\u901a\u8fc7\u672c\u6587\uff0c\u4f60\u53ef\u4ee5\u4e86\u89e3\u5230\uff1a<\/strong><\/p>\n\n\n\n
1\u3001\u805a\u7c7b\uff08clustering\uff09<\/strong><\/p>\n\n\n\n
\u805a\u7c7b\uff0c\u5c31\u662f\u6309\u7167\u67d0\u4e2a\u7279\u5b9a\u6807\u51c6\uff08\u6bd4\u5982ā\u8ddd\u79bb\u51c6\u5219ā\uff09\uff0c\u5c06\u4e00\u4e2a\u6570\u636e\u96c6\u5212\u5206\u4e3a\u6709\u610f\u4e49\u4e0d\u540c\u7684\u7c7b\u578b\u6216\u7ec4\uff08\u7c07\uff09\uff0c\u4f7f\u5f97\u76f8\u4f3c\u6027\u5c3d\u53ef\u80fd\u5927\u7684\u5ba2\u6237\u88ab\u5212\u5206\u5230\u540c\u4e00\u7fa4\uff0c\u540c\u65f6\u5728\u4e0d\u540c\u7fa4\u95f4\u80fd\u8868\u73b0\u51fa\u660e\u663e\u7684\u5dee\u5f02\u6027\u3002<\/p>\n\n\n\n
\u7b80\u8a00\u4e4b\uff0c\u5c31\u662f\u805a\u7c7b\u540e\uff0c\u540c\u4e00\u7c7b\u6570\u636e\u5c3d\u53ef\u80fd\u805a\u5728\u4e00\u8d77\uff0c\u4e0d\u540c\u7c7b\u7684\u6570\u636e\u5c3d\u91cf\u5206\u79bb\u3002<\/strong><\/p>\n\n\n\n
\u5bf9\u4e8e\u4e0d\u540c\u5ba2\u6237\u7279\u6027\u7684\u76f8\u4f3c\u6027\uff0c\u4f1a\u4f9d\u636e\u89c2\u6d4b\u5ba2\u6237\u95f4\u7684\u8ddd\u79bb\u8fdb\u884c\u5ea6\u91cf\uff0c\u6bd4\u5982\u6570\u5b66\u610f\u4e49\u4e0a\u7684\u6b27\u6c0f\u8ddd\u79bb\uff08euclidean distance\uff09\u548c\u57fa\u4e8e\u76f8\u5173\u6027\u7684\u8ddd\u79bb\uff08correlation-based distance\uff09\u3002<\/p>\n\n\n\n
\u5bf9\u4e8e\u805a\u7c7b\u548c\u5206\u7c7b\uff0c\u4e24\u8005\u4e4b\u95f4\u662f\u6709\u4e00\u5b9a\u7684\u5dee\u522b\u7684\uff0c\u63a2\u8ba8\u8fd9\u4e2a\u95ee\u9898\u9700\u8981\u5148\u5f15\u5165\u4e24\u4e2a\u6982\u5ff5\uff0c\u76d1\u7763\u5b66\u4e60(supervised learning)\u548c\u65e0\u76d1\u7763\u5b66\u4e60\uff08unsupervised learning\uff09<\/strong>\u3002<\/p>\n\n\n\n
\u5206\u7c7b\u4f1a\u5177\u5907\u660e\u786e\u7684\u89c4\u5219\u548c\u6761\u4ef6\uff0c\u50cf\u56fe\u4e66\u9986\u7684\u85cf\u4e66\u5206\u7c7b\uff0c\u6309\u4e3b\u9898\uff0c\u6309\u5e74\u4ee3\u3001\u5730\u57df\u3001\u8bed\u8a00\u7b49\u7b49\u3002\u4ee5\u8ba1\u7b97\u673a\u7684\u601d\u7ef4\u8fdb\u884c\u7406\u89e3\uff0c\u5373\u8ba1\u7b97\u673a\u53ef\u4ee5\u4ece\u5df2\u77e5\u7684\u8bad\u7ec3\u6570\u636e\u96c6\u4e2d\u8fdb\u884cā\u5b66\u4e60ā\uff0c\u4ece\u800c\u83b7\u53d6\u5bf9\u4e8e\u5206\u7c7b\u903b\u8f91\u7684\u5224\u522b\u65b9\u6cd5\u3002\u5f53\u6c47\u5165\u672a\u77e5\u7c7b\u522b\u7684\u65b0\u6570\u636e\u8fdb\u884c\u5206\u7c7b\u65f6\uff0c\u53ef\u4ee5\u4f9d\u7167\u8bad\u7ec3\u6240\u5f97\u7684\u7ecf\u9a8c\u8fdb\u884c\u81ea\u52a8\u5224\u65ad\uff0c\u800c\u8fd9\u79cd\u63d0\u4f9b\u8bad\u7ec3\u6570\u636e\u7684\u8fc7\u7a0b\u901a\u5e38\u53eb\u505a\u76d1\u7763\u5b66\u4e60\uff08supervised learning\uff09\u3002<\/p>\n\n\n\n
