Naïve Bayes is a supervised machine learning algorithm based on Bayes' Theorem. It is primarily used for classification tasks such as spam detection, sentiment analysis, document classification, disease prediction, and recommendation systems. The algorithm is called "naïve" because it assumes that all input features are conditionally independent of one another given the class label. Although this assumption is rarely true in real-world datasets, Naïve Bayes often performs remarkably well due to its simplicity, efficiency, and ability to handle high-dimensional data. Bayes' Theorem provides the mathematical foundation of the algorithm and is expressed as: [ P(C|X)=\frac{P(X|C)\times P(C)}{P(X)} ] where (P(C|X)) is the posterior probability of class (C) given the feature vector (X), (P(X|C)) is the likelihood of observing the features in class (C), (P(C)) is the prior probability of the class, and (P(X)) is the evidence. Since (P(X)) remains constant for all classes, classification is performed by selecting the class with the highest posterior probability. The working of the Naïve Bayes algorithm involves several steps. First, a labeled training dataset is collected. Next, the prior probability of each class is calculated. The likelihood of every feature corresponding to each class is then estimated from the training data. When a new instance is provided, the posterior probability for every