Predicting Stock Price Movement with LLM-Enhanced Tweet Emotion Analysis

2025-10-03

AI Summary

Short-term stock price prediction remains highly challenging due to market volatility and sensitivity to investor sentiment. To address this, we propose an LSTM-based forecasting framework that jointly models historical price sequences and LLM-enhanced sentiment features to identify significant next-day price movements. Our method innovatively employs Llama 3.1-8B-Instruct for semantic reconstruction and noise filtering of financial tweets, thereby improving sentiment representation fidelity. Multi-dimensional sentiment features are then extracted via a fine-tuned DistilRoBERTa model combined with the NRC Emotion Lexicon, and fed into an LSTM to capture temporal dynamics. Experiments on TSLA, AAPL, and AMZN demonstrate that our approach boosts sentiment classification accuracy from 23.6% to 38.5%, substantially outperforming a price-only baseline. These results validate the efficacy of LLM-driven fine-grained sentiment features in enhancing short-horizon financial forecasting.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Text Classification & Sentiment Analysis
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Accurately predicting short-term stock price movement remains a challenging task due to the market's inherent volatility and sensitivity to investor sentiment. This paper discusses a deep learning framework that integrates emotion features extracted from tweet data with historical stock price information to forecast significant price changes on the following day. We utilize Meta's Llama 3.1-8B-Instruct model to preprocess tweet data, thereby enhancing the quality of emotion features derived from three emotion analysis approaches: a transformer-based DistilRoBERTa classifier from the Hugging Face library and two lexicon-based methods using National Research Council Canada (NRC) resources. These features are combined with previous-day stock price data to train a Long Short-Term Memory (LSTM) model. Experimental results on TSLA, AAPL, and AMZN stocks show that all three emotion analysis methods improve the average accuracy for predicting significant price movements, compared to the baseline model using only historical stock prices, which yields an accuracy of 13.5%. The DistilRoBERTa-based stock prediction model achieves the best performance, with accuracy rising from 23.6% to 38.5% when using LLaMA-enhanced emotion analysis. These results demonstrate that using large language models to preprocess tweet content enhances the effectiveness of emotion analysis which in turn improves the accuracy of predicting significant stock price movements.

Problem

• Predicting short-term stock price movements using tweet emotion analysis
• Integrating emotion features with historical stock data via LSTM model
• Enhancing emotion analysis quality through LLM preprocessing of tweets

Innovation

• Uses LLaMA model for preprocessing tweet data
• Combines emotion features with historical stock prices
• Trains LSTM model for stock movement prediction