This tutorial demonstrates how to use Selenium to crawl short comments from a Douban movie page, extract them via XPath, paginate through results, and finally create a visual word cloud using Python's wordcloud library.
Using Python, the author collected and visualized Douban and Weibo comments on the film “Chinese Doctor,” showing rating distributions, keyword word clouds, and fan sentiment, and compares the 2021 movie with the 2019 documentary to illustrate audience perception of the pandemic drama.
This article demonstrates how to crawl Douban short reviews for the TV show "Actors Please Take Your Place" Season 2, clean and deduplicate the data, apply Baidu's SKEP sentiment model, and visualize word clouds, rating distributions, posting times, and sentiment scores, providing full Python code for replication.
This tutorial walks through building a Python web scraper that fetches Douban TV show titles, ratings, detail links, and images, parses JSON responses, handles anti‑scraping measures, and stores the results in a CSV file with step‑by‑step code examples.
This tutorial shows how to log in to Douban with Python, scrape short comments for the animated film “Nezha”, process the text using jieba for Chinese word segmentation, and generate a visual word cloud that highlights the most frequent audience sentiments.
This article demonstrates how to scrape the full 2018 Chinese variety‑show list from Douban using Python Selenium and BeautifulSoup, compile detailed metadata and actor information into Excel, and then analyze popularity rankings, rating distributions, frequent celebrity appearances, and common negative feedback.
The article walks through a full Python workflow that scrapes up to 500 Douban movie reviews for "Dying to Survive" and "Hidden Blade," cleans and stores them in pandas, performs SnowNLP sentiment analysis, and visualizes city distribution, rating trends, and word clouds with pyecharts.
This article scrapes over 22,000 Douban comments for the TV series “The Hunt,” cleans the data, applies SnowNLP‑based sentiment classification and word‑cloud visualization, and shows that while overall ratings are mixed, detailed short reviews are overwhelmingly positive.
This tutorial walks through using Python 3.5 to fetch the latest movies from Douban, extract their IDs and titles, crawl user comments, clean the text with regular expressions, segment Chinese words using Jieba, remove stopwords, compute word frequencies, and finally generate a word‑cloud visualization of the reviews.
The article details Douban’s platform architecture, covering online services, the BeansDB key‑value store, the DAE PaaS layer, offline big‑data processing with MooseFS and DPark, and the team’s engineering practices that keep the system reliable and extensible.
