写出基于TF-IDF+KNN进行中文文本分类的代码

以下是基于TF-IDF+KNN进行中文文本分类的代码示例：

import jieba
import os
import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.neighbors import KNeighborsClassifier

# 数据预处理
def load_data(root):
    labels = []
    corpus = []
    for filename in os.listdir(root):
        label = int(filename.split('_')[0])
        labels.append(label)
        with open(os.path.join(root, filename), 'r', encoding='utf-8') as f:
            text = f.read()
            words = jieba.cut(text)
            corpus.append(' '.join(words))
    return labels, corpus

# 加载数据集
train_labels, train_corpus = load_data('train')
test_labels, test_corpus = load_data('test')

# 构建TF-IDF特征向量
vectorizer = TfidfVectorizer()
train_features = vectorizer.fit_transform(train_corpus)
test_features = vectorizer.transform(test_corpus)

# 构建KNN分类模型
knn = KNeighborsClassifier(n_neighbors=3)
knn.fit(train_features, train_labels)

# 模型评估
accuracy = knn.score(test_features, test_labels)
print(f'Accuracy: {accuracy}')

# 预测新数据
new_text = '这是一个测试样本'
new_words = jieba.cut(new_text)
new_feature = vectorizer.transform([' '.join(new_words)])
predicted_label = knn.predict(new_feature)[0]
print(f'Predicted label: {predicted_label}')

需要注意的是，上述代码中使用了jieba库进行中文分词，需要先安装jieba库。另外，该代码示例仅适用于数据集比较小的情况，对于大规模数据集需要采用分布式计算等技术进行优化。