Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 56099 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Soleymani Baghshah, Mahdieh; Beigy, Hamid
- Abstract:
- Forecasting and analysing financial markets has always been an interesting research topic for fields ranging from financial sciences to mathematics and statistics. With the rapid development of artificial intelligence in the recent years, there has been a growing interest in using deep neural networks to predict market future trends. The price in these markets is determined by mechanisms of demand and supply. When there is a tendancy to buy a stock, there will be an increase in demand resulting a positive growth for price. On the other hand, when a large group of investors decide to sell their assets, market will experience an increase in supply and subsequently the prices drop. Availability of various data sources (such as news websites, media, books etc.) has added uncertainty to the way investors behave. Using different data modalities in deep neural networks can improve the results by training complex models that are close to reality. In this research, we have proposed a comprehensive model that includes price and textual data from social media. The extense of complex relations and numerous factors in a stock market makes graphs a suitable data structure for modeling such systems. These graphs can model financial, social and price relations that exist among companies. The remarkable performance of graph neural networks on tasks such as node classification has pushed us to use graph architectures in this project for means of predicting future price behaviour. The proposed method in this research models the market by using price and social network data in a multi-relation graph neural network. These multi-relation graphs are constructed in a way to not only captures the fundamental and structural realtions among caompanies, but also detects similar price and social media exposure patterns in the market. We will see that our model outperforms previous baselines on financial metrics, such as investment return rate, and we will examine the effects of constructed price and context graphs in our model
- Keywords:
- Deep Neural Networks ; Graph Neural Network ; Financial Market ; Financial Technologies ; Market Prediction
Digital Object List-
محتواي کتاب - view
Bookmark