机器学习项目

前置要求

pip install scikit-learn pandas numpy matplotlib

1. 房价预测（回归）

目标：使用线性回归预测房价

from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
import pandas as pd

# 读取数据
df = pd.read_csv('house_prices.csv')

# 特征与标签
X = df[['area', 'rooms', 'floor']]
y = df['price']

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

# 训练模型
model = LinearRegression()
model.fit(X_train, y_train)

# 预测与评估
predictions = model.predict(X_test)
mse = mean_squared_error(y_test, predictions)
print(f"均方误差: {mse:.2f}")
print(f"模型系数: {model.coef_}")

# 预测新房
new_house = [[120, 3, 8]]  # 120平, 3室, 8楼
predicted = model.predict(new_house)
print(f"预测价格: {predicted[0]:.2f} 万元")

知识点：train_test_split、线性回归、模型评估

---

2. 鸢尾花分类（分类）

目标：使用 KNN 算法对鸢尾花进行分类

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import accuracy_score, classification_report

# 加载数据
iris = load_iris()
X, y = iris.data, iris.target

# 划分数据
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 训练 KNN 模型
knn = KNeighborsClassifier(n_neighbors=3)
knn.fit(X_train, y_train)

# 预测
predictions = knn.predict(X_test)

# 评估
accuracy = accuracy_score(y_test, predictions)
print(f"准确率: {accuracy:.2%}")
print(classification_report(y_test, predictions, target_names=iris.target_names))

# 预测新样本
new_sample = [[5.1, 3.5, 1.4, 0.2]]
result = knn.predict(new_sample)
print(f"预测类别: {iris.target_names[result[0]]}")

知识点：KNeighborsClassifier、分类报告、准确率评估

---

3. 客户聚类（聚类）

目标：使用 K-Means 对客户进行分群

from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
import pandas as pd
import matplotlib.pyplot as plt

# 读取客户数据
df = pd.read_csv('customers.csv')
X = df[['age', 'income', 'spending_score']]

# 特征标准化
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

# K-Means 聚类
kmeans = KMeans(n_clusters=3, random_state=42, n_init=10)
df['cluster'] = kmeans.fit_predict(X_scaled)

# 查看各群特征
print(df.groupby('cluster')[['age', 'income', 'spending_score']].mean())

# 可视化（收入 vs 消费评分）
plt.scatter(df['income'], df['spending_score'], c=df['cluster'], cmap='viridis')
plt.xlabel('收入')
plt.ylabel('消费评分')
plt.title('客户聚类结果')
plt.colorbar(label='群集')
plt.savefig('clusters.png')
plt.show()

知识点：KMeans 聚类、StandardScaler 标准化、数据可视化

---

4. 手写数字识别

目标：使用 MNIST 数据集和 MLP 分类器

from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
from sklearn.neural_network import MLPClassifier
from sklearn.metrics import accuracy_score

# 加载数据（8x8 手写数字）
digits = load_digits()
X, y = digits.data, digits.target

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 多层感知机
mlp = MLPClassifier(hidden_layer_sizes=(64, 32), max_iter=500, random_state=42)
mlp.fit(X_train, y_train)

predictions = mlp.predict(X_test)
accuracy = accuracy_score(y_test, predictions)
print(f"手写数字识别准确率: {accuracy:.2%}")

知识点：load_digits 数据集、MLPClassifier、神经网络

---

5. 泰坦尼克号生存预测

目标：完整的机器学习流程

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, classification_report
from sklearn.preprocessing import LabelEncoder

# 加载数据（使用 sklearn 内置或 CSV）
url = 'https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv'
df = pd.read_csv(url)

# 特征工程
features = ['Pclass', 'Sex', 'Age', 'Fare', 'SibSp', 'Parch']
df = df[features + ['Survived']].dropna()

le = LabelEncoder()
df['Sex'] = le.fit_transform(df['Sex'])

X = df[features]
y = df['Survived']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 随机森林
rf = RandomForestClassifier(n_estimators=100, random_state=42)
rf.fit(X_train, y_train)
preds = rf.predict(X_test)

print(f"准确率: {accuracy_score(y_test, preds):.2%}")
print("特征重要性:", dict(zip(features, rf.feature_importances_)))

知识点：RandomForestClassifier、特征工程、LabelEncoder、特征重要性

---

6. 情感分析

目标：使用朴素贝叶斯进行文本分类

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# 训练数据
texts = [
    "这部电影太好了，非常喜欢", "质量不错，推荐购买",
    "服务态度很差，再也不来了", "太差了，完全不能用",
    "外观好看，功能强大", "包装破损，物流太慢",
    "性价比很高，满意", "虚假宣传，很失望"
]
labels = [1, 1, 0, 0, 1, 0, 1, 0]  # 1=正面 0=负面

# 文本向量化
vectorizer = CountVectorizer()
X = vectorizer.fit_transform(texts)
y = labels

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)

# 朴素贝叶斯
nb = MultinomialNB()
nb.fit(X_train, y_train)
preds = nb.predict(X_test)

print(f"准确率: {accuracy_score(y_test, preds):.2%}")

# 测试新文本
new_text = vectorizer.transform(["这个产品真的很棒"])
print(f"情感预测: {'正面' if nb.predict(new_text)[0] == 1 else '负面'}")

知识点：CountVectorizer 文本向量化、朴素贝叶斯、文本分类