Decision tree: Basic classification and regression method, which can be converted into a set of if-then rules and can be regarded as a conditional probability distribution of classes defined on the feature space division.
-Advantages: The model is readable and the classification speed is fast
-The target decision tree has less conflict with the training set data and has good generalization ability.
The essence of decision tree learning: to summarize a set of classification rules from the training data set; purpose: to build a model that fits the training data well and has low complexity Decision tree
- Commonly used algorithms: ID3, C4.5, CART
Feature selection:
The generation of decision tree corresponds to The local selection of the model; the decision tree corresponds almost to the global selection of the model
-Generation of the decision tree:
-Pruning of the decision tree:
(5) Decision tree
Guess you like
Origin blog.csdn.net/daisyxyr/article/details/125900654
Ranking