Machine learning algorithms emulate human learning, reasoning, and decision making, using techniques borrowed from artificial intelligence, data mining, and statistics. There are two classes of machine learning algorithms—supervised and unsupervised. For the majority of decision-making problems in machine learning, we must construct a model (or algorithm) that adequately captures and processes the required input and, in the case of supervised learning, also provides robust (accurate and precise) predictions of the output. Logistic regression is the most common classification algorithm used in machine learning. Machine learning, or supervised learning, is a set of approaches that adds the capacity to discriminate between clusters in a statistically defined algorithm. It is distinct from unsupervised learning in that it is prospective, having the ability to apply the algorithm to new samples rather than describe patterns in the original dataset.