Enhancement of efficiency by thrifty search of interlocking neighbor grids approach for grid-based data clustering

This investigation presents a new grid-based data clustering algorithm. Firstly, a parameter setting step sets a grid parameter and a threshold parameter. A diving step segments a space with a plurality of data points according to the grid parameter. A categorizing step determines whether a number of the data points contained in each grid is larger than or equal to a value of the threshold parameter. Moreover, the grid is categorized as a valid grid if the number of the data points contained therein is larger than or equal to the value of the threshold parameter, and the grid is categorized as an invalid grid if the number of the data points contained therein is smaller than the value of the threshold parameter. Finally, the clustering step retrieves one of the valid grids. If the retrieved valid grid is not yet clustered, the clustering step conducts horizontal and vertical searching/merging operations on the valid grid.