UX & E-Learning
What Drives Continuance Intention in a Self-directed Specialised Technical E-Learning Platform
Asks what keeps people coming back to musicinformationretrieval.com, an open source Music Information Retrieval platform with no enrolment, no instructor and no credential. Extends the Expectation Confirmation Model with knowledge management constructs, then tests it twice. First with PLS-SEM on survey data. Second with an NLP-assisted validation procedure that embeds users' own accounts of why they return, clusters them without reference to the model, and reads the themes back against the structural result.
UX Research
E-Learning
PLS-SEM
Sentence Embeddings
2026 · Queen Mary, University of London · MSc Project
Machine Learning & Music
Classifying EDM Subgenres Using Spotify Audio Features
Investigates whether Spotify audio features can distinguish between closely related EDM subgenres including drum-and-bass, dubstep, garage, house, techno, and trance. Compares multinomial logistic regression and Support Vector Classification using stratified 10-fold cross-validation. The SVC model outperformed logistic regression with 65.5% accuracy, demonstrating that non-linear classifiers better capture the subtle differences between EDM genres.
Machine Learning
Music Classification
Spotify
SVC
2026 · Queen Mary, University of London
Marketing & AI
Redefining Marketing in an AI-Driven World
Explores how generative AI tools like ChatGPT and Microsoft Copilot affect the creativity, strategic decision-making, and job satisfaction of marketing professionals, and the barriers that prevent effective adoption. Based on a qualitative case study of an international technology company, using semi-structured interviews with eight marketers.
Generative AI
Marketing
Job Satisfaction
Creativity
2024/25 · University of Birmingham · Dissertation
