Inspiration
La reproducció no és només cosa de dos: hackejant els secrets del microbioma seminal
What it does
It uses the given data to train the implemented model and be able to predict whether a given subject is infertile or not.
How we built it
We used a jupyter notebook script. For the model, we used two dimensionality reduction techniques: Principal Component Analysis (PCA) and Neighborhood Components Analysis (NCA), combined with K-nearest neighbours (KNN), both approaches give a similar result.
Challenges we ran into
Finding the right algorithm or combination of algorithms to find the influence of a combination of bacteria on fertility.
What's next for microbioma-pca-knn
We would like to finish implementing the graphical interface, as we didn't have the time to finish this (so it has not been submitted).