msc-dissertation

MSc Informatics dissertation project - University of Edinburgh: Curiosity in Multi-Agent Reinforcement Learning

Exoplanet Transit Detection using Deep Neural Networks written in Python

Bias-aware multimodal hate-speech detection - BERT + CNN + BiLSTM with structured features. Full MSc dissertation record, including the branches that failed.

An explainable multi-model NLP framework for post-earnings stock direction prediction. MSc Business Analytics capstone, Trinity College Dublin x EY.

Correlating Windows 11 forensic artefacts (Prefetch, Event Logs, ShellBags) to reconstruct user activity. MSc Cyber Security & Forensics dissertation, University of Westminster

A Hard Negative-Story Threshold: Training-Period Association, Temporal Non-Replication and Economic Limits. UCL MSc Computational Finance research using FinBERT, with Python tooling and reproducible evidence.

MSc dissertation: comparing five federated learning aggregation strategies for client and demographic fairness in chest X-ray classification (NIH ChestX-ray14).

A multimodal deep learning framework integrating MobileNet and BiLSTM architectures to synchronize facial landmark detection and audio feature extraction for robust emotion recognition.

MSc Dissertation | Dublin Property Price Analysis (2017-2025) using Python, Tableau and Geospatial Analytics

MSc dissertation project. Measures whether vision-language models preserve a flowchart's control-flow structure when generating code, separately from whether the code passes its tests. Four models, two datasets, plus the scoring tool and raw results.

MSc dissertation: from-scratch JPEG codec implementing ITU-T T.81 — DCT, quantization, Huffman coding, zig-zag ordering, LZW compression. Validated against libjpeg.

MSc dissertation evaluating targeted website fingerprinting on Tor Browser. Built a traffic collection pipeline and feature extraction framework, then trained Random Forest and SVM classifiers on encrypted Tor traffic to identify both visited websites and browser configuration characteristics, achieving up to 88% accuracy

MSc Business Analytics dissertation — DENUE-based market intelligence pipeline for Mexico

UCL MSc Information Security dissertation on APK-grounded agentic Android vulnerability discovery and restricted evidence-backed validation.

MSc Data Science dissertation investigating seismic sensing for traffic monitoring and environmental assessment in Greater Manchester.

MSc AI dissertation code for mammogram classification, ROI-guided attention, localisation, multimodal fusion, and anomaly detection.

MSc Dissertation - Data Science - Text Classification

Simulation framework for NHS emergency department triage optimization using a Mixture-of-Agents (MoA) architecture — built with SimPy, LangGraph, and FHIR-compliant synthetic data to reduce patient wait times and improve resource utilization.

MSc dissertation (KCL): re-measuring IoT intrusion detection under an attacker-machine holdout, and testing whether LLM-based packet prediction improves it. macro-F1 0.90 -> 0.60 once leakage is removed.