Gender-Specific Hotspot Detection of Literate and Workers in Uttar Pradesh, India using a Rough Graph-based Approach

Abstract

Spatial polygon data represents the area or region of specific events, such as disease cases, crime, medical facilities, earthquakes, and fires. In spatial data analysis, locating the hotspot is essential. However, it is challenging to identify a spatially significant hotspot. This paper proposes a novel method for finding statistically significant hotspots based on the rough graph. First, the Global Moan index is used to determine the presence of spatial dependence in the data set. Then, the HSDRG algorithm is implemented to find the hotspot of the polygon vector data. Two spatial neighbour search techniques, BFS and DFS, are employed to find the spatial neighbour. The algorithm is evaluated using socio-economic data from Uttar Pradesh, India. Four variables were chosen to find the hotspot: female literacy, male literacy, female workers, and male workers. A percentage value is calculated for each variable to find the hotspot. The analysis reveals that the generated hotspots are denser, the PAI value is high, and the running time is less than the other methods found in the literature. The running time of the HSDRH algorithm using DFS as the search technique is 69.48%, 72.91%, and 73.08% less compared to the methods Moran’s I, Getis Ord Gi, and Getis Ord Gi*, respectively. Therefore, the HDSRG algorithm using a rough graph is considered the optimal method for hotspot detection. This type of analysis is vital to know whether the area has good literacy concerning males and females and to know the area has hotspot workers.