Text Quality Analysis of Emergency Response Plans

Abstract

Emergency response plans are regarded as effective guidance for natural disasters and these plans describe emergency response processes in natural language. More specifically, they are textual process descriptions and describe not only how all departments perform their own response tasks, but also how different departments interact with each other. Analyzing text quality of emergency response plans as a typical evaluation approach is an important concern in emergency responses. Because of the flexibility of natural language, emergency response plans normally contain unwanted ambiguities, and it is difficult to check consistency and completeness. Automatic text quality analysis of emergency response plans written in Chinese from the perspective of process descriptions is proposed in this paper. Firstly, three types of response tasks including message sending tasks, message receiving tasks and regular tasks are extracted through Bi-LSTM-CRF networks (a Conditional Random Fields network is combined with a Bidirectional Long Short-Term Memory network). Then, a series of text quality analysis rules associated with extracted response tasks are created. These rules focus on the progressive relationship of four levels of emergency responses, completeness of response tasks, ambiguity and redundancy of emergency response plans. Finally, real-world data is collected to validate the proposed approach, which consists of four types of emergency response plans of natural disasters including district, municipal, provincial and national emergency response plans. It is demonstrated that the proposed approach can be used to facilitate revisions and improvements of emergency response plans. INDEX TERMS Emergency response plans, process information extraction, text quality analysis, emergency plan evaluation, textual process descriptions.

Traffic accident duration prediction using multi-mode data and ensemble deep learning

“…The results reported herein also broadly support the work of other studies demonstrating that the processing and extraction of text features is a complex issue in the field of traffic accident duration prediction [ 11 , 30 , 32 ]. This follows from the surprising results demonstrating that the prediction performances obtained by deep-learning-based models when employing data modes that included unstructured text data extracted using pre-trained models (i.e., Modes 3–6) were generally not better than the performances of conventional machine learning methods using only standard structured feature variables (i.e., Modes 1 and 2).…”

“…The proposed approach extracted message sending tasks, message receiving tasks, and regular tasks. This was achieved through the use of Bi-LSTM-CRF networks, which combined a Conditional Random Fields network with a Bidirectional Long Short-Term Memory network [ 30 ].…”

Decision Support for Plan Adaptation in Unforeseen Situations

“…When using this document, the response team knows who is carrying out specific actions, the resources available, and coordinating all efforts. Besides the diversity of representations and the difficulty in formalizing plans in complex environments, it is possible to identify a set of common elements to describe them (Alexander, 2016;Bénaben et al, 2016;Ferreira et al, 2015;Guo et al, 2020;Penadés et al 2011;Savino et al, 2014) (Figure 4): action, a task performed to achieve a goal, which takes the phenomenon from one state to another by changing the state variables values; state, variables that characterize the phenomenon at a specific moment, thus having associated values that may change over time; resource, necessary elements to perform an action or that influence its performance; event, which may occur during handling and has an impact on the state variables; and a goal, what should be achieved. These elements are not always formally structured, being in the response teams' feeling, but should be instantiated in plans for complex environments.…”

Rapid Generation of Emergency Response Plans for Unconventional Emergencies

“…The results achieved, while beneficial to assist decision-making of unconventional emergency response, do not necessarily result in a complete ERP because much more comprehensive information about handling an unconventional emergency should be included in ERPs. Moreover, ERPs currently used in real-world emergency management agencies are mostly in the form of text [11], because natural language is more human-friendly than formalized models [12], [13]. Because of the gap between formalized decisionmaking models and textual ERPs, the existing theoretical methodologies proposed in the unconventional emergency management literature are hardly to be leveraged to construct textual ERPs straightaway.…”