Which classifier models belong to linear classifiers in machine learning? Which belong to nonlinear classifiers?

Linear classifier: single-layer perceptron network, Bayesian.

Factors affecting the error rate of classifier;

Number of records in the training set. The generator should use the training set to learn, so the larger the training set, the more reliable the classifier. However, the larger the training set, the longer it takes the generator to build the classifier. The improvement of error rate decreases with the increase of training set size.

The number of attributes. More attributes mean more combinations for the generator, which makes the generator more difficult and takes longer. Sometimes random relations will lead the generator astray, and as a result, an inaccurate classifier may be constructed (this is technically called over-fitting).

Extended data:

Decision tree classifier: This process is similar to identifying plants by their characteristics. Such a classifier can be used to judge a person's credit degree. For example, the decision tree can conclude that "a person with a family, a car with a value of 654.38+0.5 million to 23,000 dollars and two children" have good credit.

The decision tree generator generates a decision tree from the training set. The visualization tool provided by MineSet, a data mining tool of SGI company, shows the structure of decision tree classifier with tree diagram. In the graph, each decision is represented by a node of the tree.

Baidu encyclopedia-classifier