SYSTEM AND METHOD FOR DATA CLUSTERING USING GRID-BASED SHAKING

摘要

PURPOSE: A method for clustering data using grid-based shaking and a system using the same are provided to draw a correct cluster result by preventing outlier data from being merged with other clusters. CONSTITUTION: A plurality of data points becomes partially sampling in a grid type in a shaking sampling step(S110). One or more partial samples are clustered in a partial sample clustering step(S120). A final cluster is formed by merging the cluster in a cluster merging step(S130). The shaking sampling stage comprises the following steps: a basis grid sampling step of sampling a plurality of data points into the grip type; a movement sampling step of partially sampling one or more grid; and a partial sample arrangement step of listing the partial samples.