Create Stunning Word Clouds from WeChat Moments with Python: Step‑by‑Step Guide

Previously we used a Python web crawler to collect data from WeChat Moments; this article shows how to visualize that data with word clouds.

First, install the required libraries wordcloud and jieba via pip install wordcloud jieba.

You also need a Chinese font such as simhei.ttf; download it from the internet.

Create a file analyse.py next to items.py and define an analyse_words function. Use jieba.cut to obtain word segmentation and load the JSON data from moment.json for processing.

Note: when opening moment.json, specify encoding='utf-8' to avoid GBK encoding errors.

Run the program; it generates keys.png in the items.py directory.

Open keys.png to view the basic word cloud.

The initial cloud is functional but visually unappealing; we improve it using the WordArt online tool.

For example, applying an animal-shaped template yields a more attractive result.

To avoid clutter, preprocess the JSON: extract all text into moment.txt.

Write code to segment the text; the implementation is shown below.

Running the script produces moment_outputs.txt, which clearly displays the segmentation results.

Next, write code to count word frequencies; the snippet is illustrated below.

After execution, a text file and an Excel sheet containing the frequency statistics are generated.

Import these keywords into WordArt to create visually appealing word clouds.

Adjust shape, font, layout, and colors to produce vibrant clouds, such as a dog-shaped word cloud.

Another example is a cloud-shaped word cloud that offers a visual feast.