\u800c\u805a\u7c7b\u6ca1\u6709\u786e\u5207\u7684\u5b9a\u4e49\uff0c\u5e76\u4e0d\u4f1a\u77e5\u9053\u4efb\u4f55\u6837\u672c\u7684\u7c7b\u522b\u6807\u53f7\u3002\u5e0c\u671b\u901a\u8fc7\u67d0\u79cd\u7b97\u6cd5\u6765\u628a\u4e00\u7ec4\u672a\u77e5\u7c7b\u522b\u7684\u6837\u672c\u5212\u5206\u6210\u82e5\u5e72\u7c7b\u522b\uff0c\u671f\u95f4\u4e0d\u9700\u8981\u4f7f\u7528\u8bad\u7ec3\u6570\u636e\u96c6\u8fdb\u884c\u5b66\u4e60\u3002\u6240\u4ee5\uff0c\u805a\u7c7b\u53c8\u88ab\u79f0\u4e3a\u65e0\u76d1\u7763\u5b66\u4e60\uff08unsupervised learning\uff09\u3002<\/strong><\/p>\n\n\n\n
\u4e24\u8005\u76f8\u6bd4\uff0c\u805a\u7c7b\u65e8\u5728\u9a8c\u8bc1\u6570\u636e\u4e4b\u95f4\u7684\u76f8\u4f3c\u6027\u6216\u4e0d\u76f8\u4f3c\u6027\uff0c\u66f4\u4fa7\u91cd\u4e8e\u8fb9\u754c\u6761\u4ef6\u3002<\/strong><\/p>\n\n\n\n
2\u3001\u805a\u7c7b\u57fa\u672c\u601d\u60f3<\/strong><\/p>\n\n\n\n
\u5148\u770b\u4e00\u4e2a\u6848\u4f8b\u3002\u5728\u67d0\u5e74\u7684\u7f8e\u56fd\u603b\u7edf\u5927\u9009\u4e2d\uff0c\u5019\u9009\u4eba\u7684\u5f97\u7968\u6570\u975e\u5e38\u63a5\u8fd1\u3002\u76f8\u4e92\u7ade\u4e89\u7684\u5019\u9009\u4eba\u7684\u666e\u9009\u7968\u6570 48.7% : 47.9%\u3002\u8fd9\u65f6\u5019\uff0c\u5982\u679c\u6709\u529e\u6cd5\u8ba9 1% \u7684\u9009\u6c11\u5012\u6208\uff0c\u5c06\u624b\u4e2d\u7684\u9009\u7968\u6295\u5411\u53e6\u5916\u7684\u5019\u9009\u4eba\uff0c\u90a3\u4e48\u9009\u4e3e\u7ed3\u679c\u5c31\u4f1a\u622a\u7136\u4e0d\u540c\u3002<\/p>\n\n\n\n
\u5b9e\u9645\u4e0a\uff0c\u5982\u679c\u6709\u9488\u5bf9\u6027\u5730\u59a5\u5584\u52a0\u4ee5\u5f15\u5bfc\uff0c\u5c11\u90e8\u5206\u9009\u6c11\u5c31\u4f1a\u8f6c\u6362\u7acb\u573a\u3002\u5c3d\u7ba1\u8fd9\u7c7b\u9009\u4e3e\u8005\u5360\u7684\u6bd4\u4f8b\u8f83\u4f4e\uff0c\u4f46\u5f53\u5019\u9009\u4eba\u7684\u9009\u7968\u63a5\u8fd1\u65f6\uff0c\u8fd9\u4e9b\u4eba\u7684\u7acb\u573a\u65e0\u7591\u4f1a\u5bf9\u9009\u4e3e\u7ed3\u679c\u4ea7\u751f\u975e\u5e38\u5927\u7684\u5f71\u54cd\u3002<\/strong><\/p>\n\n\n\n
\u9996\u5148\u6536\u96c6\u9009\u6c11\u7684\u57fa\u672c\u4fe1\u606f\uff0c\u542c\u53d6\u9009\u6c11\u58f0\u97f3\u4e2d\u53cd\u9988\u6ee1\u610f\u6216\u4e0d\u6ee1\u610f\u7684\u4fe1\u606f\uff0c\u56e0\u4e3a\u9009\u6c11\u5173\u6ce8\u7684\u91cd\u8981\u8bae\u9898\uff0c\u5f88\u5927\u53ef\u80fd\u5de6\u53f3\u9009\u6c11\u7684\u6295\u7968\u7ed3\u679c\u3002<\/p>\n\n\n\n
\u7136\u540e\uff0c\u5c06\u8fd9\u4e9b\u4fe1\u606f\u8f93\u5165\u5230\u67d0\u4e2a\u805a\u7c7b\u7b97\u6cd5\u4e2d\u3002\u5bf9\u805a\u7c7b\u7ed3\u679c\u4e2d\u7684\u6bcf\u4e00\u4e2a\u7fa4\uff08\u6700\u597d\u9009\u62e9\u6700\u5927\u7fa4 \uff09\uff0c \u7cbe\u5fc3\u6784\u9020\u80fd\u591f\u5438\u5f15\u8be5\u7fa4\u9009\u6c11\u7684\u6d88\u606f\u3002<\/p>\n\n\n\n
