An assertive reasoning method for emergency response management based on knowledge elements C4.5 decision tree

The correct selection of knowledge elements is the key to emergency management. Using emergency knowledge elements, this study constructs an assertive reasoning selection methodology by improving acquisition on the balance coefficient in the C4.5 algorithm. Through hierarchical representation, a two level model selection method, based on a top model construction process as well as an underlying model selection process, is proposed. Specifically, the top model construction process is based on assertive reasoning and an underlying model selection process is based on the C4.5 decision tree. This method reduces the requirements for emergency domain knowledge, and improves the accuracy and timeliness. Lastly, the methodology is applied to the 2015 Tianjin explosions as a case study. (C) 2018 Elsevier Ltd. All rights reserved.