用 Python 看清岗位结构变化：数据采集、入库与可视化

import csv
import time

import requests


def fetch_page(keyword: str, page: int) -> list:
    """按关键词拉取一页职位列表，接口地址请替换为合规数据源。"""
    resp = requests.get(
        "https://你的数据源地址/api/jobs",
        params={"kw": keyword, "page": page},
        headers={"User-Agent": "research-script/1.0"},
        timeout=10,
    )
    resp.raise_for_status()
    return resp.json()["items"]


rows = []
for kw in ["智能体", "数字人", "AI运营", "客服", "文案"]:
    for p in range(1, 4):
        rows += fetch_page(kw, p)
        time.sleep(2)

with open("jobs.csv", "w", newline="", encoding="utf-8") as f:
    writer = csv.DictWriter(
        f, fieldnames=["title", "salary_min", "salary_max", "city", "publish_date"]
    )
    writer.writeheader()
    writer.writerows(rows)

CREATE TABLE job_posting (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    title TEXT NOT NULL,
    category TEXT,
    salary_min REAL,
    salary_max REAL,
    city TEXT,
    publish_date TEXT
);
CREATE INDEX idx_category ON job_posting(category);

import pandas as pd

df = pd.read_csv("jobs.csv")

AI_KEYS = ["智能体", "数字人", "AI", "大模型", "提示词"]


def classify(title: str) -> str:
    return "AI应用类" if any(k in title for k in AI_KEYS) else "传统运营类"


df["category"] = df["title"].map(classify)
df["salary_mid"] = (df["salary_min"] + df["salary_max"]) / 2

summary = df.groupby("category").agg(
    职位数=("title", "count"),
    薪资中位数=("salary_mid", "median"),
)
print(summary)

import matplotlib.pyplot as plt

summary["职位数"].plot(kind="bar")
plt.ylabel("职位数")
plt.tight_layout()
plt.savefig("jobs.png", dpi=150)