pip install scikit-learn pandas numpy

import imaplib
import email
from email.header import decode_header

def fetch_emails(username, password, imap_server='imap.qq.com', limit=50):
    # 連接到服務(wù)器
    mail = imaplib.IMAP4_SSL(imap_server)
    mail.login(username, password)
    
    # 選擇收件箱
    mail.select('INBOX')
    
    # 搜索所有郵件
    status, messages = mail.search(None, 'ALL')
    email_ids = messages[0].split()
    
    # 取最新的 limit 封郵件
    emails_data = []
    for e_id in email_ids[-limit:]:
        status, msg_data = mail.fetch(e_id, '(RFC822)')
        
        # 解析郵件內(nèi)容
        for response_part in msg_data:
            if isinstance(response_part, tuple):
                msg = email.message_from_bytes(response_part[1])
                
                # 解碼主題
                subject, encoding = decode_header(msg["Subject"])[0]
                if isinstance(subject, bytes):
                    subject = subject.decode(encoding if encoding else 'utf-8')
                
                # 提取發(fā)件人
                from_ = msg.get("From")
                
                # 提取正文 (簡(jiǎn)單處理，僅取第一個(gè)text/plain部分)
                body = ""
                if msg.is_multipart():
                    for part in msg.walk():
                        content_type = part.get_content_type()
                        content_disposition = str(part.get("Content-Disposition"))
                        if content_type == "text/plain" and "attachment" not in content_disposition:
                            body = part.get_payload(decode=True).decode()
                            break
                else:
                    body = msg.get_payload(decode=True).decode()
                
                emails_data.append({
                    "id": e_id,
                    "subject": subject,
                    "from": from_,
                    "body": body
                })
    mail.close()
    mail.logout()
    return emails_data

# 示例調(diào)用 (請(qǐng)?zhí)鎿Q為你自己的賬號(hào)和授權(quán)碼)
# emails = fetch_emails('your_email@qq.com', 'your_auth_code')
# print(f"獲取了 {len(emails)} 封郵件")

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.pipeline import make_pipeline
from sklearn.metrics import classification_report

# 1. 模擬訓(xùn)練數(shù)據(jù) (實(shí)際中應(yīng)從歷史郵件提取并標(biāo)注)
# 0: 重要/工作, 1: 垃圾/無(wú)關(guān)
data = [
    ("【報(bào)警】服務(wù)器CPU使用率超過(guò)90%", 0),
    ("項(xiàng)目周會(huì)通知 - 請(qǐng)準(zhǔn)時(shí)參加", 0),
    ("恭喜你獲得100元優(yōu)惠券", 1),
    ("發(fā)票確認(rèn)：請(qǐng)查收附件", 0),
    ("澳門(mén)首家賭場(chǎng)上線啦", 1),
    ("Python技術(shù)交流群邀請(qǐng)", 1),
    ("緊急：數(shù)據(jù)庫(kù)連接失敗", 0),
    ("Weekly Report - Team A", 0),
    ("代開(kāi)發(fā)票，誠(chéng)信合作", 1),
    ("系統(tǒng)維護(hù)通知", 0)
]

# 轉(zhuǎn)換為DataFrame
df = pd.DataFrame(data, columns=['text', 'label'])

# 2. 構(gòu)建管道 (Pipeline)
# TfidfVectorizer: 將文本轉(zhuǎn)換為數(shù)值向量
# MultinomialNB: 樸素貝葉斯分類器
model = make_pipeline(TfidfVectorizer(), MultinomialNB())

# 3. 訓(xùn)練模型
# 這里的X是郵件內(nèi)容，y是標(biāo)簽
X_train = df['text']
y_train = df['label']

model.fit(X_train, y_train)

print("模型訓(xùn)練完成！")

# 模擬新的郵件
new_emails = [
    "你好，這是本周的工作總結(jié)",
    "急急急！內(nèi)部系統(tǒng)升級(jí)維護(hù)",
    "注冊(cè)即送大禮包，點(diǎn)擊領(lǐng)取",
    "財(cái)務(wù)部門(mén)請(qǐng)注意查收郵件"
]

predictions = model.predict(new_emails)

for email, pred in zip(new_emails, predictions):
    label = "重要" if pred == 0 else "垃圾/忽略"
    print(f"郵件: '{email}' -> 預(yù)測(cè)結(jié)果: {label}")

def auto_process_emails(username, password, model):
    # 1. 獲取郵件
    emails = fetch_emails(username, password, limit=20)
    
    # 2. 連接服務(wù)器準(zhǔn)備操作
    mail = imaplib.IMAP4_SSL('imap.qq.com')
    mail.login(username, password)
    mail.select('INBOX')
    
    for item in emails:
        # 提取特征 (這里簡(jiǎn)單用subject + body)
        text = item['subject'] + " " + item['body']
        
        # 3. 預(yù)測(cè)
        prediction = model.predict([text])[0]
        
        if prediction == 1: # 假設(shè)1是垃圾/低優(yōu)先級(jí)
            print(f"正在處理低優(yōu)先級(jí)郵件: {item['subject']}")
            
            # 獲取郵件ID (注意：fetch_emails里的ID是bytes，需要轉(zhuǎn)碼查找)
            # 這里為了演示簡(jiǎn)化，實(shí)際需要根據(jù)RFC822頭匹配或索引處理
            # 簡(jiǎn)單的IMAP操作示例：
            # 標(biāo)記為已讀
            # mail.store(item['id'], '+FLAGS', '\\Seen')
            # 移動(dòng)到其他文件夾 (需要先創(chuàng)建該文件夾)
            # mail.copy(item['id'], 'ProcessedBox')
            
        else:
            print(f"發(fā)現(xiàn)重要郵件: {item['subject']} (保持原樣)")

# 注意：在實(shí)際運(yùn)行前，請(qǐng)確保你已經(jīng)用足夠多的數(shù)據(jù)訓(xùn)練了模型
# auto_process_emails('user', 'pass', model)