, species are undergoing range shifts. Range shifts are generally documented using databases such as the Global Biodiversity Information Facility (GBIF), which largely contain data from monitoring schemes and wildlife surveys. Such databases have two major limitations: (i) data may be spatially biased because traditionally surveyed areas are in rural habitats and (ii) there is a time lag between formal monitoring and survey data collection and assimilation into GBIF, which means rapid range shifts cannot be tracked. Alternative data sources, such as social media, could provide information on species distributions and range shifts that compensate for spatial biases in GBIF records because social media data may be collected outside traditionally surveyed areas. Such data are also usually shared online immediately after a wildlife sighting. The complementarity of GBIF and social media data, however, has rarely been assessed, particularly when tracking range shifts. Despite their potential utility, social media data may be particularly prone to temporary trends or geographic variation in behaviour that are not understood. We lack tools with which to counter these biases. To address these knowledge gaps, we compare the habitat usage revealed by biological records of the Jersey tiger moth from GBIF and from social media data sources (Instagram and Flickr). We develop a novel method to investigate recorder bias in social media data and compare between data sources. We find that biological records from Instagram reveal greater than expected occurrence in urban environments. Recorder effort differs notably between data sources and Instagram complements GBIF by recording species in areas unaccounted for by GBIF. By incorporating recorder effort metrics, data from social media sources could be used to improve monitoring of range‐shifting species in urban spaces.

Indonesia is a country that has a relatively high level of vulnerability to natural disasters. Data on natural disaster events in Indonesia in 2023 is 4,800 disaster events and in 2024 there are 2,093 disaster events. School and student preparedness is an important key in developing strategies when facing disasters because there are still disasters that cannot be predicted when they will occur. Guidance and counseling teachers must have basic knowledge and the ability to deal with disaster situations well before a disaster, during a disaster and after a disaster. Providing guidance and counseling services is one of the right steps in an effort to increase student readiness. This research uses literature study as a research method. The research results show that guidance and counseling (BK) teachers play an important role in providing disaster preparedness education such as information about types of disasters, how to protect themselves from disasters, as well as steps that students must take when a disaster occurs. In addition, guidance and counseling teachers play a role in supporting psychosocial services to students by helping train them to manage their anxiety and fear of disasters and providing emotional management strategies such as relaxation techniques to increase appropriate psychosocial support..

PurposeUsing intermodal transportation is vital for the delivery of relief supplies when single mode alternative becomes unusable or infeasible. The objective of this paper is to investigate the use of intermodal freight transportation in humanitarian logistics.MethodsThis paper first identifies the differences between multimodal and intermodal transportation. Then, we examine the use of each transportation mode for specific disaster types and phases. When combinations of transportation modes (i.e. air, road, rail and sea) for intermodal transportation are considered together with different disaster types (e.g. earthquake, flood and famine), the feasible decision space becomes rather large. To explore this decision space, we have reviewed the academic and practitioner studies as well as several non-governmental organizations (NGO)’ disaster archives.ResultsFrom this exploration, we developed a transportation mode/disaster-type combination matrix and a transportation mode/disaster-phase combination matrix. We then discuss examples of real life usage of intermodal transportation in humanitarian logistics and share our findings and analyses. Of 369 academic humanitarian logistics articles, only 20 have mentioned transportation mode changes. In practitioner studies, we found a decreasing percentage of the usage of slower modes (e.g. sea and rail) in the disaster response phase over time. We were not able to find a significant relationship between a specific transportation mode and a specific disaster-type or -phase. Road transportation seems to cover most of the disaster operations regardless of the disaster-type or -phase.ConclusionsWe can conclude that intermodality and the transportation unit concept is not being studied extensively in humanitarian logistics. Most of the relief organizations do not share transported freight amounts in their reports and those that do share transported freight amounts in their reports do not explicitly mention mode changes. We discuss the enablers of and obstacles to the effective use of intermodal transportation in humanitarian logistics and propose future research directions. We anticipate that intermodal transportation in humanitarian logistics will garner greater research attention and increased utilization in coming years.

Is user-generated social media content useful for informing planning and management of emergency events? – An investigation of an active shooting event in a U.S. Airport

People are increasingly sharing information on social media during disaster events. This information could be valuable to emergency responders, but there remain challenges for using it to inform response efforts---including filtering relevant information from the large volumes of noise. Previous research has largely focused on identifying information that can contribute to a generalized concept of situational awareness. Our work explores the value of approaching this problem from a different perspective---one of actionablity---with the idea that information relevance may vary across responder role, domain, and other factors. This approach asks how we can get the right information to the right person at the right time? We interviewed and surveyed diverse responders to understand what "actionable" information is, allowing that actionability might differ from one responder to another. Through the findings, we (a) offer a nuanced understanding of actionability and differentiate it from situational awareness; (b) describe responders' perspective of what distinguishes good information when making rapid judgments; and (c) suggest opportunities for augmenting social media use to highlight information that needs immediate attention. We offer researchers an opportunity to frame different models of actionability to suit the requirements of a responding role.

