This chapter focuses on machine learning, deep learning, artificial neural network, and Natural language processing-based sentiment analyses, which are all important application in healthcare and drug discovery. Machine learning has been considered in terms of categories and subcategories, viz., (i) supervised, (ii) unsupervised, (iii) semi-supervised, and (iv) reinforcement learning. Similarly drug discovery is also classified in stages and steps. The development process has been categorized as classification, regression, protein modeling, de novo molecular designs, protein folding, virtual screening, toxicity predication, etc., and these when made part of AI transforms into (ML, DL, ANN, NLP) algorithms when applied to performance. Also, this chapter classifies analysis in terms of regression analysis type (Gaussian process, ensemble methods, decision trees, PCA, SVR, LASSO, etc.) and classifier type (SVMs, kernel methods, Bayesian classifier, etc.) under the category “supervised learning” techniques. It also highlights the clustering methods (K-means, GANs, Kohorien maps, Markova, etc.,) of the unsupervised learning techniques, into specific process and steps of drug discovery and development. These include target drug ability, ADME properties in targets and chemical synthesis, image-based diagnosis, compound bioactivity, drug sensitivity predication, phenol typing of cellular images, CT image analysis and, many more.