Who Invented Naive Bayes?
1 Naive Bayes History
Naive Bayes algorithm was invented by Thomas Bayes and introduced in 1730s. Thomas Bayes didn’t live a very long life and his work was rigorously studied, curated, structured, corrected and even published by his friend Richard Price.
“An Essay towards solving a Problem in the Doctrine of Chances” was published in 1763 by Price on behalf of Thomas Bayes which was the first publication including Bayes Theorem.
More than two centuries since its discovery, Bayes Theorem inspired and influenced many discussions, algorithms, research and technology in statistics field and pioneered probabilistic statistics.
2 Chronological publications of Thomas Bayes
Thomas Bayes only published two papers when he was alive and a third paper introducing Bayes Theorem was edited and published by his friend Richard Price. You can see the dates and titles of these three papers below:

Thomas Bayes – Divine Benevolence, or an Attempt to Prove That the Principal End of the Divine Providence and Government is the Happiness of His Creatures : 1731

Thomas Bayes – An Introduction to the Doctrine of Fluxions, and a Defence of the Mathematicians Against the Objections of the Author of The Analyst : 1736

Richard Price – An Essay towards solving a Problem in the Doctrine of Chances : 1763 : read here
3 What Inspired Thomas Bayes to Invent Bayes' Theorem?
Thomas Bayes, an English mathematician and theologian, developed the foundation of what is now known as Bayesian probability theory. Unfortunately, there is limited information available about the specific inspiration behind Bayes’ work, as he did not publish his ideas during his lifetime. His influential paper, “An Essay towards solving a Problem in the Doctrine of Chances,” was published posthumously in 1763 by his friend Richard Price.
The inspiration for Bayes’ theorem likely came from his interest in the philosophy of probability and his desire to understand how to update beliefs in the face of new evidence. Bayes was influenced by earlier works in probability theory, including the works of Thomas Simpson and PierreSimon Laplace.
 One possible inspiration for Bayes’ theorem could be attributed to the “inverse probability” method proposed by Thomas Simpson. Simpson suggested a method for calculating the probability of an event based on observed data. Bayes expanded on this idea by considering how prior beliefs (prior probabilities) could be updated using observed data (likelihoods) to obtain posterior probabilities.
 Another influence could be PierreSimon Laplace’s work on probability. Laplace made significant contributions to probability theory and statistical inference, and his ideas may have influenced Bayes. Laplace developed what is known as the “principle of indifference” or the “principle of insufficient reason,” which suggests that in the absence of prior knowledge, all outcomes are equally likely. Bayes’ theorem provides a way to update these prior beliefs using observed data.
It is important to note that while Bayes’ theorem is attributed to Thomas Bayes, the formulation and development of the theorem might have involved contributions from others, including Richard Price, who edited and published Bayes’ work after his death.
Overall, the exact inspirations that led to Thomas Bayes’ formulation of Bayesian theorem are not welldocumented. However, we can make pretty sophisticated speculations that his work was influenced by earlier ideas in probability theory and his interest in updating beliefs based on new evidence. This is a particularly interesting point considering it lines up with his progressive theological views and active position in the Protestant church.
4 Significance of Bayes' Theorem for Modern Statistics & Artificial Intelligence
Bayes’ theorem played a crucial role in the development of modern statistics by introducing a formal framework for updating beliefs and making inferences based on probabilistic reasoning. Here’s how Bayes’ theorem contributed to the field of modern statistics:
 Bayesian Inference: Bayes’ theorem forms the foundation of Bayesian inference, which is a fundamental concept in modern statistics. Bayesian inference provides a coherent framework for incorporating prior beliefs or knowledge and updating them based on observed data. It allows for the quantification of uncertainty and the estimation of unknown quantities through the calculation of posterior probabilities.
 Prior and Posterior Distributions: Bayes’ theorem enables the construction of prior distributions, which represent the initial beliefs or knowledge about the parameters or variables of interest. By combining the prior distributions with observed data, Bayesian inference produces posterior distributions that reflect the updated beliefs after considering the data. These distributions serve as a powerful tool for summarizing uncertainty and making probabilistic statements about the parameters or variables.
 Quantifying Uncertainty: One of the strengths of Bayes’ theorem is its ability to handle uncertainty in a principled manner. Through the use of prior and posterior distributions, Bayesian inference provides a natural way to quantify uncertainty and express it in terms of probabilities. This is in contrast to frequentist statistics, where uncertainty is typically expressed through confidence intervals or pvalues.
 Decision Theory: Bayes’ theorem is intimately connected to decision theory, which focuses on making optimal decisions under uncertainty. Bayesian decision theory provides a framework for decision making by explicitly considering the probabilities of different outcomes and their associated costs or benefits. It allows for rational decision making by balancing the available information with the potential consequences of different actions.
 Complex Models and Data: Bayes’ theorem allows for the development of complex statistical models that can handle a wide range of data and variables. Through the use of prior distributions, Bayesian inference can incorporate prior knowledge, expert opinions, and regularization techniques to handle limited data or highdimensional problems. This flexibility has led to the development of sophisticated Bayesian models, such as hierarchical models, Gaussian processes, and Bayesian neural networks.
 Computational Advances: The field of Bayesian statistics has benefited from significant computational advancements that have made Bayesian inference more accessible and feasible. Techniques such as Markov chain Monte Carlo (MCMC) and variational inference have facilitated the estimation of posterior distributions even for complex models. These computational tools have played a crucial role in expanding the practical application of Bayesian methods.
In conclusion, Bayes’ theorem laid the groundwork for modern statistics by introducing the concepts of prior and posterior distributions, enabling the quantification of uncertainty, providing a foundation for decision theory, accommodating complex models and data, and benefiting from computational advancements. The Bayesian framework continues to be a vibrant area of research and application in statistics, offering a powerful and flexible approach to modeling and inference.
5 Summary
In this brief article we have covered the history of Naive Bayes Theorem which Naive Bayes Machine Learning algorithm is based on.