根据您提供的文档内容，我将为您分模块提供Python程序的主要框架和关键代码段。完整的程序较为复杂，涉及网络请求、数据分析、可视化等多个方面，这里给出主要思路和核心代码片段： ### 项目一：豆瓣电影信息爬取与存储 #### 1.1 网页内容分析与爬取 **准备工作** ```python import requests from bs4 import BeautifulSoup import time import pandas as pd import json ``` **爬取函数** ```python def fetch_movie_data(category, start): url = f"https://movie.douban.com/j/chart/top_list?type={category}&interval_id=100:90&action=&start={start}&limit=20" headers = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36' } response = requests.get(url, headers=headers) if response.status_code == 200: return response.json() else: print(f"Failed to retrieve {url}") return None ``` **主逻辑** ```python categories = ["plot", "comedy", "action", "love", "sci-fi", "suspense", "thriller", "horror", "history", "war", "crime", "fantasy", "adventure"] all_movies = [] for category in categories: for start in range(0, 100, 20): # 每次抓取20条，直到100条 movies = fetch_movie_data(category, start) all_movies.extend(movies) time.sleep(5) df = pd.DataFrame(all_movies) df.to_csv('film_info.csv', index=False) ``` #### 项目二：国家发行各类别电影的可视化分析 **数据预处理** ```python df = pd.read_csv('film_info.csv') df['regions'] = df['regions'].str.split('/') df_exploded = df.explode('regions') df_pivot = pd.crosstab(df_exploded['regions'], df_exploded['types']) df_pivot['总计'] = df_pivot.sum(axis=1) df_pivot.loc['总计'] = df_pivot.sum() df_sorted = df_pivot.sort_values(by='总计', ascending=True) df_sorted.to_csv('df_counties.csv') ``` **可视化分析** ```python import matplotlib.pyplot as plt select_data = df_sorted.tail(10)[df_sorted.columns[:5]] ax = select_data.plot(kind='bar', stacked=True, figsize=(10, 8)) plt.title("各国发行电影数量") plt.xlabel("国家") plt.ylabel("电影数量") plt.savefig("各国家分类电影堆叠图.jpg") plt.show() ``` #### 项目三：基于近邻的电影推荐研究 **数据预处理** ```python from sklearn.preprocessing import MinMaxScaler df_films = pd.read_csv('film_info.csv').drop_duplicates(subset='title', keep='first') scaler = MinMaxScaler() scaled_features = scaler.fit_transform(df_films[['score', 'vote_count']]) ``` **计算相似度** ```python from scipy.spatial.distance import cdist import numpy as np df_sim = pd.DataFrame(cdist(scaled_features, scaled_features), index=df_films['title'], columns=df_films['title']) recommendations = {} for movie in df_films['title']: sim_scores = df_sim[movie].sort_values()[1:6] recommendations[movie] = list(zip(sim_scores.index, sim_scores.values)) data_tuijian = pd.DataFrame.from_dict(recommendations, orient='index') data_tuijian.to_csv('data_tuijian.csv') ``` 以上代码仅为简化版本，实际应用时需要根据具体需求调整参数和细节。希望这些代码能帮助您完成作业！如果有任何不明白的地方，请随时提问。