環球資訊：【案例分享】降維案例探究

降維案例

(相關資料圖)

案例一步驟步驟一步驟二步驟三

案例一

探究: 用戶對物品類別的喜好細分降維.

數據:

products.csv 商品信息order_products__prior.csv 訂單與商品信息orders.csv 用戶的訂單信息aisles.csv 商品所屬具體物品類別

步驟

合并各張表到一張表當中: pd.merge()建立一個類似行, 列數據使用 PCA 分析

步驟一

import pandas as pdfrom sklearn.decomposition import PCA# 讀取四張表的數據prior = pd.read_csv("order_products__prior.csv")products = pd.read_csv("products.csv")orders = pd.read_csv("orders.csv")aisles = pd.read_csv("aisles.csv")# 合并四張表到一張表_mg = pd.merge(prior, products, on=["product_id", "product_id"])_mg = pd.merge(_mg, orders, on=["order_id", "order_id"])mt = pd.merge(_mg, aisles, on=["aisle_id", "aisle_id"])print(mt.head())輸出結果:0         2       33120  ...                     8.0   eggs1        26       33120  ...                     7.0   eggs2       120       33120  ...                    10.0   eggs3       327       33120  ...                     8.0   eggs4       390       33120  ...                     9.0   eggs

步驟二

import pandas as pdfrom sklearn.decomposition import PCA# 讀取四張表的數據prior = pd.read_csv("order_products__prior.csv")products = pd.read_csv("products.csv")orders = pd.read_csv("orders.csv")aisles = pd.read_csv("aisles.csv")# 合并四張表到一張表_mg = pd.merge(prior, products, on=["product_id", "product_id"])_mg = pd.merge(_mg, orders, on=["order_id", "order_id"])mt = pd.merge(_mg, aisles, on=["aisle_id", "aisle_id"])# 交叉表 (特殊的分組工具)cross = pd.crosstab(mt["user_id"],mt["aisle"])# 輸出頭5條數據print(cross.head())輸出結果:aisle    air fresheners candles  asian foods  ...  white wines  yogurtuser_id                                       ...                     1                             0            0  ...            0       12                             0            3  ...            0      423                             0            0  ...            0       04                             0            0  ...            0       05                             0            2  ...            0       3

步驟三

import pandas as pdfrom sklearn.decomposition import PCA# 讀取四張表的數據prior = pd.read_csv("order_products__prior.csv")products = pd.read_csv("products.csv")orders = pd.read_csv("orders.csv")aisles = pd.read_csv("aisles.csv")# 合并四張表到一張表_mg = pd.merge(prior, products, on=["product_id", "product_id"])_mg = pd.merge(_mg, orders, on=["order_id", "order_id"])mt = pd.merge(_mg, aisles, on=["aisle_id", "aisle_id"])# 交叉表 (特殊的分組工具)cross = pd.crosstab(mt["user_id"], mt["aisle"])# 進行主成分分析pca = PCA(n_components=0.9)data = pca.fit_transform(cross)# 輸出數據print(data)輸出結果:[[-2.42156587e+01  2.42942720e+00 -2.46636975e+00 ...  6.86800336e-01   1.69439402e+00 -2.34323022e+00] [ 6.46320806e+00  3.67511165e+01  8.38255336e+00 ...  4.12121252e+00   2.44689740e+00 -4.28348478e+00] [-7.99030162e+00  2.40438257e+00 -1.10300641e+01 ...  1.77534453e+00  -4.44194030e-01  7.86665571e-01] ... [ 8.61143331e+00  7.70129866e+00  7.95240226e+00 ... -2.74252456e+00   1.07112531e+00 -6.31925661e-02] [ 8.40862199e+01  2.04187340e+01  8.05410372e+00 ...  7.27554259e-01   3.51339470e+00 -1.79079914e+01] [-1.39534562e+01  6.64621821e+00 -5.23030367e+00 ...  8.25329076e-01   1.38230701e+00 -2.41942061e+00]]

查看 data.shape, 我們可以發現 類別由 134 個變為了 27 個.