The Research Of Emergency Response System Based On Data Mining Technical

In recent years, accompanied by human development process, there have been frequently reports of many kinds of disaster and emergency incidents, in additional to natural disasters, kinds of terrorist incidents, the spread of the disease, man-made disasters, and criminal cases have also gradually changed in some way. Under this high pressure, how to rapidly and effectively response to these emergency incident, and how to dispatch suitable resources to handle the incidents becomes the key points in this responding procedure. Predefined Action Plan (PAP) is the scientific summary up on the historic response procedures; it plays a very important guiding role when handling the similar following occurrence. Once PAP is established, the following issue need to be solved is how to search and execute the plans at the first time and to eliminate the human factor errors.Case-Based Reasoning (CBR) technology is now widely used in artificial intelligence, statistics, machine learning and databases etc cross-cutting areas. In this paper, we will study and explore the way to apply CBR technology in management of predefined action plan (PAP) in an emergency response system. The traditional way to create PAP is summarized by experienced operator from the historic handling process. The personal limitation will give big impaction on the created PAP. While after we adopt CBR technology, the PAP is created automatically from the historic data. In this paper, we will demonstrate a way to generate PAP from the detailed responding action steps executed in emergency response procedure, so the PAP is also a list of action steps, it can be executed directly. We will design and develop a CBR based PAP management system in this paper, and prove it valid through a sample emergency system for transportation bureau.