Generative adversarial network for sentiment‐based stock prediction

Abstract

Financial markets received more attention due to technological advancements, such as Artificial Intelligence (AI). In addition to the price index, traders and investors constantly monitor stock news on social media. Therefore, predicting the market by analyzing public opinions is an important issue. In this research, we propose three models based on Generative Adversarial Network (GAN), namely Price-GAN, Price-Sentiment-GAN, and Price-Sentiment-WGAN. The first model uses only optimized price features, and the two other models use sentiment features collected from social media as well as optimized price features. All the proposed GAN models include Long Short-Term Memory (LSTM) as generators and Convolution Neural Networks (CNN) as discriminators. To evaluate the proposed models, two different social media datasets in English and Persian are used. Our proposed models predict the close stock price for 15 English and 5 Persian stocks. All of the proposed GAN models outperform the state-of-the-art models by enhancing the performance of the English dataset by 2.44% and the Persian dataset by 12.11%.

Improved Generative Adversarial Network Stock Price Prediction Method

“…Liu YL, et al [7] addressed imbalanced text token distribution issues in convolutional neural network sentiment analysis for stock price prediction through a sentiment analysis and GAN-based approach. Furthermore, Sonkiya P, et al [8] proposed an ensemble approach incorporating sentiment analysis and GANs to predict stock prices and Asgarian S, et al [9] introduced three GAN-based models for market trend prediction by analyzing public sentiment. Polamuri SR, et al [10] innovatively applied reinforcement learning and Bayesian optimization, overcoming hyperparameter challenges in a hybrid prediction algorithm based on Generative Adversarial Networks (GAN-HPA).…”

“…As articulated in Zhao T, et al [1], nonlinear prediction methods encompass approaches grounded in BP neural networks, support vector machines, recurrent neural networks, generative adversarial networks, and reinforcement learning. Among these, generative adversarial network-based prediction methods [2][3][4][5][6][7][8][9][10], trained through adversarial learning, stand out for their adaptability to the nonlinearity, instability, and complexity of the stock market. They generate data samples aligning more closely with actual market conditions, offering valuable support for stock prediction.…”

The Intersection of Ai and Quantum Computing in Financial Markets: A Critical Review

“…Additionally, the evolution of AI algorithms and models in financial applications is a key area of interest. Research has shown that AI, particularly deep learning and machine learning, has been extensively utilized for sentiment-based stock prediction and time series data analysis in financial markets (Asgarian et al, 2022;Biju et al, 2023). Furthermore, the integration of AI and quantum computing in finance is an emerging trend that is gaining traction.…”