Efficient Event Modeling and Multimodal Integration for Stock Price Prediction

Tian, Guangyang

Efficient Event Modeling and Multimodal Integration for Stock Price Prediction

Tian, Guangyang

Permalink

Publication Type:: Thesis
Issue Date:: 2026

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (9.13 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Tian, Guangyang
dc.date.accessioned	2026-10-27T23:43:53Z
dc.date.available	2026-10-27T23:43:53Z
dc.date.issued	2026
dc.identifier.uri	http://hdl.handle.net/10453/197275
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	Stock market forecasting remains challenging because financial markets are nonlinear, non-stationary, and strongly influenced by heterogeneous information sources. Historical prices alone cannot fully explain market movements, as financial news, investor sentiment, and online discussions often shape expectations and trigger price fluctuations. This thesis investigates efficient event modeling and multimodal integration for stock price prediction, with the aim of improving forecasting accuracy, robustness, and contextual understanding. To address these challenges, this thesis develops a series of forecasting frameworks. It first proposes a hybrid deep learning model to capture complementary temporal dependencies in financial time series. It then introduces a regime-aware learning strategy with volatility-weighted training to improve robustness under changing market conditions. The thesis further develops a multimodal framework for modeling interactions between financial news and market data, and uses large language models to extract structured features from social media discussions. Experimental results show that the proposed methods improve predictive performance and provide practical value for financial forecasting.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2026 Guangyang Tian
dc.rights	au.edu.uts.lib/nph
dc.title	Efficient Event Modeling and Multimodal Integration for Stock Price Prediction	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

Stock market forecasting remains challenging because financial markets are nonlinear, non-stationary, and strongly influenced by heterogeneous information sources. Historical prices alone cannot fully explain market movements, as financial news, investor sentiment, and online discussions often shape expectations and trigger price fluctuations. This thesis investigates efficient event modeling and multimodal integration for stock price prediction, with the aim of improving forecasting accuracy, robustness, and contextual understanding. To address these challenges, this thesis develops a series of forecasting frameworks. It first proposes a hybrid deep learning model to capture complementary temporal dependencies in financial time series. It then introduces a regime-aware learning strategy with volatility-weighted training to improve robustness under changing market conditions. The thesis further develops a multimodal framework for modeling interactions between financial news and market data, and uses large language models to extract structured features from social media discussions. Experimental results show that the proposed methods improve predictive performance and provide practical value for financial forecasting.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/197275