\u6700\u540e\uff0c\u5f00\u5c55\u7ade\u9009\u6d3b\u52a8\u5e76\u89c2\u5bdf\u4e0a\u8ff0\u505a\u6cd5\u662f\u5426\u6709\u6548\uff0c\u4e0d\u65ad\u8fed\u4ee3\u8c03\u6574\uff0c\u8fd9\u5c31\u662f\u805a\u7c7b\u7684\u5927\u81f4\u539f\u7406\u3002<\/strong><\/p>\n\n\n\n
\u5bf9\u4e8e\u4e00\u5806\u6563\u843d\u7684\u70b9\uff0c\u5148\u786e\u5b9a\u8fd9\u4e9b\u6563\u843d\u7684\u70b9\u6700\u540e\u9700\u8981\u805a\u6210\u51e0\u7c7b\uff0c\u7136\u540e\u6311\u9009\u968f\u673a\u6311\u9009\u51e0\u4e2a\u70b9\u4f5c\u4e3a\u521d\u59cb\u4e2d\u5fc3\u70b9\uff0c\u518d\u7136\u540e\u4f9d\u636e\u9884\u5148\u5b9a\u597d\u7684\u542f\u53d1\u5f0f\u7b97\u6cd5\uff08heuristic algorithms\uff09\u7ed9\u6570\u636e\u70b9\u505a\u8fed\u4ee3\u91cd\u7f6e\uff08iterative relocation\uff09\uff0c\u76f4\u5230\u6700\u540e\u5230\u8fbe\u201c\u7c7b\u5185\u7684\u70b9\u90fd\u8db3\u591f\u8fd1\uff0c\u7c7b\u95f4\u7684\u70b9\u90fd\u8db3\u591f\u8fdc\u201d\u7684\u76ee\u6807\u6548\u679c\u3002<\/p>\n\n\n\n
\u542f\u53d1\u5f0f\u7b97\u6cd5\u4e3b\u8981\u5305\u62ec\u4e86K-Means\uff0c\u53ca\u5176\u53d8\u4f53K-Medoids\u3001K-Modes\u3001K-Medians\u3001Kernel K-means\u7b49\u7b97\u6cd5\u3002<\/p>\n\n\n\n
\u672c\u6b21\u5c06\u4f1a\u9009\u62e9\u6700\u4e3a\u7ecf\u5178\u7684 K-Means \u805a\u7c7b+ RFM \u6a21\u578b\uff0c\u5176\u4ed6\u7b97\u6cd5\u4f1a\u5728\u540e\u7eed\u6587\u7ae0\u4e2d\u66f4\u65b0\u3002<\/p>\n\n\n\n
3\u3001K-Means<\/strong><\/p>\n\n\n\n
K-Means \u662f\u4e00\u79cd\u8fed\u4ee3\u6c42\u89e3\u7684\u805a\u7c7b\u5206\u6790\u7b97\u6cd5\uff0c\u7531\u4e8e\u5b83\u53ef\u4ee5\u53d1\u73b0 K \u4e2a\u4e0d\u540c\u7684\u7fa4, \u4e14\u6bcf\u4e2a\u7fa4\u7684\u4e2d\u5fc3\u91c7\u7528\u7fa4\u4e2d\u6240\u542b\u503c\u7684\u5747\u503c\u8ba1\u7b97\u800c\u6210\uff0c\u4e5f\u79f0\u4e4b\u4e3a K-\u5747\u503c\u3002\u5176\u4e2d\uff0c\u7fa4\u6570 K \u987b\u7531\u7528\u6237\u6307\u5b9a\u3002<\/p>\n\n\n\n
K-Means \u805a\u7c7b\u8fc7\u7a0b\u56fe\u89e3<\/p>\n\n\n\n
\u7b97\u6cd5\u6d41\u7a0b\u5982\u4e0b\uff1a<\/strong><\/p>\n\n\n\n
4\u3001K-Means \u4f18\u7f3a\u70b9<\/strong><\/p>\n\n\n\n
K-Means \u4f18\u70b9\u5728\u4e8e\u539f\u7406\u7b80\u5355\uff0c\u5bb9\u6613\u5b9e\u73b0\uff0c\u805a\u7c7b\u6548\u679c\u597d\uff0c\u5bf9\u4e8e\u5927\u578b\u6570\u636e\u96c6\u4e5f\u662f\u7b80\u5355\u9ad8\u6548\u3001\u65f6\u95f4\u590d\u6742\u5ea6\u3001\u7a7a\u95f4\u590d\u6742\u5ea6\u4f4e\u3002<\/strong><\/p>\n\n\n\n
