Abstract: Assessing the effectiveness of crisis response is key to improving preparedness and adapting policies. One method for response evaluation is reviewing actual response activities and interactions. Response reports are often available in the form of natural language text data. Analyzing a large number of such reports requires automated or semi automated solutions. To improve the trustworthiness of methods for this purpose, we empirically validate the reliability of three relation extraction methods that we used to construct interorganizational collaboration networks by comparing them against human-annotated ground truth (crisis-specific situational reports and tweets). For entity extraction, we find that using a combination of two off-the-shelf methods (FlairNLP and SpaCy) is optimal for situational reports data and one method (SpaCy) for tweets data. For relation extraction, we find that a heuristics-based model that we built by leveraging word co-occurrence and deep and shallow syntax as features and training it on domain-specific text data outperforms two state-of-the-art relation extraction models (Stanford OpenIE and OneIE) that were pre-trained on general domain data. We also find that situational reports, on average, contain less entities and relations than tweets, but the extracted networks are more closely related to collaboration activities mentioned in the ground truth. As it is widely known that general domain tools might need adjustment to perform accurately in specific domains, we did not expect the tested off-the-shelf tools to perform highly accurately. Our point is to rather identify what accuracy one could reasonably expect when leveraging available resources as-is for domain specific work (in this case, crisis informatics), what errors (in terms of false positives and false negatives) to expect, and how to account for that.

Abstract: It is generally agreed in crisis management that a comprehensive visualization of the situation is crucial for an appropriate situational awareness of the staff personnel in control rooms. Therefore an expert group of fire officers in the German State North Rhine Westphalia developed a system for this purpose known as the “tactical wall”. The core of the system is a situation map of the relevant area with so-called tactical signs, i. e. defined graphic symbols for hazards, response units and tactical measures. Moreover, the assignment of response units to tactical sectors or staging areas as well as other relevant information such as the management organization is displayed at defined places within the wall. While the system was purely manual in its original version, a new digital version was recently developed. The user interfaces of this system are web-based and can by intuitively operated after a minor training effort.

Abstract: During the COVID-19 health crisis, local public offcials expend considerable energy encouraging citizens to comply with prevention measures in order to reduce the spread of infection. During the pandemic, mask-wearing has been accepted among health offcials as a simple preventative measure; however, some local areas have been more likely to comply than others. This paper explores methods to better understand local attitudes towards mask-wearing as a tool for public health offcials' situational awareness when preparing public messaging campaigns. This exploration compares three methods to explore local attitudes: sentiment analysis, n-grams, and hashtags. We also explore hashtag co-occurrence networks as a starting point to begin the filtering process. The results show that while sentiment analysis is quick and easy to employ, the results oer little insight into specific local attitudes towards mask-wearing, while examining hashtags and hashtag co-occurrence networks may be used a tool for a more robust understanding of local areas when attempting to gain situational awareness.