Implementing and evaluating the performance of various Machine Learning algorithms with different datasets

Abstract

Machine learning algorithms are used to train the machine to learn on its own and improve from experience. It involves building the mathematical models to help in understand the data. When these models are applied with tunable parameters to the observed data. Using this program can be considered to be learning from the data. Once the models learned enough from the data given as input, they could be used for predicting and understand different features of new data. The supervised learning involves modelling the relationship between measured features of data and some label associated with data.  Once the model is trained with enough data and features, then new data can be given to the model for classification purpose. It is further classified into classification tasks and regression tasks. Unsupervised learning involves modelling the features of a dataset without reference to any label, and in this based on some similarity features data are grouped into some form. The similarity features are nothing but distance between the data is very minimum.  These models include tasks such as clustering and dimensionality reduction. Clustering algorithms identify distinct groups of data, while dimensionality reduction algorithms search for more simple representations of the data.