Abstract: The paper presents a four-phase action research project that was (and still is) conducted at the department of Information Management Customer Support and Operations (IM\CS&O) of a large multi-national company. The department is in charge of ICT-service continuity and has to produce ICT recovery plans that are integrated with the organization's overall Business Continuity plan. Interviews, Group Support System (GSS) technologies as well as a risk survey have been used to gather information and identify risks and threats. A systematic quantitative classification, measuring the impact of loss of ICT services on the company's business processes in terms of cost and risk will allow in the near future to utilize an economic decision model to prioritize the core activities of training and implementation of a recovery disaster plan. The research has made clear to the involved protagonists the necessity to share information, to develop awareness, and to formulate a shared recovery disaster plan to ensure ICT/business continuity and/or recovery when ICT disruptions occurs. © Proceedings ISCRAM 2004.

Abstract: During major disaster events, human operators in a crisis center will be overloaded with under-stress a flood of phone calls. As an increasing number of people in and around big cities do not master the native language, the need for automated systems that automatically process the context and content of information about disaster situations from the communicated messages becomes apparent. To support language-independent communication and to reduce the ambiguity and multitude semantics, we developed an icon-based reporting observation system. Contrast to previous approaches of such a system, we link icon messages to disaster events without using Natural Language Processing. We developed a dedicated set of icons related to the context and characteristic features of disaster events. The developed system is able to compute the probability of the appearance of possible disaster events using Bayesian reasoning. In this paper, we present the reporting system, the developed icons, the Bayesian model, and the results of two experiments.

Abstract: A central part of the work in Incident Commands (ICs) deals with handling messages that contain relevant information. Classification schemes for messages can be exploited by command staff and assisting tools to support this work, given that a common understanding of the scheme is shared among participants. We present user studies on two such classifications, which imply some disagreement among participants. Interpretations of the studies and a revised scheme are presented. All users in our studies are highly trained experts and represent the state of the art in german IC work.

Abstract: Microblogging sites such as Twitter can play a vital role in spreading information during “natural” or man-made disasters. But the volume and velocity of tweets posted during crises today tend to be extremely high, making it hard for disaster-affected communities and professional emergency responders to process the information in a timely manner. Furthermore, posts tend to vary highly in terms of their subjects and usefulness; from messages that are entirely off-topic or personal in nature, to messages containing critical information that augments situational awareness. Finding actionable information can accelerate disaster response and alleviate both property and human losses. In this paper, we describe automatic methods for extracting information from microblog posts. Specifically, we focus on extracting valuable “information nuggets”, brief, self-contained information items relevant to disaster response. Our methods leverage machine learning methods for classifying posts and information extraction. Our results, validated over one large disaster-related dataset, reveal that a careful design can yield an effective system, paving the way for more sophisticated data analysis and visualization systems.

Abstract: Many machine learning and natural language processing techniques, including supervised and domain adaptation algorithms, have been proposed and studied in the context of filtering crisis tweets. However, applying these approaches in real-time is still challenging because of time-critical requirements of emergency response operations and also diversities and unique characteristics of emergency events. In this paper, we explore the idea of building “generalized” classifiers for filtering crisis tweets that can be pre-trained, and are thus ready to use in real-time, while generalizing well on future disasters/crises data. We propose to achieve this using simple feature based adaptation with tweet representations based on word embeddings and also sentence-level embeddings, representations which do not rely on unlabeled data to achieve domain adaptations and can be easily implemented. Given that there are different types of word/sentence embeddings that are widely used, we propose to compare them to get a general idea about which type works better with crisis tweets classification tasks. Our experimental results show that GloVe embeddings in general work better with the datasets used in our evaluation, and that the supervised algorithms used in our experiments benefit from GloVe embeddings trained specifically on crisis data. Furthermore, our experimental results show that following GloVe, the sentence embeddings have great potential in crisis tweet tasks.

Abstract: Social media generates large amounts of almost real-time data which has proven valuable in disaster response. Specially for providing information within the first 48 hours after a disaster occurs. However, this potential is poorly exploited in operational environments due to the challenges of curating social media data. This work builds on top of the latest research on automatic classification of social media content, proposing the use of ensemble learning to help in the classification of social media images for disaster response. Ensemble methods use multiple learning algorithms to obtain better predictive performance than could be obtained from any of the constituent learning algorithms alone. Experimental results show that ensemble learning is a valuable technology for the analysis of social media images for disaster response,and could potentially ease the integration of social media data within an operational environment.

