Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

TeleScope: A Longitudinal Dataset for Investigating Online Discourse and Information Interaction on Telegram

Susmita Gangopadhyay, Danilo Dessi, Dimitar Dimitrov, Stefan Dietze

TL;DR

TeleScope addresses the need for a large-scale, longitudinal, multilingual Telegram data resource by building a seed-driven, snowball-expanded dataset of ~534K channels (with ~71K fully downloaded) containing ~120M messages, enhanced with language, temporal, and entity enrichments, plus new cross-channel propagation and interaction graphs. The approach enables replication of social-media methodologies on Telegram, supports network and community discovery, and facilitates entity-based search and low-resource language research through rich forward-flow and engagement data. Key contributions include the largest publicly available Telegram dataset to date, forward-propagation flows, a channel-to-channel graph, and FAIR data release with reproducible tooling. TeleScope’s data enable scalable investigations into information diffusion, discourse dynamics, and multilingual communication patterns with practical impact for social science, NLP, and platform studies.

Abstract

Telegram is a globally popular instant messaging platform known for its strong emphasis on security, privacy, and unique social networking features. It has recently emerged as the host for various cross-domain analysis and research works, such as social media influence, propaganda studies, and extremism. This paper introduces TeleScope, an extensive dataset suite that, to our knowledge, is the largest of its kind. It comprises metadata for about 500K Telegram channels and downloaded message metadata for about 71K public channels, accounting for around 120M crawled messages. We also release channel connections and user interaction data built using Telegram's message-forwarding feature to study multiple use cases, such as information spread and message forwarding patterns. In addition, we provide data enrichments, such as language detection, active message posting periods for each channel, and Telegram entities extracted from messages, that enable online discourse analysis beyond what is possible with the original data alone. The dataset is designed for diverse applications, independent of specific research objectives, and sufficiently versatile to facilitate the replication of social media studies comparable to those conducted on platforms like X (formerly Twitter)

TeleScope: A Longitudinal Dataset for Investigating Online Discourse and Information Interaction on Telegram

TL;DR

TeleScope addresses the need for a large-scale, longitudinal, multilingual Telegram data resource by building a seed-driven, snowball-expanded dataset of ~534K channels (with ~71K fully downloaded) containing ~120M messages, enhanced with language, temporal, and entity enrichments, plus new cross-channel propagation and interaction graphs. The approach enables replication of social-media methodologies on Telegram, supports network and community discovery, and facilitates entity-based search and low-resource language research through rich forward-flow and engagement data. Key contributions include the largest publicly available Telegram dataset to date, forward-propagation flows, a channel-to-channel graph, and FAIR data release with reproducible tooling. TeleScope’s data enable scalable investigations into information diffusion, discourse dynamics, and multilingual communication patterns with practical impact for social science, NLP, and platform studies.

Abstract

Telegram is a globally popular instant messaging platform known for its strong emphasis on security, privacy, and unique social networking features. It has recently emerged as the host for various cross-domain analysis and research works, such as social media influence, propaganda studies, and extremism. This paper introduces TeleScope, an extensive dataset suite that, to our knowledge, is the largest of its kind. It comprises metadata for about 500K Telegram channels and downloaded message metadata for about 71K public channels, accounting for around 120M crawled messages. We also release channel connections and user interaction data built using Telegram's message-forwarding feature to study multiple use cases, such as information spread and message forwarding patterns. In addition, we provide data enrichments, such as language detection, active message posting periods for each channel, and Telegram entities extracted from messages, that enable online discourse analysis beyond what is possible with the original data alone. The dataset is designed for diverse applications, independent of specific research objectives, and sufficiently versatile to facilitate the replication of social media studies comparable to those conducted on platforms like X (formerly Twitter)
Paper Structure (31 sections, 9 figures, 4 tables)

This paper contains 31 sections, 9 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Data Collection Methodology
  4. Seed List Channel Collection
  5. Data Collection Process and Statistics
  6. Channel Discovery and Data Coverage
  7. Dataset Description
  8. Enriched Channel Metadata
  9. Channel Metadata.
  10. Channel Language.
  11. Temporal Channel Metadata.
  12. Enriched Crawl of Telegram Message Metadata
  13. Message Metadata.
  14. Telegram Entities.
  15. User Interaction Data and Cross-Channel Message Propagation Flows
  16. ...and 16 more sections

Figures (9)

  • Figure 1: Overlap of discovered channels across different seed selection criteria.
  • Figure 2: Depicts the data collection pipeline. Channels collected using three criteria are merged into a seedlist, enriched via the Telethon API, expanded through snowballing, and stored in a database.
  • Figure 3: Channels discovered via snowball sampling over time.
  • Figure 4: An example of message forwarding flows in TeleScope.
  • Figure 5: Example of channel-to-channel graph.
  • ...and 4 more figures