Naive Bayesian classifier is a simple probability classifier based on independent hypothesis Bayesian theorem. The reason why you become naive should be the literal translation of Naive, which means simple, naive and naive.

1, Bayesian method

Bayesian method is based on Bayesian principle, which uses the knowledge of probability and statistics to classify sample data sets. Because of its solid mathematical foundation, Bayesian classification algorithm has a low misjudgment rate.

Bayesian method is characterized by combining prior probability and posterior probability, avoiding the subjective deviation of using only prior probability and the over-fitting phenomenon of using only sample information. Bayesian classification algorithm shows high accuracy in the case of large data sets, and the algorithm itself is relatively simple.

2. Naive Bayesian algorithm

Naive Bayesian algorithm is one of the most widely used classification algorithms.

Naive Bayesian method is simplified on the basis of Bayesian algorithm, that is, given the target value, the attributes are assumed to be independent of each other. In other words, no attribute variable accounts for a large proportion of the decision-making results, and no attribute variable accounts for a small proportion of the decision-making results.

Although this simplified method reduces the classification effect of Bayesian classification algorithm to a certain extent, it greatly simplifies the complexity of Bayesian method in practical application scenarios.

Extended data

Research significance

People need to estimate the probability of various conclusions when reasoning and making decisions based on uncertain information. This kind of reasoning is called probabilistic reasoning. Probabilistic reasoning is not only the research object of probability theory and logic, but also the research object of psychology, but the research angle is different. Probability and logic study formulas or rules for objective probability calculation.

Psychology studies the cognitive processing law of subjective probability estimation. The problem of Bayesian inference is conditional probabilistic inference. The discussion in this field is of great theoretical and practical significance for revealing people's cognitive processing process and law of probability information and guiding people to learn and judge effectively.