| Author: | Zhu, Yiming |
| Title: | Pre-trained knowledge assisted stock price prediction based on financial events |
| Advisors: | Li, Wenjie (COMP) |
| Degree: | M.Sc. |
| Year: | 2021 |
| Subject: | Stock price forecasting Machine learning Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Computing |
| Pages: | viii, 45 pages : color illustrations |
| Language: | English |
| Abstract: | Due to the complexity of the capital market, stock price prediction has always been a difficult problem for researchers. Researchers in the past tend to see it as a time series problem or a sentiment analysis problem based on social media text. However, changes in stock prices are closely related to events in the financial market. An important financial event can trigger a chain effect on the market. Financial events can be presented in financial news or reports by securities researchers. However, the background knowledge of the relevant financial entities in the event is lacking, and the amount of information contained in the news text or report text is limited. Therefore, in this paper, we propose an event-driven stock price prediction model that combines financial event text and event-related entity representation. The accuracy score of our model on the financial news data set and stock research report data set is better than the current state of art BERT model in the NLP field and beats a series of baselines. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 5870.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 1.52 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/11338