A Comparative Analysis of Supervised Machine Learning Methods using Disaster Datasets

Abstract

Supervised machine learning is one of the machine learning task that generates required function from the training data which is labelled. The aim of supervised machine learning is to build or construct a model that makes predictions by using the function inferred from the labelled training data. This paper put a light on how the supervised machine-learning techniques are used to build a predictive model from the dataset of titanic disaster and also a comparative analysis of supervised machine learning methods like Random Forests and Decision Trees are implemented. In this work, with a training dataset containing features or labels like sex, age and class, survivors are predicted from the four test datasets. And from the observations of results a comparative analysis of both supervised machine learning methods namely Decision Trees and Random Forests is implemented.