Unique Presentation Identifier:
66
Program Type
Undergraduate
Faculty Advisor
Rebecca Cunningham
Document Type
Presentation
Location
Face-to-face
Start Date
9-4-2026 9:00 AM
End Date
9-4-2026 9:30 AM
Abstract
Financial markets increasingly react to social media discourse, yet investors lack tools to translate this unstructured commentary into measurable indicators. Platforms such as YouTube host extensive discussions about publicly traded equities, but extracting reliable sentiment trends from high-volume, noisy comment streams remains technically challenging. This project develops a stock sentiment and market intelligence platform that transforms YouTube comment data into aggregated sentiment indicators aligned to specific equities. Comments are mapped to equities using ticker specific keyword identification combined with contextual filtering to reduce false associations from ambiguous or off-topic mentions. The system assigns numerical sentiment scores to individual comments and aggregates them across defined timeframes to evaluate directional market perception. The architecture consists of a Vue 3 and Vuetify frontend, a Django REST API backend, and a Supabase-hosted relational database. A local large language model (LLM) executed through Ollama serves as the reasoning layer. The backend manages ingestion, batching of LLM inference for efficiency, sentiment aggregation, and secure API exposure. The current implementation includes a watchlists system, data visualization module, and comment ingestion pipeline. Outputs are evaluated through structured manual sampling and consistency checks across repeated inference runs, along with qualitative comparison against observable short-term market trends to assess reliability and alignment with observable market trends. Key engineering challenges addressed include filtering off-topic comments and aligning temporal sentiment data with financial price movements. This platform demonstrates a scalable framework for converting unstructured social media discourse into structured market intelligence suitable for analytical and decision-support applications.
Recommended Citation
Thrower, Joshua; Pinkerton, Andrew; Duggan, Ian; and Lester, Wyatt, "LLM-Based Stock Sentiment and Market Intelligence Platform" (2026). ATU Scholars Symposium. 17.
https://orc.library.atu.edu/atu_rs/2026/2026/17
Included in
Artificial Intelligence and Robotics Commons, Databases and Information Systems Commons, Software Engineering Commons, Systems Architecture Commons
LLM-Based Stock Sentiment and Market Intelligence Platform
Face-to-face
Financial markets increasingly react to social media discourse, yet investors lack tools to translate this unstructured commentary into measurable indicators. Platforms such as YouTube host extensive discussions about publicly traded equities, but extracting reliable sentiment trends from high-volume, noisy comment streams remains technically challenging. This project develops a stock sentiment and market intelligence platform that transforms YouTube comment data into aggregated sentiment indicators aligned to specific equities. Comments are mapped to equities using ticker specific keyword identification combined with contextual filtering to reduce false associations from ambiguous or off-topic mentions. The system assigns numerical sentiment scores to individual comments and aggregates them across defined timeframes to evaluate directional market perception. The architecture consists of a Vue 3 and Vuetify frontend, a Django REST API backend, and a Supabase-hosted relational database. A local large language model (LLM) executed through Ollama serves as the reasoning layer. The backend manages ingestion, batching of LLM inference for efficiency, sentiment aggregation, and secure API exposure. The current implementation includes a watchlists system, data visualization module, and comment ingestion pipeline. Outputs are evaluated through structured manual sampling and consistency checks across repeated inference runs, along with qualitative comparison against observable short-term market trends to assess reliability and alignment with observable market trends. Key engineering challenges addressed include filtering off-topic comments and aligning temporal sentiment data with financial price movements. This platform demonstrates a scalable framework for converting unstructured social media discourse into structured market intelligence suitable for analytical and decision-support applications.