\u5f53\u7136\uff0c\u4e5f\u6709\u4e00\u4e9b\u7f3a\u70b9\u3002\u9700\u8981\u9884\u5148\u8bbe\u5b9a K \u503c\uff0c\u5bf9\u6700\u5148\u7684 K \u4e2a\u70b9\u9009\u53d6\u5f88\u654f\u611f\uff1b\u5bf9\u566a\u58f0\u548c\u79bb\u7fa4\u503c\u975e\u5e38\u654f\u611f\uff1b\u53ea\u7528\u4e8e numerical \u7c7b\u578b\u6570\u636e\uff1b\u4e0d\u80fd\u89e3\u51b3\u975e\u51f8\uff08non-convex\uff09\u6570\u636e\u3002\u53d7\u79bb\u7fa4\u503c\u5f71\u54cd\u5927\uff0c\u6700\u91cd\u8981\u662f\u6570\u636e\u96c6\u5927\u65f6\u7ed3\u679c\u5bb9\u6613\u5c40\u90e8\u6700\u4f18\u3002<\/p>\n\n\n\n
\u6bcf\u79cd\u7c7b\u578b\u7684\u7b97\u6cd5\u90fd\u4f1a\u6709\u5b83\u7684\u7279\u70b9\uff0c\u90fd\u6709\u4fa7\u91cd\u89e3\u51b3\u95ee\u9898\u7684\u9762\u5411\u3002\u5728\u5e94\u7528\u4e4b\u524d\uff0c\u5148\u7406\u89e3\u4e0d\u540c\u7b97\u6cd5\u7684\u5185\u5728\u903b\u8f91\u3001\u4f5c\u7528\u3001\u5e94\u7528\u573a\u666f\uff0c\u7ed3\u5408\u5b9e\u8df5\u7ecf\u9a8c\uff0c\u9009\u51fa\u6700\u5408\u9002\u7684\u7b97\u6cd5\u6a21\u578b\u6765\u8fbe\u5230\u4e1a\u52a1\u76ee\u6807\u3002<\/p>\n\n\n\n
5\u3001RFM + K-Means \u6848\u4f8b\u5e94\u7528<\/strong><\/p>\n\n\n\n
\u7ec3\u4e60\u6570\u636e\u96c6\u548c\u5b9e\u73b0\u4ee3\u7801\u53ef\u4ee5\u901a\u8fc7\u516c\u4f17\u53f7\u540e\u53f0\u56de\u590d\u3010RFM<\/strong>\u3011\u83b7\u53d6\u3002<\/p>\n\n\n\n
\u7b14\u8005\u4f7f\u7528\u7684\u662f R Studio \u8fdb\u884c\u6848\u4f8b\u6f14\u793a\u3002\u5982\u679c\u4f60\u6709\u5176\u4ed6\u7684\u6e90\u6570\u636e\uff0c\u8981\u60f3\u5728 R Studio \u4e2d\u8fdb\u884c\u805a\u7c7b\u5206\u6790\uff0c\u5e94\u8be5\u6309\u7167\u5982\u4e0b\u8981\u6c42\u51c6\u5907\u6570\u636e\uff1a<\/p>\n\n\n\n
\u6b65\u9aa4 1\uff1a<\/strong><\/p>\n\n\n\n
install.packages(\"factoextra\")\ninstall.packages(\"cluster\")<\/code><\/pre>\n\n\n\n
\u6570\u636e\u9884\u5904\u7406\u5b8c\u540e\uff0c\u5982\u679c\u6ca1\u6709\u5b89\u88c5 fact<\/a><\/span>oextra\uff08\u7528\u4e8e\u5bf9\u805a\u7c7b\u7ed3\u679c\u8fdb\u884c\u53ef\u89c6\u5316\uff09 \u3001cluster\uff08\u7528\u4ee5\u5bf9\u6570\u636e\u8fdb\u884c\u805a\u7c7b\u8ba1\u7b97\uff09\uff0c\u9700\u8981\u5148\u8fdb\u884c\u5b89\u88c5\u3002<\/p>\n\n\n\n
\u6b65\u9aa4 2\uff1a<\/strong><\/p>\n\n\n\n
#\u8f7d\u5165\u5305\nlibrary(factoextra)\nlibrary(cluster)\n\n\n#\u5bfc\u5165\u5df2\u7ecf\u5904\u7406\u597d\u7684\u6570\u636e\u96c6 \u2192 RFM_Dataset.csv\nrfm_data <- read.csv(\"\/Users\/guofu.long\/Desktop\/RFM_Dataset.csv\", header = TRUE)\n\n\ndata_1 <- rfm_data[,1:4]\nhead(data_1)\n\n\n#\u67e5\u770b\u6570\u636e\u884c\u5217\u6570\u3001\u5b57\u7b26\u7c7b\u578b\u3001\u63cf\u8ff0\u6027\u7edf\u8ba1\u91cf\ndim(data_1)\nstr(data_1)\nsummary(data_1)<\/code><\/pre>\n\n\n\n
