以下是根据文档核心要点编写的代码实现，涵盖了几种经典机器学习算法的应用，如KNN、SVM、朴素贝叶斯、决策树和随机森林。我们将使用`scikit-learn`库来实现这些算法，并以心脏病预测为例展示它们的效果。此外，还会加入网格搜索来优化随机森林的超参数。 ### 1. 数据准备 ```python import numpy as np import pandas as pd from sklearn.model_selection import train_test_split from sklearn.preprocessing import StandardScaler from sklearn.metrics import accuracy_score, f1_score, confusion_matrix import seaborn as sns import matplotlib.pyplot as plt # 读取数据 df_heart = pd.read_csv("../input/heart-dataset/heart.csv") # 特征和目标变量分离 X = df_heart.drop(columns=['target']) y = df_heart['target'] # 数据集划分 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) # 特征缩放 scaler = StandardScaler() X_train = scaler.fit_transform(X_train) X_test = scaler.transform(X_test) ``` ### 2. KNN算法实现 ```python from sklearn.neighbors import KNeighborsClassifier # 定义KNN分类器并训练 K = 5 knn = KNeighborsClassifier(n_neighbors=K) knn.fit(X_train, y_train) # 预测并评估 y_pred_knn = knn.predict(X_test) print(f"KNN预测准确率: {accuracy_score(y_test, y_pred_knn)*100:.2f}%") print(f"KNN预测F1分数: {f1_score(y_test, y_pred_knn)*100:.2f}%") print('KNN混淆矩阵:') print(confusion_matrix(y_test, y_pred_knn)) # 寻找最佳K值 acc_scores = [] f1_scores = [] for k in range(1, 15): knn = KNeighborsClassifier(n_neighbors=k) knn.fit(X_train, y_train) y_pred = knn.predict(X_test) acc_scores.append(accuracy_score(y_test, y_pred)) f1_scores.append(f1_score(y_test, y_pred)) plt.plot(range(1, 15), acc_scores, label='Accuracy') plt.plot(range(1, 15), f1_scores, label='F1 Score', linestyle='--') plt.xlabel('K Value') plt.ylabel('Score') plt.legend() plt.show() ``` ### 3. SVM算法实现 ```python from sklearn.svm import SVC # 定义SVM分类器并训练 svm = SVC(random_state=42) svm.fit(X_train, y_train) # 预测并评估 y_pred_svm = svm.predict(X_test) print(f"SVM预测准确率: {accuracy_score(y_test, y_pred_svm)*100:.2f}%") print(f"SVM预测F1分数: {f1_score(y_test, y_pred_svm)*100:.2f}%") print('SVM混淆矩阵:') print(confusion_matrix(y_test, y_pred_svm)) ``` ### 4. 朴素贝叶斯算法实现 ```python from sklearn.naive_bayes import GaussianNB # 定义朴素贝叶斯分类器并训练 nb = GaussianNB() nb.fit(X_train, y_train) # 预测并评估 y_pred_nb = nb.predict(X_test) print(f"朴素贝叶斯预测准确率: {accuracy_score(y_test, y_pred_nb)*100:.2f}%") print(f"朴素贝叶斯预测F1分数: {f1_score(y_test, y_pred_nb)*100:.2f}%") print('朴素贝叶斯混淆矩阵:') print(confusion_matrix(y_test, y_pred_nb)) ``` ### 5. 决策树算法实现 ```python from sklearn.tree import DecisionTreeClassifier # 定义决策树分类器并训练 dtc = DecisionTreeClassifier(random_state=42) dtc.fit(X_train, y_train) # 预测并评估 y_pred_dtc = dtc.predict(X_test) print(f"决策树预测准确率: {accuracy_score(y_test, y_pred_dtc)*100:.2f}%") print(f"决策树预测F1分数: {f1_score(y_test, y_pred_dtc)*100:.2f}%") print('决策树混淆矩阵:') print(confusion_matrix(y_test, y_pred_dtc)) ``` ### 6. 随机森林算法实现及超参数优化 ```python from sklearn.ensemble import RandomForestClassifier from sklearn.model_selection import GridSearchCV # 定义随机森林分类器 rf = RandomForestClassifier(random_state=42) # 设置超参数网格 param_grid = { 'n_estimators': [100, 300], 'max_features': ['auto', 'sqrt'], 'max_depth': [None, 10, 20, 30], 'min_samples_split': [2, 5, 10], 'min_samples_leaf': [1, 2, 4], 'bootstrap': [True, False] } # 使用网格搜索进行超参数优化 grid_search = GridSearchCV(estimator=rf, param_grid=param_grid, cv=5, scoring='accuracy', n_jobs=-1) grid_search.fit(X_train, y_train) # 输出最佳参数 print("最佳参数组合:", grid_search.best_params_) # 使用最佳参数重新训练并评估 best_rf = grid_search.best_estimator_ y_pred_best_rf = best_rf.predict(X_test) print(f"参数优化后随机森林预测准确率: {accuracy_score(y_test, y_pred_best_rf)*100:.2f}%") print(f"参数优化后随机森林预测F1分数: {f1_score(y_test, y_pred_best_rf)*100:.2f}%") print('参数优化后随机森林混淆矩阵:') print(confusion_matrix(y_test, y_pred_best_rf)) # 绘制混淆矩阵热力图 plt.figure(figsize=(6, 6)) sns.heatmap(confusion_matrix(y_test, y_pred_best_rf), annot=True, cmap="Blues", fmt="d", cbar=False) plt.title("Random Forest (Best Parameters) Confusion Matrix") plt.show() ``` 这段代码涵盖了经典的机器学习算法实现及其评估，同时还展示了如何使用网格搜索来优化随机森林的超参数。希望这对您有所帮助！如果有任何问题或需要进一步的解释，请随时告诉我。