An AI Trading Bot to Increase Investment Portfolio Performance

Esparza Giacomelli, Rodrigo (2025) An AI Trading Bot to Increase Investment Portfolio Performance. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (584kB) | Preview

Abstract

The volatility in the stock market, understood as the unpredictable variation of prices, makes it significantly harder to make accurate investment decisions. However, different AI techniques have started to be used to detect patterns in the market and provide a stronger base for making investment decisions. Still, the wide variety of ways these tools can be applied to decide the best times to buy or sell makes their practical use a big challenge. This work proposes different neural network implementations to increase portfolio investment through trading. The neural networks that will be implemented will use the DQN reinforcement learning algorithm. In these networks, different financial strategies and architectures will be tested to find which model gives the best performance with more stable and secure behaviour. After that, the best DQN model will be combined with an NLP pipeline that analyses international news and predicts its impact on the stock market, creating a final neural network that uses both quantitative and qualitative data for trading. The results will show which neural network architecture, training method, exploration parameters, and financial strategy are the best to use for trading, and how much the model can change if international news is taken into account or if only the numerical history of the stock is considered.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Simiscuka, Anderson UNSPECIFIED
Uncontrolled Keywords:	Stock Market; Reinforcement Learning; DQN; Deep Learning; NLP; Volatility
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HG Finance > Investment Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning H Social Sciences > HG Finance > Investment > Stock Exchange
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	28 May 2026 13:44
Last Modified:	28 May 2026 14:40
URI:	https://norma.ncirl.ie/id/eprint/9318

Actions (login required)

View Item