Aim: Given the heavy damages caused by natural hazards such as earthquakes to the built environment, the present study aims to assess the vulnerability of residential areas to earthquake risk in the city of Neyshabur. First, it delineates the earthquake-prone areas of the city and then evaluates the vulnerability levels of different urban areas.
Material & Method: The present research is descriptive-analytical. The research is applied. Data collection was library-based and field-based. The most important software used in this research is Weka data mining software, for the purpose of implementing machine learning algorithms, and ArcMap software, in order to prepare the data for algorithm implementation. The data used in this research include two indicators: natural factors and human factors.
Finding: The findings of this research indicated that the four supervised algorithms used—Random Tree, Rep Tree, MP5, and Random Forest—were able to accurately zone areas at risk of earthquakes in the city of Neyshabur.
Conclusion: Based on the research findings, it can be stated that approximately 42 percent of the area of the city of Neyshabur is located in a zone of high to very high vulnerability. Therefore, the city of Neyshabur, especially in the central areas, is highly vulnerable to the natural phenomenon of earthquakes.
Innovation: Due to limited access to information regarding vulnerability assessment against earthquakes within the geographical area of Neyshabur, this research initially focused on zoning the earthquake risk in Neyshabur County. Subsequently, by integrating human factors effective in increasing the severity of earthquakes within the urban area with the outputs of machine algorithms, the vulnerability of the urban area was assessed. The machine learning algorithms used in the research are of the integrated type.