Spatial Data Analysis using DBSCAN Method and KNN classification

Abstract

Spatial Data Clustering is one of the most important technical techniques used to obtain information about knowledge about number boundaries in databases from various applications. This technique can determine groups of forms that cannot be arranged and can be used effectively with a budget. Exploring interesting and useful spatial boundary patterns is more difficult to extract traditional and categorical numerical polymers because of the difficulty of species, the relationship between autocorrelation of spatial boundaries. One of the pioneering techniques in the development of facial and technical grouping technologies is DBSCAN. This technique can determine groups of shapes that cannot be arranged and can be arranged in an effective way. the groups that have already received the next classification process are carried out in order to obtain information on the classes already formed. The K-Nearest Neighbour classification technique is based on learning by analogy. When there is new data, K-Nearest Neighbor will look for a class of data from the learning sample that is closest to the new data. This closeness can be defined using the Euclidean Distance calculation method.