\u8fd9\u91cc\u4f7f\u7528\u7684 RFM_Dataset.csv \u6570\u636e\u662f\u5728\u300a\u5e94\u7528 RFM \u6a21\u578b\u5ba2\u6237\u5206\u7fa4\u64cd\u4f5c\u7bc7\uff0c\u63d0\u6548\u5ba2\u6237\u4f53\u9a8c\u7ba1\u7406\u300b\u4e2d\u5904\u7406\u597d RFM \u5bf9\u5e94\u503c\u3002\u5bfc\u5165\u6570\u636e\u540e\uff0c\u5bf9\u6570\u636e\u8fdb\u884c\u9002\u5f53\u7684\u89c2\u5bdf\uff0c\u6bd4\u5982\u6570\u636e\u884c\u5217\u6570\u3001\u5b57\u7b26\u7c7b\u578b\u3001\u63cf\u8ff0\u6027\u7edf\u8ba1\u91cf\u7b49\u3002<\/p>\n\n\n\n
\u6b65\u9aa4 3\uff1a<\/strong><\/p>\n\n\n\n
data_2 <- data_1[,2:4]\n#\u6570\u636e\u6807\u51c6\u5316\ndata_3 <- scale(data_2)\nhead(data_3)<\/code><\/pre>\n\n\n\n
\u7531\u4e8e\u540c\u4e00\u4e2a\u6570\u636e\u96c6\u5408\u4e2d\u7ecf\u5e38\u5305\u542b\u4e0d\u540c\u7c7b\u522b\u7684\u53d8\u91cf\u3002\u8fd9\u6837\u4f1a\u5bfc\u81f4\u8fd9\u4e9b\u53d8\u91cf\u7684\u503c\u57df\u53ef\u80fd\u5927\u4e0d\u76f8\u540c\uff0c\u5982\u679c\u4f7f\u7528\u539f\u503c\u57df\u5c06\u4f1a\u4f7f\u5f97\u503c\u57df\u5927\u7684\u53d8\u91cf\u88ab\u8d4b\u4e88\u66f4\u591a\u7684\u6743\u91cd\u3002<\/p>\n\n\n\n
\u9488\u5bf9\u8fd9\u4e2a\u95ee\u9898\uff0c\u6807\u51c6\u5316\u53ef\u4ee5\u4f7f\u5f97\u4e0d\u540c\u7684\u7279\u5f81\u5177\u6709\u76f8\u540c\u7684\u5c3a\u5ea6\uff0c\u6d88\u9664\u7279\u5f81\u4e4b\u95f4\u7684\u5dee\u5f02\u6027\u3002\u5f53\u539f\u59cb\u6570\u636e\u4e0d\u540c\u7ef4\u5ea6\u4e0a\u7684\u7279\u5f81\u7684\u5c3a\u5ea6\uff08\u5355\u4f4d\uff09\u4e0d\u4e00\u81f4\u65f6\uff0c\u9700\u8981\u6807\u51c6\u5316\u6b65\u9aa4\u5bf9\u6570\u636e\u8fdb\u884c\u9884\u5904\u7406\u3002<\/p>\n\n\n\n
RFM_Dataset.csv \u6570\u636e\u96c6\u7b14\u8005\u5176\u5b9e\u5df2\u7ecf\u5c06\u5c3a\u5ea6\u8f6c\u5316\u7edf\u4e00\u8ba1\u5206\u65b9\u5f0f\uff0c\u4e3a\u4e86\u51f8\u663e\u8fd9\u4e2a\u6b65\u9aa4\u7684\u91cd\u8981\u6027\uff0c\u7279\u522b\u518d\u52a0\u4ee5\u8bf4\u660e\u3002<\/p>\n\n\n\n
R \u8bed\u8a00\u4e2d scale \u51fd\u6570\u63d0\u4f9b\u6570\u636e\u6807\u51c6\u5316\u529f\u80fd\uff0c\u6307\u4e2d\u5fc3\u5316\u4e4b\u540e\u7684\u6570\u636e\u5728\u9664\u4ee5\u6570\u636e\u96c6\u7684\u6807\u51c6\u5dee\uff0c\u5373\u6570\u636e\u96c6\u4e2d\u7684\u5404\u9879\u6570\u636e\u51cf\u53bb\u6570\u636e\u96c6\u7684\u5747\u503c\u518d\u9664\u4ee5\u6570\u636e\u96c6\u7684\u6807\u51c6\u5dee\u3002<\/p>\n\n\n\n
\u7279\u522b\u9700\u8981\u6ce8\u610f scale \u51fd\u6570\u4e0d\u63a5\u53d7\u542b\u6709\u5b57\u7b26\u4e32\u7684\u6570\u636e\u6846\uff0c\u4f7f\u7528\u524d\u8981\u8fdb\u884c\u8f6c\u6362\u3002<\/p>\n\n\n\n
\u6b65\u9aa4 4\uff1a<\/strong><\/p>\n\n\n\n
#\u8bbe\u7f6e\u968f\u673a\u6570\u79cd\u5b50\uff0c\u4fdd\u8bc1\u53ef\u91cd\u590d\nset.seed(1234)\n#\u624b\u8098\u6cd5\uff0c\u786e\u5b9a\u6700\u4f73\u805a\u7c7b\u6570\u76ee\nfviz_nbclust(data_3, kmeans, method = \"wss\") + geom_vline(xintercept = 3, linetype = 2)<\/code><\/pre>\n\n\n\n
