Projects
A selection of technical projects involving machine learning, dashboard analytics, and predictive modelling. Each card includes a short description, a skills snapshot, and a link to explore the full work.
Discogs Recommendation Model
A personalised recommendation system built using the Discogs API to analyse my collection and wantlist, then branch out through connected labels and artists. The model includes a desirability and rarity score that boosts high-want/low-have records while down-weighting over-owned ones. The result is a personalised engine for surfacing genuinely obscure (and hopefully cheap!) dance records I’m likely to enjoy.
- Python (Pandas, Numpy, Scikit-learn, SciPy)
- JSON API integration and rate-limited ETL pipeline design
- Content-based recommender modelling (TF-IDF, cosine similarity)
- Custom scoring logic for weighting and feature scaling
- LLM prompt engineering for quick iteration and scaling
Apple Music Listening Analytics Dashboard
A Tableau Public dashboard visualising listening patterns across songs, artists, albums, and genres. Includes parameters, ranking logic, calculated fields, and an integrated UX-focused layout for dynamic analysis – happy clicking!
- Tableau (calculated fields, parameters, LODs, dynamic sorting and grouping)
- Script Editor and Python-based ETL
- Time-based filtering
- Dashboard UX design
- Interactive storytelling
Term Deposit Subscription Prediction Model
A supervised machine learning classifier predicting customer conversion for term deposit campaigns. Includes preprocessing, feature engineering, model training, and evaluation (ROC-AUC, CV).
- Python (Pandas, Numpy, Seaborn, Scikit-learn)
- Feature engineering
- Model evaluation (ROC-AUC, cross-validation)
- Performance comparison
- Jupyter notebook workflow