朴素贝叶斯垃圾邮件文档识别代码

以下是一个简单的Python实现朴素贝叶斯垃圾邮件文档识别的代码：

import os
import string

# 定义停用词表
stopwords = []

# 加载停用词表
with open('stopwords.txt', 'r', encoding='utf-8') as f:
    for line in f:
        stopwords.append(line.strip())

# 定义训练数据集和测试数据集的路径
train_dir = 'train'
test_dir = 'test'

# 定义文档类别
categories = os.listdir(train_dir)

# 定义词典
words_dict = {}

# 统计词频
def count_words(filepath):
    words_dict = {}
    with open(filepath, 'r', encoding='utf-8') as f:
        for line in f:
            line = line.strip()
            for word in line.split():
                word = word.strip(string.punctuation)
                if word and word not in stopwords:
                    if word not in words_dict:
                        words_dict[word] = 0
                    words_dict[word] += 1
    return words_dict

# 训练朴素贝叶斯分类器
def train():
    global words_dict
    for category in categories:
        category_path = os.path.join(train_dir, category)
        for filename in os.listdir(category_path):
            filepath = os.path.join(category_path, filename)
            words_dict_temp = count_words(filepath)
            for word in words_dict_temp:
                if word not in words_dict:
                    words_dict[word] = [0] * len(categories)
                words_dict[word][categories.index(category)] += words_dict_temp[word]

# 预测分类
def predict(filepath):
    words_dict_temp = count_words(filepath)
    scores = [0] * len(categories)
    for word in words_dict_temp:
        if word in words_dict:
            for i in range(len(categories)):
                scores[i] += words_dict[word][i]
    return categories[scores.index(max(scores))]

# 测试分类器
def test():
    total = 0
    correct = 0
    for category in categories:
        category_path = os.path.join(test_dir, category)
        for filename in os.listdir(category_path):
            filepath = os.path.join(category_path, filename)
            total += 1
            if predict(filepath) == category:
                correct += 1
    print('Accuracy:', correct / total)

if __name__ == '__main__':
    train()
    test()

这个代码实现了一个简单的朴素贝叶斯垃圾邮件文档识别器，它可以从训练数据集中学习每个单词在不同类别中的出现频率，并根据测试数据集中的文档来预测文档所属的类别。注意，在实际应用中，还需要对模型进行优化和改进，以提高准确率和鲁棒性。