Analyzing and Predicting the Box Office of "The Battle at Lake Changjin" Using Python Data Scraping and Visualization

The article begins by describing the popularity of the movie "The Battle at Lake Changjin" and introduces a data‑driven analysis using Python to scrape comments from the Maoyan platform.

Data acquisition : The comments are fetched by constructing the Maoyan API request and the JSON response is saved locally.

df_new = df.drop_duplicates(['comment_id'])

Data storage : A function is written to write the retrieved JSON data to a local file.

Data cleaning : Duplicate entries are removed based on comment_id , and non‑Chinese characters are filtered out from the comment text.

Visualization analysis includes:

Removing duplicate comments.

Ranking comments by number of likes.

City‑wise comment distribution, showing a correlation with regional economic development.

Gender distribution, revealing a slight female majority.

Watch status, indicating most users comment after watching.

Rating distribution, converting the 10‑point UI scale to a 5‑point internal scale.

User level distribution, where most users are level 2.

Frequency of mentions of the film’s creators, with 易烊千玺 ranking highest.

All visualizations are generated with pyecharts and displayed as bar, pie, and line charts.

Box‑office prediction : The daily box‑office data is plotted, showing a rise until day 7 and a decline thereafter. A linear regression model from sklearn.linear_model is fitted to the data.

attr = ['其他','男','女'] b = (Pie() .add(...))

The fitted model is then used to forecast the next day's box‑office, and the prediction result is presented as an image.

Overall, the article provides a step‑by‑step guide for data scraping, cleaning, visualization, and simple predictive modeling applied to movie comment data.