\u4e3a\u4fdd\u8bc1\u5b9e\u9a8c\u53ef\u91cd\u590d\u8fdb\u884c\uff0c\u9700\u8bbe\u5b9a\u968f\u673a\u6570\u79cd\u5b50\u3002<\/p>\n\n\n\n
factoextra \u5305\u4e2d\u5305\u542b\u8bb8\u591a\u7528\u4e8e\u805a\u7c7b\u5206\u6790\u548c\u53ef\u89c6\u5316\u7684\u51fd\u6570\uff0c\u5305\u62ec\uff1a<\/p>\n\n\n\n
\u51fd\u6570<\/strong><\/td> \u529f\u80fd<\/td><\/tr> dist(fviz_dist, get_dist)<\/td> \u8ddd\u79bb\u77e9\u9635\u7684\u8ba1\u7b97\u4e0e\u53ef\u89c6\u5316<\/td><\/tr> get_clust_tendency<\/td> \u8bc4\u4f30\u805a\u7c7b\u8d8b\u52bf<\/td><\/tr> fviz_nbclust(fviz_gap_stat)<\/td> \u786e\u5b9a\u6700\u4f73\u7684\u805a\u7c7b\u6570<\/td><\/tr> fviz_dend<\/td> \u6811\u72b6\u56fe\u7684\u589e\u5f3a\u7248\u53ef\u89c6\u5316<\/td><\/tr> fviz_cluster<\/td> \u805a\u7c7b\u7ed3\u679c\u7684\u53ef\u89c6\u5316<\/td><\/tr> fviz_mclust<\/td> \u57fa\u4e8e\u6a21\u578b\u7684\u805a\u7c7b\u7ed3\u679c\u7684\u53ef\u89c6\u5316<\/td><\/tr> fviz_silhouette<\/td> \u805a\u7c7b\u4e2d\u7684\u8f6e\u5ed3\u4fe1\u606f\u7684\u53ef\u89c6\u5316<\/td><\/tr> hcut<\/td> \u5206\u5c42\u805a\u7c7b\u7684\u8ba1\u7b97\u5e76\u526a\u5207\u6811<\/td><\/tr> hkmeans<\/td> \u5206\u5c42\u7684k\u5747\u503c\u805a\u7c7b<\/td><\/tr> eclust<\/td> \u805a\u7c7b\u5206\u6790\u7684\u53ef\u89c6\u5316\u589e\u5f3a\u7248<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n \u5229\u7528\u201c\u624b\u8098\u6cd5\u201d\uff0c\u627e\u5230\u6700\u4f73\u805a\u7c7b\u6570\u76ee\u3002\u968f\u7740\u805a\u7c7b\u6570 K \u7684\u589e\u5927\uff0c\u6837\u672c\u5212\u5206\u4f1a\u66f4\u52a0\u7cbe\u7ec6\uff0c\u6bcf\u4e2a\u7fa4\u7684\u805a\u5408\u7a0b\u5ea6\u4f1a\u9010\u6e10\u63d0\u9ad8\uff0c\u90a3\u4e48\u8bef\u5dee\u5e73\u65b9\u548c SSE \u81ea\u7136\u4f1a\u9010\u6e10\u53d8\u5c0f\u3002<\/strong><\/p>\n\n\n\n
\u5373\u968f\u7740\u805a\u7c7b\u6570\u76ee\u589e\u591a\uff0c\u6bcf\u4e00\u4e2a\u7c7b\u522b\u4e2d\u6570\u91cf\u8d8a\u6765\u8d8a\u5c11\uff0c\u8ddd\u79bb\u8d8a\u6765\u8d8a\u8fd1\uff0c\u56e0\u6b64 WSS \u503c\u80af\u5b9a\u662f\u968f\u7740\u805a\u7c7b\u6570\u76ee\u589e\u591a\u800c\u51cf\u5c11\u7684\uff0c\u6240\u4ee5\u5173\u6ce8\u7684\u662f\u659c\u7387\u7684\u53d8\u5316<\/strong>\u3002<\/p>\n\n\n\n
<\/figure>\n\n\n\n
\u805a\u7c7b\u6570\u8d8b\u52bf\u56fe<\/p>\n\n\n\n
\u5728 WWS \u968f\u7740 K \u503c\u7684\u7ee7\u7eed\u589e\u5927\u800c\u51cf\u5c11\u5f97\u5f88\u7f13\u6162\u65f6\uff0c\u8ba4\u4e3a\u8fdb\u4e00\u6b65\u589e\u5927\u805a\u7c7b\u6570\u6548\u679c\u4e5f\u5e76\u4e0d\u80fd\u589e\u5f3a\uff0c\u4e5f\u5c31\u662f\u8bf4 SSE \u548c K \u7684\u5173\u7cfb\u56fe\u662f\u4e00\u4e2a\u624b\u8098\u7684\u5f62\u72b6\uff0c\u5b58\u5728\u7684\u8fd9\u4e2a\u201c\u8098\u70b9\u201d\u5c31\u662f\u6700\u4f73\u805a\u7c7b\u6570\u76ee<\/strong>\u3002\u4ece 1 \u7c7b\u5230 3 \u7c7b\u4e0b\u964d\u5f97\u5f88\u5feb\uff0c\u4e4b\u540e\u4e0b\u964d\u5f97\u5f88\u6162\uff0c\u6240\u4ee5\u6700\u4f73\u805a\u7c7b\u4e2a\u6570\u9009\u4e3a 3\u3002<\/strong><\/p>\n\n\n\n
