机器学习中的分类：决策树、随机森林及其应用

import pandas as pd
import random
import string

# 随机生成数据的数量
num_records = 1000

# 随机生成用户ID
def generate_user_id():
    return random.randint(1, 1000)

# 随机生成性别
def generate_gender():
    return random.choice(['M', 'F'])

# 随机生成年龄（假设年龄范围在18到60之间）
def generate_age():
    return random.randint(18, 60)

# 随机生成优惠券ID
def generate_coupon_id():
    return ''.join(random.choices(string.digits, k=4))

# 随机生成优惠力度（假设优惠力度范围在5%到50%之间）
def generate_discount():
    return f"{random.randint(5, 50)}%"

# 随机生成购买历史（假设每个用户的购买历史是一个整数，范围在1到20之间）
def generate_purchase_history():
    return random.randint(1, 20)

# 随机生成优惠券是否被使用（1表示使用，0表示未使用）
def generate_coupon_used():
    return random.choice([0, 1])

# 生成数据
data = []
for _ in range(num_records):
    data.append([
        generate_user_id(),
        generate_age(),
        generate_gender(),
        generate_coupon_id(),
        generate_discount(),
        generate_purchase_history(),
        generate_coupon_used()
    ])

# 创建DataFrame
columns = ['user_id', 'age', 'gender', 'coupon_id', 'discount', 'purchase_history', 'coupon_used']
df = pd.DataFrame(data, columns=columns)

# 保存为CSV文件
df.to_csv('o2o_data.csv', index=False)

print("CSV文件已生成：o2o_data.csv")

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder

# 加载数据
data = pd.read_csv('o2o_data.csv')

# 处理缺失值
data = data.dropna()  # 丢弃含有缺失值的行

# 特征选择与标签编码
features = data[['age', 'gender', 'discount', 'purchase_history']]  # 选择特征
labels = data['coupon_used']  # 目标变量

# 性别特征标签编码
le = LabelEncoder()
features['gender'] = le.fit_transform(features['gender'])

# 将折扣从百分比转化为数值
features['discount'] = features['discount'].apply(lambda x: float(x.strip('%')) / 100)

# 划分训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(features, labels, test_size=0.2, random_state=42)

from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix
import matplotlib.pyplot as plt
from sklearn.tree import plot_tree

# 初始化决策树模型
dt_model = DecisionTreeClassifier(random_state=42)

# 训练模型
dt_model.fit(X_train, y_train)

# 预测结果
y_pred_dt = dt_model.predict(X_test)

# 评估模型性能
print("决策树模型准确率:", accuracy_score(y_test, y_pred_dt))
print("\n分类报告:\n", classification_report(y_test, y_pred_dt))
print("\n混淆矩阵:\n", confusion_matrix(y_test, y_pred_dt))

# 可视化决策树
plt.figure(figsize=(12,8))
plot_tree(dt_model, filled=True, feature_names=features.columns, class_names=['Not Used', 'Used'], rounded=True)
plt.show()

from sklearn.ensemble import RandomForestClassifier

# 初始化随机森林模型
rf_model = RandomForestClassifier(n_estimators=100, random_state=42)

# 训练模型
rf_model.fit(X_train, y_train)

# 预测结果
y_pred_rf = rf_model.predict(X_test)

# 评估模型性能
print("随机森林模型准确率:", accuracy_score(y_test, y_pred_rf))
print("\n分类报告:\n", classification_report(y_test, y_pred_rf))
print("\n混淆矩阵:\n", confusion_matrix(y_test, y_pred_rf))

# 可视化特征重要性
feature_importances = rf_model.feature_importances_
plt.barh(features.columns, feature_importances)
plt.xlabel('Feature Importance')
plt.title('Random Forest Feature Importance')
plt.show()