CognitiveSky: Scalable Sentiment and Narrative Analysis for Decentralized Social Media
Gaurab Chhetri, Anandi Dutta, Subasish Das
2025-09-16
Summary
This paper introduces CognitiveSky, a new system for automatically understanding what people are saying and feeling on Bluesky, a social media platform similar to Twitter. It's designed to analyze large amounts of text data from Bluesky to track things like overall sentiment, specific emotions, and the main topics being discussed.
What's the problem?
Analyzing what's happening on social media in real-time is really useful for understanding public opinion, spotting potential crises, or even tracking the spread of misinformation. However, newer, decentralized social media platforms like Bluesky present a challenge because they aren't controlled by a single company, making it harder to collect and analyze data. Existing tools often don't work well with these new platforms or are too expensive to use.
What's the solution?
The researchers built CognitiveSky, which connects directly to Bluesky using its API to collect posts. Then, it uses powerful AI models, specifically 'transformer-based models,' to automatically label each post with information about the emotions expressed, the overall sentiment (positive, negative, neutral), and the main topics being discussed. All this data is then displayed on a dashboard so you can easily see trends and patterns. Importantly, they built it using free online services to keep costs down and make it accessible to anyone.
Why it matters?
CognitiveSky is important because it provides a free and open way to study what's happening on decentralized social media. This is crucial as more people move to these platforms. It's not just limited to Bluesky either; the system is designed to be flexible and could be used to analyze conversations about mental health, identify fake news, respond to emergencies, or understand public opinion on important issues. It helps researchers and others understand the changing landscape of online communication.
Abstract
The emergence of decentralized social media platforms presents new opportunities and challenges for real-time analysis of public discourse. This study introduces CognitiveSky, an open-source and scalable framework designed for sentiment, emotion, and narrative analysis on Bluesky, a federated Twitter or X.com alternative. By ingesting data through Bluesky's Application Programming Interface (API), CognitiveSky applies transformer-based models to annotate large-scale user-generated content and produces structured and analyzable outputs. These summaries drive a dynamic dashboard that visualizes evolving patterns in emotion, activity, and conversation topics. Built entirely on free-tier infrastructure, CognitiveSky achieves both low operational cost and high accessibility. While demonstrated here for monitoring mental health discourse, its modular design enables applications across domains such as disinformation detection, crisis response, and civic sentiment analysis. By bridging large language models with decentralized networks, CognitiveSky offers a transparent, extensible tool for computational social science in an era of shifting digital ecosystems.