Continuous Assessment for ECM3420 - Learning From Data, set by Dr. Chico Camargo, Dr. Diogo Pacheco and Dr. Marcos Oliveira (Year 3, Semester 1). Involves the use of machine learning methods, specifically a multi-layer perceptron (MLP), to explore which is the best predictor of the outcome of a UFC fight - the fighters' physical metrics or their historical data.
This work received a final mark of 75/100.
Please see specification.md
for specification. (Unfortunately, original specification does not exist; this is a replica.)
pandas
, numpy
and sklearn
are required to run src/physical-fp.py
and src/historical-fp.py
. These can be installed with:
pip install -r requirements.txt
Please run Python source files with
python physical-fp.py
and
python historical-fp.py
Results are printed to stdout
, and can be redirected to a file if you wish.
Please see doc/report.pdf
and doc/slides.pdf
for results. A YouTube video discussing the results is also available; please click here for the link.
This research makes use of a dataset that has not been included due to size limitations. The dataset can be accessed here. All credits go to their respective owners.