Feature Selection using Swarm Intelligence Techniques
Final MSc Project (Dissertation) in Data Science, University of Greenwich — Supervised by Dr Mohammed Majid Al-Rifaie and Dr Hooman Oroojeni
Overview
High-dimensional datasets often contain redundant or irrelevant features that can negatively impact classification accuracy. This project aimed to reduce dimensionality and enhance classification performance using advanced optimization techniques.
- Implemented PSO and DFO algorithms to select relevant features.
- Reduced dimensionality of datasets while enhancing model performance.
- Evaluated selected features on classification tasks to demonstrate accuracy improvements.
Technologies Used
Python
NumPy
Pandas
Jupyter Notebook
Scikit-Learn
Resources
- The Dispersive Flies Optimisation (DFO) Algorithm - https://github.com/mohmaj/DFO
- Al-Rifaie, M. M. (2014). Dispersive flies optimisation. In 2014 Federated Conference on Computer Science and Information Systems (pp. 529-538). IEEE