Analyzing and Visualizing Maoyan Movie Reviews for “Chinese Doctors” Using Python

This article explains a complete workflow for collecting, processing, and visualizing user comments of the Chinese movie "Chinese Doctors" from the Maoyan platform. It begins with a web‑scraping step that retrieves comment JSON data, including user ID, nickname, city, content, score, and timestamp.

Data acquisition is performed with a simple function that sends an HTTP GET request with appropriate headers:

# 获取数据，根据url获取
def get_data(url):
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.123 Safari/537.36',
    }
    response = requests.get(url=url, headers=headers)
    html_data = response.text
    return html_data

The JSON response is parsed to extract the cmts list, and each comment is transformed into a dictionary containing the fields mentioned above:

# 处理数据
def parse_data(html):
    data = json.loads(html)['cmts']
    comments = []
    for item in data:
        comment = {
            'id': item['id'],
            'nickName': item['nickName'],
            'cityName': item.get('cityName', ''),
            'content': item['content'].replace('\n', ' ', 10),
            'score': item['score'],
            'startTime': item['startTime']
        }
        comments.append(comment)
    return comments

All comments are saved to a plain‑text file 中国医生.txt for later analysis:

# 存储数据，存储到文本文件
def save_to_data():
    start_time = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
    end_time = '2021-07-09 00:00:00'
    while start_time > end_time:
        url = 'https://m.maoyan.com/mmdb/comments/movie/1337700.json?_v_=yes&offset=0&startTime=' + start_time.replace(' ', '%20')
        html = None
        try:
            html = get_data(url)
        except Exception as e:
            time.sleep(0.5)
            html = get_data(url)
        else:
            time.sleep(0.1)
        comments = parse_data(html)
        for item in comments:
            with open('中国医生.txt', 'a', encoding='utf-8') as f:
                f.write(str(item['id']) + ',' + item['nickName'] + ',' + item['cityName'] + ',' + item['content'] + ',' + str(item['score']) + ',' + item['startTime'] + '\n')
        start_time = comments[14]['startTime']
        start_time = datetime.strptime(start_time, '%Y-%m-%d %H:%M:%S') - timedelta(seconds=1)
        start_time = datetime.strftime(start_time, '%Y-%m-%d %H:%M:%S')

Once the data file is ready, the script performs several analyses:

Rating distribution: the score field is read, grouped into five star levels, and visualized with a pie chart using pyecharts .

Geographic distribution: the cityName field is counted, the top 25 cities are selected, and a map of China is rendered.

Top‑viewer ranking: the same city counts are displayed as a bar chart.

Word cloud: all comment texts are concatenated, segmented with jieba , filtered by a custom stop‑word list, and rendered with wordcloud on a background image.

Example code for the rating pie chart:

pie = (
    Pie()
    .add("", [list(z) for z in zip(attr, value)])
    .set_global_opts(title_opts=opts.TitleOpts(title="《中国医生》评分比例饼图"))
    .set_series_opts(label_opts=opts.LabelOpts(formatter="{b}:{d}%"))
)
pie.render('评分.html')

Example code for the city distribution map:

def geo_base():
    c = (
        Geo()
        .add_schema(maptype="china")
        .add("", [list(x) for x in data])
        .set_series_opts(label_opts=opts.LabelOpts(is_show=False))
        .set_global_opts(
            visualmap_opts=opts.VisualMapOpts(is_piecewise=True, min_=min2, max_=max2),
            title_opts=opts.TitleOpts(title="《中国医生》观影用户分布", subtitle="数据来源：猫眼 -- Dragon少年", pos_left="300px")
        )
    )
    return c

Word‑cloud generation snippet:

# 设置分词
comment_after_split = jieba.cut(str(comments), cut_all=False)
words = ' '.join(comment_after_split)
# 设置屏蔽词
stopwords = STOPWORDS.copy()
for w in ['电影','我','我们','的','是','了','没有','什么','有点','不是','真的','感觉','觉得','还是','但是']:
    stopwords.add(w)
# 生成词云
wc = WordCloud(width=1024, height=768, background_color='white', mask=bg_image,
               font_path='STKAITI.TTF', stopwords=stopwords, max_font_size=400, random_state=50)
wc.generate_from_text(words)
wc.to_file('词云图.jpg')

Finally, the article shows the resulting visualizations – a pie chart where five‑star reviews account for about 89.5%, a bar chart of the top 25 cities, a geographic heat map of viewer distribution, and a word cloud highlighting frequent terms in the comments. The analysis concludes that the film enjoys a strong positive reception and that the most active viewing cities correspond to regions with higher GDP.