中文文本分类python代码

以下是一个简单的中文文本分类的 Python 代码示例，使用了朴素贝叶斯分类器：

import jieba
import os
import random
from sklearn.naive_bayes import MultinomialNB
from sklearn.feature_extraction.text import TfidfVectorizer

# 读取数据集
def read_dataset(folder_path):
    data = []
    labels = []
    for root, dirs, files in os.walk(folder_path):
        for file in files:
            file_path = os.path.join(root, file)
            with open(file_path, 'r', encoding='utf-8') as f:
                content = f.read()
                data.append(' '.join(jieba.cut(content)))
                labels.append(root.split('\\')[-1])
    return data, labels

# 分割数据集
def split_dataset(data, labels, test_ratio=0.2):
    data_size = len(data)
    index = list(range(data_size))
    random.shuffle(index)
    test_size = int(data_size * test_ratio)
    test_index = index[:test_size]
    train_index = index[test_size:]
    train_data = [data[i] for i in train_index]
    train_labels = [labels[i] for i in train_index]
    test_data = [data[i] for i in test_index]
    test_labels = [labels[i] for i in test_index]
    return train_data, train_labels, test_data, test_labels

# 特征提取
def feature_extraction(train_data, test_data):
    vectorizer = TfidfVectorizer(max_features=3000)
    train_features = vectorizer.fit_transform(train_data)
    test_features = vectorizer.transform(test_data)
    return train_features, test_features

# 训练模型
def train_model(train_features, train_labels):
    classifier = MultinomialNB(alpha=0.01)
    classifier.fit(train_features, train_labels)
    return classifier

# 预测
def predict(classifier, test_features):
    predict_labels = classifier.predict(test_features)
    return predict_labels

if __name__ == '__main__':
    # 数据集路径
    data_folder_path = 'dataset'
    # 读取数据集
    data, labels = read_dataset(data_folder_path)
    # 分割数据集
    train_data, train_labels, test_data, test_labels = split_dataset(data, labels)
    # 特征提取
    train_features, test_features = feature_extraction(train_data, test_data)
    # 训练模型
    classifier = train_model(train_features, train_labels)
    # 预测
    predict_labels = predict(classifier, test_features)
    # 计算准确率
    accuracy = sum(1 for i in range(len(predict_labels)) if predict_labels[i] == test_labels[i]) / float(len(predict_labels))
    print('Accuracy: {:.2%}'.format(accuracy))

其中，数据集的格式为：

dataset/
├── class1/
│   ├── file1.txt
│   ├── file2.txt
│   └── ...
├── class2/
│   ├── file1.txt
│   ├── file2.txt
│   └── ...
└── ...

其中 `jieba` 库用于中文分词，`TfidfVectorizer` 用于提取特征，`MultinomialNB` 用于训练朴素贝叶斯分类器。