| Author: | Li, Dongwei |
| Title: | On machine learning and deep learning to financial data analysis and applications |
| Advisors: | You, Jane (COMP) |
| Degree: | Eng.D. |
| Year: | 2023 |
| Subject: | Finance -- Data processing Machine learning Hong Kong Polytechnic University -- Dissertations |
| Department: | Faculty of Engineering |
| Pages: | xv, 116 pages : color illustrations |
| Language: | English |
| Abstract: | Machine learning is a branch of artificial intelligence that aims to build systems that can learn or achieve specific targets based on the given data. Machine learning algorithms can complete predictions and perform tasks based on the learned data features by learning patterns from input samples. Deep learning is a branch of machine learning that utilizes various neural networks with many layers and parameters to implement machine learning tasks better. At present, financial technology is a field that has received more attention and is participating in various businesses in the financial area by combining the latest machine learning and deep learning technologies. Fintech has played a vital role in the financial field. At present, a large number of financial practitioners are inseparable from it every day. At present, the application of financial technology is not thorough enough, and there are still many places in the industry where financial technology can help. Some financial technology applications can assist human beings, and some applications can even replace human beings. This research attempts to apply the methods of machine learning and deep learning to financial analysis, mainly researching financial data prediction and company classification and rating. First, the study tries to predict future data values based on past financial values. In the research, some econometrics methods will be tested and compared. At the same time, various regression methods in machine learning, including support vector regression method based on support vector machines, will be used and analyzed whether they are suitable for financial prediction. This research will also focus on neural networks, attempting with the input of financial statements, and trying to predict future data. The results showed that the single variable prediction results are not ideal, and the models cannot simply be used to make predictions. With a change to the original data and using multiple variables as input data, the performances are good enough for real world use. This research will also use machine learning and deep learning technology to rate and classify the company's financial data and try to use the K-means algorithm and deep learning model to establish an automated company rating classification system. The models generated good classification results, which can help financial analysts in rating activities. In the research, a prototype system was built, integrating some of the models used in the study. The model is used in the bank's internal systems and has undergone some internal testing by the bank. The results show that the prototype system can play a positive role in the bank's business. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 6916.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 4.33 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/200/12467