A Sentiment Based LSTM for Stock Prediction
Faculty Mentor
Dr. Hayden Wimmer, Atef Shalan
Location
Poster 214
Session Format
Poster Presentation
Academic Unit
Department of Information Technology
Background
Sentiment Analysis is the utilization of unstructured text-data to numerically estimate how a group of persons feels about a certain item/topic. Sentiment analysis can come from many different media selections such as social media, news reports, and speeches. Sentiment Score is the numerical data that is associated with the text. A positive sentiment means that a product is well received by the public; whereas, a negative sentiment means that the product was not received well.
Keywords
Allen E. Paulson College of Engineering and Computing Student Research Symposium, Long Short-term Memory, LSTM, Recurrent Neural Network, RNN, Relative Strength Index, RSI
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Presentation Type and Release Option
Presentation (File Not Available for Download)
Start Date
2022 12:00 AM
January 2022
A Sentiment Based LSTM for Stock Prediction
Poster 214
Stock market prediction has been challenging researchers for decades; however, new technology and methods becoming available are making a case that it could be done. These new technologies and methods provide a hopeful outlook for investors and researchers alike. One of the more recent statistical approaches to stock market prediction is Time Series Forecasting. This approach is used to make future predictions based on historical data. There are different Neural Networks designed for Time Series Forecasting; however, one form of neural network has been becoming more and more popular, Long Short-term Memory. Long Short-term Memory, or LSTM, is a Recurrent Neural Network (RNN) architecture that has an emphasis on feedback connections. The differentiation in this LSTM comes from one of the inputs, which is sentiment analysis.