Abstract: In times of emergency, crisis response agencies need to quickly and accurately assess the situation on the ground in order to deploy relevant services and resources. However, authorities often have to make decisions based on limited information, as data on affected regions can be scarce until local response services can provide first-hand reports. Fortunately, the widespread availability of smartphones with high-quality cameras has made citizen journalism through social media a valuable source of information for crisis responders. However, analyzing the large volume of images posted by citizens requires more time and effort than is typically available. To address this issue, this paper proposes the use of state-of-the-art deep neural models for automatic image classification/tagging, specifically by adapting transformer-based architectures for crisis image classification (CrisisViT). We leverage the new Incidents1M crisis image dataset to develop a range of new transformer-based image classification models. Through experimentation over the standard Crisis image benchmark dataset, we demonstrate that the CrisisViT models significantly outperform previous approaches in emergency type, image relevance, humanitarian category, and damage severity classification. Additionally, we show that the new Incidents1M dataset can further augment the CrisisViT models resulting in an additional 1.25% absolute accuracy gain.

Abstract: We report on the development of a classifier to identify Twitter users contributing first-hand information during a disaster. Identifying such users helps social media monitoring teams identify critical information that might otherwise slip through the cracks. A parallel study (St. Denis et al., 2020) demonstrates that Twitter user filtering creates an information-rich stream of content, but the best way to approach this task is unexplored. A user's profile contains many different “modalities” of data, including numbers, text, and images. To integrate these different data types, we constructed a multimodal neural network that combines the loss function of all modalities, and we compared the results to many individual unimodal models and a decision-level fusion approach. Analysis of the results suggests that unimodal models acting on Twitter users' recent tweets are sufficient for accurate classification. We demonstrate promising classification of Twitter users for crisis response with methods that are (1) easy to implement and (2) quick to both optimize and infer.

Abstract: Social media data analysis is important for disaster management. Lots of prior studies have focused on classifying a tweet based on its text or based on its images, independently, even if the tweet contains both text and images. Under the assumptions that text and images may contain complementary information, it is of interest to construct classifiers that make use of both modalities of the tweet. Towards this goal, we propose a multimodal classification model which aggregates text and image information. Our study aims to provide insights into the benefits obtained by combining text and images, and to understand what type of modality is more informative with respect to disaster tweet classification. Experimental results show that both text and image classification can be improved by the multimodal approach.

Abstract: In this paper, we show that the text-image relationship of disaster tweets can be used to improve the classification of tweets from emergency situations. To this end, we introduce DisRel, a dataset which contains 4,600 multimodal tweets, collected during the disasters that hit the USA in 2017, and manually annotated with coherence image-text relationships, such as Similar and Complementary. We explore multiple models to detect these relationships and perform a comprehensive analysis into the robustness of these methods. Based on these models, we build a simple feature augmentation approach that can leverage the text-image relationship. We test our methods on 2 tasks in CrisisMMD: Humanitarian Categories and Damage Assessment, and observe an increase in the performance of the relationship-aware methods.

Abstract: Social media applications such as Twitter and Facebook are fast becoming a key instrument in gaining situational awareness (understanding the bigger picture of the situation) during disasters. This has provided multiple opportunities to gather relevant information in a timely manner to improve disaster response. In recent years, identifying crisis-related social media posts is analysed as an automatic task using machine learning (ML) or deep learning (DL) techniques. However, such supervised learning algorithms require labelled training data in the early hours of a crisis. Recently, multiple manually labelled disaster-related open-source twitter datasets have been released. In this work, we create a large dataset with 186,718 tweets by combining a number of such datasets and evaluate the performance of multiple ML and DL algorithms in classifying disaster-related tweets in three settings, namely ``in-disaster'', ``out-disaster'' and ``cross-disaster''. Our results show that the Bidirectional LSTM model with Word2Vec embeddings performs well for the tweet classification task in all three settings. We also make available the preprocessing steps and trained weights for future research.

Abstract: Sentiment analysis has been widely researched in the domain of online review sites with the aim of generating summarized opinions of product users about different aspects of the products. However, there has been little work focusing on identifying the polarity of sentiments expressed by users during disaster events. Identifying sentiments expressed by users in an online social networking site can help understand the dynamics of the network, e.g., the main users' concerns, panics, and the emotional impacts of interactions among members. Data produced through social networking sites is seen as ubiquitous, rapid and accessible, and it is believed to empower average citizens to become more situationally aware during disasters and coordinate to help themselves. In this work, we perform sentiment classification of user posts in Twitter during the Hurricane Sandy and visualize these sentiments on a geographical map centered around the hurricane. We show how users' sentiments change according not only to users' locations, but also based on the distance from the disaster.