A Study on Grid Coupling-based Big Data Clustering Method for Massive Electricity Carbon Emission with Mixed Attributes

To comprehensively and stably categorize and analyze the power carbon emission data for different sizes and types of power carbon emission datasets, a clustering method based on grid coupling is studied for massive power carbon emission big data with mixed attributes. In the offline stage, based on the updated core grid and the center of mass of the grid, we search for grids that are connected with the density, complete the clustering, and obtain the clustering results of the massive electricity carbon emission mixed-attribute big data clustering. It is proved that the method can effectively divide the spatial grid of massive electricity carbon emission mixed-attribute big data and calculate the weight of each data sample, and can quickly cluster massive electricity carbon emission mixed-attribute big data. Under different degrees of interference information, the clustering of the method has a higher normalized mutual information (NMI) index and Riese’s (RI) index, that is, the clustering accuracy is higher.