Machine Learning: What, Why, and How?

Publication: Bioinformatics: Sequences, Structures, Phylogeny
Software: ADMET Predictor®


Since the beginning of computers era, an enormous amount of data generated seems to be ever-increasing and could be of great use with efficient learning techniques. Learning from the data to make reliable predictions, discovering new patterns and theories has been the most challenging task for the researchers. Machine learning detects the hidden insights in the data, learns from them, and makes reliable predictions on the unseen data. It is used in a range of applications that include bioinformatics, cheminformatics, marketing, linguistics, email filtering, optical character recognition, and many more. In the present chapter, we have described the types of learning, applications of machine learning, steps to generate machine learning models using various learning algorithms, and validation of generated models. Additionally, step-by-step generation of models using Weka workbench which is a collection of machine learning algorithms and data preprocessing tools has also been discussed.