We explored the transformation of social media content into three radically different formats; a book, a triptych of photographs and film.Using interviews, we investigated users' responses to their remediated data.Our findings were used to develop a digital curation process to assist the development of different tools for social media analysis and display.Our paper establishes design implications to aid reflection on social media use through remediation. Increasingly our digital traces are providing new opportunities for self-reflection. In particular, social media (SM) data can be used to support self-reflection, but to what extent is this affected by the form in which SM data is presented? Here, we present three studies where we work with individuals to transform or remediate their SM data into a physical book, a photographic triptych and a film. We describe the editorial decisions that take place as part of the remediation process and show how the transformations allow users to reflect on their digital identity in new ways. We discuss our findings in terms of the application of Goffman's (1959) self-presentation theories to the SM context, showing that a fluid rather than bounded interpretation of our social media spaces may be appropriate. We argue that remediation can contribute to the understanding of digital self and consider the design implications for new SM systems designed to support self-reflection.

Compared to natural language processing research investigating suicide risk prediction with social media (SM) data, research utilizing data from clinical settings are scarce. However, the utility of models trained on SM data in text from clinical settings remains unclear. In addition, commonly used performance metrics do not directly translate to operational value in a real-world deployment. The objectives of this study were to evaluate the utility of SM-derived training data for suicide risk prediction in a clinical setting and to develop a metric of the clinical utility of automated triage of patient messages for suicide risk. Using clinical data, we developed a Bidirectional Encoder Representations from Transformers-based suicide risk detection model to identify messages indicating potential suicide risk. We used both annotated and unlabeled suicide-related SM posts for multi-stage transfer learning, leveraging customized contemporary learning rate schedules. We also developed a novel metric estimating predictive models' potential to reduce follow-up delays with patients in distress and used it to assess model utility. Multi-stage transfer learning from SM data outperformed baseline approaches by traditional classification performance metrics, improving performance from 0.734 to a best F1 score of 0.797. Using this approach for automated triage could reduce response times by 15 minutes per urgent message. Despite differences in data characteristics and distribution, publicly available SM data benefit clinical suicide risk prediction when used in conjunction with contemporary transfer learning techniques. Estimates of time saved due to automated triage indicate the potential for the practical impact of such models when deployed as part of established suicide prevention interventions. This work demonstrates a pathway for leveraging publicly available SM data toward improving risk assessment, paving the way for better clinical care and improved clinical outcomes.

Social media has become a nondetachable part of our life, with the exponential growth of usage in the past decade. Social sites like Twitter, Facebook, Instagram, Flickr, Weibo, etc., with their millions of user base, apart from being a source of entertainment, has proven to be a very useful mean for public opinion generation, news propagation and information broadcasting by authorities. Social media data analysis has been a popular research area for the past few years. Detecting subevents from social media posts to identify an unusual event that requires special attention, especially in a disaster situation, is one of the key researches in this domain. In this article, we have proposed a novel biclustering-based subevent detection method from the Twitter dataset for retrospective analysis of disaster events. First, we have clustered the data matrix using spectral co-clustering. Then we identified subevents (words) and formulated a ranking framework to find the top-ranked subevents within the clusters. Finally, through statistical analysis, we have shown that the proposed framework works better than other existing subevent detection methods.

With the rapid development of Internet information technology, the advantages of social media in terms of speed, content, form, and effect of communication are becoming increasingly significant. In recent years, more and more researchers have paid attention to the special value and role of social media tools in disaster information emergency management. Weibo is the most widely used Chinese social media tool. To effectively mine and apply the emergency function of disaster situation microblogs, a disaster situation information discovery and collection system capable of online incremental identification and collection are constructed for massive and disordered disaster microblog text streams. First, based on the deep learning- trained word vector model and a large-scale corpus, an unsupervised short-text feature representation method of disaster situation Weibo information is developed. According to the experimental results of the feature combination test and the training set scale test, the SVM algorithm was selected for disaster microblog information classification, which realized effective identification of disaster situation micro-bloggings. Then, the temporal information similarity and geographic information similarity are used to improve the single text similarity algorithm, and a Chinese disaster event online real-time detection model is constructed. Furthermore, the disaster-affected areas can be achieved in real-time based on the detection results. By crawling and classifying the micro-bloggings from the disaster-affected areas, it is possible to realize the incremental identification and collection of online disaster situation Weibo information. Finally, the empirical analysis of disaster events such as the “Leshan Earthquake” shows that the real- time intelligent identification and collection system for disaster situation Weibo micro-bloggings developed in this paper can obtain large-scale and useful data for disaster emergency management, which proving that this system is effective and efficient.

Tie strength is an essential concept in identifying different kind of social ties - strong ties and weak ties. Most present studies that evaluated tie strength from social media were carried out in a controlled environment and used private/closed social media data. Even though social media has become a very important way of networking in professional events, access to such private social media data in those events is almost impossible. There is very limited research on how to facilitate networking between event participants and especially on how to automate this networking aspect in events using social media. Tie strength evaluated using social media will be key in automating this process of networking. To create such tie strength based event participant recommendation systems and tools in the future, first, we need to understand how to evaluate tie strength using publicly available social media data. The purpose of this study is to evaluate tie strength from publicly available social media data in the context of a professional event. Our case study environment is community managers’ online discussions in social media (Twitter and Facebook) about the CMAD2016 event in Finland. In this work, we analyzed social media data from that event to evaluate tie strength and compared the social media analysis-based findings with the individuals’ perceptions of the actual tie strengths of the event participants using a questionnaire. We present our findings and conclude with directions for future work.