hreyulog/weibo-opinion-dynamic-single-dim

Weibo Sentiment Evolution Dataset

This dataset contains Weibo posts and their associated comment threads used for studying sentiment evolution and opinion dynamics in social media discussions.

The dataset is distributed as a single JSON Lines file:

• weibo_dataset.jsonl

Each line is one Weibo post record. Comments for that post are embedded in the comments field.

Dataset Details

• Number of post records: 1,379
• Number of embedded comments: 93,569
• Number of Weibo users/accounts: 7
• Language: Chinese
• Format: JSON Lines (.jsonl)

The final dataset was built using *output.json files as the source of truth. Posts or users without a corresponding *output.json entry were excluded.

Data Fields

Each row contains:

Field	Type	Description
user_id	string	Weibo user/account identifier from the source file name.
id	string	Weibo post identifier.
content	string	Text content of the Weibo post.
content_score	float	Sentiment-related score for the post content.
time	string	Post publication time.
likes	string	Number of likes for the post as stored in the source data.
comment_score	list	Sentiment-related score and like-count pairs for comments.
comments	list	Embedded comment records for the post.
comment_count_merged	integer	Number of embedded comments attached to this post.

Each item in comments contains:

Field	Type	Description
id	integer	Comment author/user identifier as stored in the source data.
comment	string	Comment text.
time	string	Comment publication time.
likes	integer	Number of likes for the comment.

Loading

Using datasets:

from datasets import load_dataset

dataset = load_dataset("json", data_files="weibo_dataset.jsonl", split="train")
print(dataset[0])

Using plain Python:

import json

with open("weibo_dataset.jsonl", "r", encoding="utf-8") as f:
    for line in f:
        record = json.loads(line)
        print(record["id"], record["content"], len(record["comments"]))
        break

Intended Use

This dataset is intended for research on:

• sentiment evolution in online discussions
• opinion dynamics modeling
• comment-thread analysis
• Chinese social media text analysis

Citation

If you use this dataset, please cite:

@misc{he2025opiniondynamicsmodelssentiment,
      title={Opinion Dynamics Models for Sentiment Evolution in Weibo Blogs}, 
      author={Yulong He and Anton V. Proskurnikov and Artem Sedakov},
      year={2025},
      eprint={2511.15303},
      archivePrefix={arXiv},
      primaryClass={cs.SI},
      url={https://arxiv.org/abs/2511.15303}, 
}

Notes

The dataset contains social media text from Weibo. Users should handle the data responsibly and use it only for research purposes consistent with applicable platform policies, privacy expectations, and local regulations.