模型模拟的基准分区格网数据和四种空间网络分区算法结果图(2022)

本数据包含由计算机程序模拟的八个基准分区格网数据（shapefile格式），以及基于此数据计算得到的四种网络分区算法结果图（jpg格式）和结果文档（csv格式）。其中基准分区格网数据由空间自回归模型构造(自回归系数γ=0.9)，网格为10*10单位格网。基准网络数据由格网数据和重力模型产生(距离衰减系数β=0.5,1,1.5,2)。四种网络分区算法分别是Louvain，louvgeo, REDCAP, SCSCD。算法结果图包含了模拟数据空间分布图、四种算法的分区结果图、算法参数调整结果图等。为测试算法的准确性，每个基准分区格网数据模拟了100遍。This dataset comprises eight benchmark partition grid datasets (Shapefile format) simulated via computer programs, alongside four sets of network partition algorithm result plots (JPEG format) and result documentation (CSV format) derived from these datasets. Specifically, the benchmark partition grid datasets are constructed using a spatial autoregressive model with an autoregressive coefficient γ=0.9, where each grid is a 10×10 unit grid. The benchmark network datasets are generated using the grid data and a gravity model, with distance decay coefficients set to 0.5, 1, 1.5, and 2 respectively. The four network partition algorithms evaluated are Louvain, louvgeo, REDCAP, and SCSCD. The algorithm result plots cover spatial distribution maps of the simulated data, partition outcome plots for each of the four algorithms, parameter adjustment performance plots, and other related visualizations. To validate the accuracy of these algorithms, each benchmark partition grid dataset was simulated 100 times.