PTNDataset

Telegram News Persian Dataset Overview This dataset contains structured data collected from public Telegram channels related to Iranian political news and information. It includes channel-level metadata and post-level content, with optional media references, collected for academic research purposes. The dataset accompanies a research article currently under review and is released to support transparency, reproducibility, and further research on political communication and information diffusion on Telegram. All data were collected from public Telegram channels only. Dataset Structure The dataset consists of two JSON files: 1. `channels.json` channel-level metadata 2. `posts.json` post-level content, grouped by channel Channels File (channels.json) The channels file is a list of channel metadata objects, one per channel. Example: [ { "id": <channel id>, "title": <channel title>, "url": <channel url>, "date": <channel creation time>, "about": <channel description>, "subscribers": <channel number of subscribers>, "source": <TGstat or recommended>, "verified": <0 or 1>, "restricted": <0 or 1>, "fake": <0 or 1> }, ... ] Field description Field Description id Unique channel identifier assigned by Telegram title Channel name at the time of collection url Channel username as in `t.me/<url>` date Channel creation date (Unix timestamp, milliseconds) about Channel description subscribers Number of subscribers at time of collection source Channel source (`TGstat` or `recommended`) verified Whether the channel is verified by Telegram restricted Whether the channel is restricted channel_type specifies the nature of the channel: including instant news services, official domestic (Iran-based) news channel, official international news channel, or political Posts File (posts.json) The posts file is a dictionary indexed by channel URL.Each key corresponds to one channel, and its value is a list of post objects. Example { "channel_x_url": [ { "id": "12345", "uid": "1", "channel_id": 10012345, "date": "2025-07-15T00:00:00+00:00", "text": "...", "views": "1000", "forwards": "100", "reactions": [ { "emoji": "❤️", "count": "1000", } ], "comments": [], "forwarded_from": "10067890", "origin_id" : "67890" "media_type": "photo", "has_media": true, "photo_id": "1234567890123456789", "size": "100000" }, ... ] } ``` Field description Field Description id Post ID assigned by Telegram (unique per channel) uid Dataset-wide unique post identifier (assigned incrementally) channel_id Telegram channel ID date Timestamp of publication text Text of the post views Number of views forwards Number of forwards reactions List of emoji reactions with counts comments List of comments forwarded_from Channel ID of the origin post origin_id Telegram Id of the origin post media_id Media ID given by Telegram upon upload media_type Type of attached media (`photo`, `video`, `document`, or `null`) size Media size in bytes Data Types and Encoding - All integer values are stored as strings to avoid overflow and parsing inconsistencies across platforms.- Text fields may contain Persian (Farsi) and Arabic characters.- Files are encoded in UTF-8.- JSON format is preferred over CSV to preserve text integrity and nested structures. Ethical Considerations - Only public Telegram channels were collected.- No private messages, private groups, or personal accounts are included.- Usernames of commenters were pseudonimized.- The dataset complies with common ethical guidelines for social media research. Intended Use This dataset is intended for:- Computational social science- Political communication research- Network analysis- Content and discourse analysis- Media diffusion studies Contact For questions or clarifications, please contact:- Maintainer: Amirsiavosh Bashardoust or Caroline Violot- Affiliation: University of Lausanne- Email: [e-mail] or [e-mail]