Analyzing the Reliability of Unstructured Data for Urban Rainfall Pattern Studies—A Case Study from Zhengzhou

Abstract

Due to the insufficient number and uneven distribution of urban rainfall stations, research on urban flooding disasters is limited. With the development of big data research, many scholars have applied big data to natural disaster research. In this paper, we analyzed the reliability of unstructured data from the urban rainfall patterns studies using the measured rainfall data for Zhengzhou City. First, web crawler technology was used on Sina Weibo, one of China’s largest social platforms, to obtain the unstructured data related to rainfall. The fuzzy recognition method was used to analyze the rain patterns of the measured rainfall data and the unstructured data, which verified the reliability of the unstructured data in the analysis of the urban rainfall patterns. Taking Zhengzhou City as an example, it was found that the matching degree of rain pattern recognition results was 45%, between the unstructured data and measured data. This showed that the application of the unstructured data in the analysis of the urban rainfall patterns has a certain degree of reliability.