The clustering of black spot using province public data

Abstract

Software is an essential element in ensuring competitiveness throughout the country. Advanced foreign and domestic software companies are leading innovation not only in the traditional IT field but also in various fields based on sophisticated software. The development of artificial intelligence machine learning opens up possibilities for the automation of intellectual activities, and its impact is expected to be very large and broad. The high accuracy of artificial intelligence and machine learning requires a large amount of data. In Korea as well as overseas, public data can be accessed and utilized through public data portals. This study clustered the public data of Gyeonggi provincial accidents in order to identify the area of accidents. In this paper, we propose a solution to the problem by specifying the range (Epsilon) and the minimum population (MinPts) in order to characterize the multi-accident area by using DBSCAN (Density Based Spatial Clustering of Applications with Noise). © Research India Publications.