\u6b65\u9aa4 5\uff1a<\/strong><\/p>\n\n\n\n
#\u8fdb\u884c\u805a\u7c7b\nresult <- kmeans(data_3,3)<\/code><\/pre>\n\n\n\n
kmeans \u51fd\u6570\u8fd8\u63d0\u4f9b\u4e86\u50cf iter.max\uff08\u6700\u5927\u8fed\u4ee3\u6b21\u6570\uff09\u3001 nstart\uff08\u8d77\u59cb\u968f\u673a\u5206\u533a\u7684\u6570\u91cf\uff09\u7b49\u7b49\uff0c\u6709\u9700\u8981\u53ef\u4ee5\u6839\u636e\u51fd\u6570\u7528\u6cd5\u81ea\u884c\u8c03\u6574\uff0c\u8fd9\u91cc\u4f7f\u7528\u7684\u662f\u9ed8\u8ba4\u7684\u53c2\u6570\u8bbe\u5b9a\u3002<\/p>\n\n\n\n
\u6b65\u9aa4 6\uff1a<\/strong><\/p>\n\n\n\n
#\u53ef\u89c6\u5316\u805a\u7c7b\nfviz_cluster(result, data = data_3)<\/code><\/pre>\n\n\n\n
\u4f7f\u7528 factoextra \u5305\u751f\u6210\u7684\u805a\u7c7b\u540e\u7684\u5206\u5e03\u56fe\u3002<\/p>\n\n\n\n
<\/figure>\n\n\n\n
\u805a\u7c7b\u5212\u5206\u56fe<\/p>\n\n\n\n
\u6b65\u9aa4 7\uff1a<\/strong><\/p>\n\n\n\n
#\u805a\u7c7b\u7ed3\u679c\u5bfc\u51fa\nresult_output <- data.frame(data_1[,1:4],result$cluster)\nwrite.csv(result_output,file=\"\/Users\/guofu.long\/Desktop\/result_output.csv\",row.names=T,quote=F)<\/code><\/pre>\n\n\n\n
\u805a\u7c7b\u7ed3\u679c\u5217\u8868\u3002<\/p>\n\n\n\n
<\/figure>\n\n\n\n
\u5ba2\u6237\u5206\u7fa4\u5339\u914d\u5217\u8868<\/p>\n\n\n\n
6\u3001\u5c0f\u7ed3<\/strong><\/p>\n\n\n\n
\u6700\u540e\uff0c\u5229\u7528 K-Means \u805a\u7c7b\u7b97\u6cd5\u548c RFM \u6a21\u578b\u5f97\u5230\u5ba2\u6237\u5206\u7fa4\u53ea\u662f\u5f00\u59cb\uff0c\u7531\u4e8e\u805a\u7c7b\uff08\u5373\u6570\u5b66\u4e0a\u7684\u76f8\u4f3c\u6027\uff09\u6240\u4ea7\u751f\u7684\u5ba2\u6237\u5206\u7fa4\uff0c\u5206\u7fa4\u672c\u8eab\u4e0d\u80fd\u76f4\u63a5\u4ea7\u751f\u4ef7\u503c\uff0c\u65e0\u8bba\u5206\u7fa4\u7528\u7684\u662f\u5565\u6a21\u578b\uff0c\u6700\u540e\u7684\u7ed3\u679c\u4e5f\u53ea\u662f\u4e00\u4e2a\u6570\u636e\u6807\u7b7e\u800c\u5df2\u3002\u8fd8\u9700\u8981\u7ed3\u5408\u6570\u636e\u672c\u8eab\u7684\u7279\u70b9\u548c\u4e1a\u52a1\u7279\u6027\u8fdb\u884c\u610f\u4e49\u8d4b\u4e88\uff0c\u624d\u80fd\u591f\u4ea7\u751f\u4e0e\u4e4b\u5339\u914d\u7684\u8fd0\u8425\u884c\u52a8\u65b9\u6848\u3002<\/strong><\/p>\n\n\n\n
