Using Machine Learning to Predict Penguin Species

Abstract

Machine learning is used to make a neural network to predict penguin species from physical attributes. The data set is first prepared, with missing and extraneous data deleted. After this, there remains 333 examples consisting of three different penguin species. These remaining examples are fed into a neural network which proceeds to train on them. The neural network consists of an input and output layer with no hidden layers and uses the softmax activation formula. 25% of the data set is used as a validation set with the remaining 75% used as the training set.