AI行业应用教程¶

from openai import OpenAI

client = OpenAI(api_key="your-api-key")

def generate_marketing_copy(product, features, target_user):
    prompt = f"""
    为以下产品生成营销文案：

    产品：{product}
    特点：{features}
    目标用户：{target_user}

    请生成包含标题、正文、CTA的完整文案。
    """

    response = client.chat.completions.create(
        model="gpt-5.5",
        messages=[{"role": "user", "content": prompt}]
    )

    return response.choices[0].message.content

# 使用Code Interpreter分析数据

# 1. 上传数据文件
# 2. 描述分析需求
# 3. AI自动生成代码并执行
# 4. 返回分析结果和图表

import pandas as pd
import matplotlib.pyplot as plt
from openai import OpenAI

client = OpenAI(api_key="your-api-key")

def analyze_sales_data(file_path):
    # 读取数据
    df = pd.read_csv(file_path)

    # 数据概览
    summary = df.describe()

    # 让AI分析
    prompt = f"""
    请分析以下销售数据：

    数据概览：
    {summary.to_string()}

    前5行数据：
    {df.head().to_string()}

    请提供：
    1. 数据洞察
    2. 关键指标
    3. 趋势分析
    4. 建议措施
    """

    response = client.chat.completions.create(
        model="gpt-5.5",
        messages=[{"role": "user", "content": prompt}]
    )

    return response.choices[0].message.content

# Dify API调用

import requests

def call_dify_workflow(workflow_id, inputs):
    url = f"https://api.dify.ai/v1/workflows/{workflow_id}/run"

    headers = {
        "Authorization": "Bearer your-api-key",
        "Content-Type": "application/json"
    }

    data = {
        "inputs": inputs,
        "response_mode": "blocking"
    }

    response = requests.post(url, headers=headers, json=data)
    return response.json()

from openai import OpenAI
import json

class AICustomerService:
    def __init__(self):
        self.client = OpenAI(api_key="your-api-key")
        self.knowledge_base = self.load_knowledge_base()

    def load_knowledge_base(self):
        # 加载FAQ知识库
        return {
            "退款": "退款政策：7天内无理由退款...",
            "发货": "发货时间：订单后24小时内发货...",
            "会员": "会员权益：享受95折优惠..."
        }

    def classify_intent(self, user_input):
        prompt = f"""
        请判断用户意图，返回JSON格式：

        用户输入：{user_input}

        可选意图：{list(self.knowledge_base.keys())}, 其他

        格式：{{"intent": "意图", "confidence": 0.95}}
        """

        response = self.client.chat.completions.create(
            model="gpt-5.5",
            messages=[{"role": "user", "content": prompt}],
            response_format={"type": "json_object"}
        )

        return json.loads(response.choices[0].message.content)

    def handle_request(self, user_input):
        intent = self.classify_intent(user_input)

        if intent["confidence"] > 0.8:
            if intent["intent"] in self.knowledge_base:
                return self.knowledge_base[intent["intent"]]

        # 低置信度转人工
        return "正在为您转接人工客服，请稍候..."

def generate_quiz(topic, difficulty, num_questions):
    prompt = f"""
    请为"{topic}"生成{num_questions}道{difficulty}难度的题目。

    要求：
    1. 题型多样（选择、填空、简答）
    2. 覆盖核心知识点
    3. 包含答案和解析

    请以JSON格式返回：
    {{
        "questions": [
            {{
                "type": "选择题",
                "question": "题目内容",
                "options": ["A", "B", "C", "D"],
                "answer": "A",
                "explanation": "解析"
            }}
        ]
    }}
    """

    response = client.chat.completions.create(
        model="gpt-5.5",
        messages=[{"role": "user", "content": prompt}],
        response_format={"type": "json_object"}
    )

    return json.loads(response.choices[0].message.content)