\u89c2\u5bdf\u805a\u7c7b\u540e\u88ab\u5206\u4e3a 3 \u79cd\u7c7b\u578b\u7684\u516c\u53f8\u5ba2\u6237\uff0c\u7fa4\u4f53\u8868\u73b0\u4e0a\u53ef\u4ee5\u770b\u51fa\uff0c3 \u79cd\u7c7b\u522b\u53ef\u4ee5\u5212\u5206\u4e3a\u3010\u7c7b\u522b1 : \u7c7b\u522b2 : \u7c7b\u522b3 = \u9ad8\u4ef7\u503c : \u4e2d\u4ef7\u503c : \u4f4e\u4ef7\u503c\u3011\u3002<\/strong><\/p>\n\n\n\n
\u5ba2\u6237\u4f53\u9a8c\u7ba1\u7406\u4f9d\u636e\u5bf9\u5e94\u7684\u7c7b\u578b\u91c7\u53d6\u884c\u52a8\u65b9\u6848\u3002<\/p>\n\n\n\n
\u672c\u6587\u63d0\u4f9b\u7684\u4ec5\u4ec5\u662f\u533a\u522b\u4e8e\u5355\u4e00\u4f7f\u7528 RFM \u6a21\u578b\u8fdb\u884c\u5ba2\u6237\u5206\u7fa4\u7684\u601d\u8def\uff0c\u5e0c\u671b\u5bf9\u4f60\u6709\u6240\u542f\u53d1\u3002<\/p>\n\n\n\n
\u672c\u6587\u5b8c.<\/p>\n\n\n\n
\u516c\u4f17\u53f7\uff1a\u9f99\u56fd\u5bcc\uff0c\u4eba\u56e0\u5de5\u7a0b\u7855\u58eb\u3002\u81f4\u529b\u4e8e\u7ec8\u8eab\u5b66\u4e60\u548c\u81ea\u6211\u63d0\u5347\uff0c\u5206\u4eab\u7528\u6237\u7814\u7a76\u3001\u5ba2\u6237\u4f53\u9a8c\u3001\u670d\u52a1\u79d1\u5b66\u7b49\u9886\u57df\u8d44\u8baf\uff0c\u89c2\u70b9\u548c\u4e2a\u4eba\u89c1\u89e3\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"
\u6211\u4eec\u77e5\u9053 RFM \u6a21\u578b\u662f\u5ba2\u6237\u5206\u7fa4\u7684\u91cd\u8981\u6a21\u578b\u4e4b\u4e00\uff0c\u5b83\u4e3b\u8981\u57fa\u4e8e\u5ba2\u6237\u884c\u4e3a\u8fdb\u884c\u5212\u5206\uff0c\u8bc6\u522b\u5ba2\u6237\u4ef7\u503c\u60c5\u51b5\uff0c\u628a\u5ba2\u6237\u5212\u5206\u4e3a 8 \u79cd\u7c7b\u578b\u3002 \u90a3\u4e48\u95ee\u9898\u6765\u4e86\uff0c\u4e00\u5b9a\u8981\u5212\u5206\u4e3a 8 \u79cd\u5ba2\u6237\u7c7b\u578b\u5417\uff1f \u5728\u5b9e\u9645\u7684\u5ba2\u2026<\/p>\n","protected":false},"author":1888,"featured_media":69227,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2036],"tags":[4373],"special":[],"class_list":["post-69226","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-quan","tag-rfm","entry"],"views":65520,"_links":{"self":[{"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/posts\/69226","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/users\/1888"}],"replies":[{"embeddable":true,"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/comments?post=69226"}],"version-history":[{"count":0,"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/posts\/69226\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/media\/69227"}],"wp:attachment":[{"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/media?parent=69226"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/categories?post=69226"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/tags?post=69226"},{"taxonomy":"special","embeddable":true,"href":"https:\/\/www.growthhk.cn\/wp-json\/wp\/v2\/special?post=69226"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}