Decision Tree (DT) is located in artificial intelligence: Artificial Intelligence -> Machine Learning -> Supervised Learning -> Decision Tree. The decision tree is mainly used to solve the classification and regression problems, but the decision tree (DT) will produce over-fitting, which leads to the weakening of generalization ability. Overfitting is one of the important challenges in building a decision tree model. In view of the shortcomings of decision trees that are prone to overfitting, the Bell Farms of the United States proposed a random forest (RF) voting mechanism to improve decision trees. Random Forest (RF) is an improved method for the over-fitting problem of decision trees (DT), and Random Forest (RF) is a recently fired algorithm. Therefore, it is necessary to further explore the random forest (RF). ^_^
The location of random forest (RF) in artificial intelligence: artificial intelligence --> machine learning --> supervised learning --> decision tree --> random forest.
Random Forest (RF) refers to a classifier that uses multiple trees to train and predict samples. The classifier was first proposed by Leo Breiman and Adele Cutler and registered as a trademark.
So what is a random forest?
RandomForests is a classifier that contains multiple decision trees, and the categories of its output are determined by the mode of the categories output by the individual trees. Leo Breiman and Adele Cutler develop and infer the algorithm of random forests. The term random forest (RF) comes from the 1995 random decision forests proposed by Bell Labs' TIn Kam Ho. This method combines Breimans' "Bootstrap aggregaTIng" idea with Ho's "random subspace method" to build a collection of decision trees.
By definition we know that Random Forest (RF) has to build multiple decision trees (DT) and combine them together to get more accurate and stable predictions. One of the great advantages of a random forest is that it can be used for both classification and regression problems, which form exactly what most current machine learning systems need to face.
Random forests are a subclass of integrated learning that relies on the voting choices of the decision tree to determine the final classification result. Integrated learning solves a single prediction problem by building several model combinations. The simple principle of integrated learning is to generate multiple classifiers/models, each learning and making predictions independently. These predictions are ultimately combined into a single prediction and are therefore better than any single classification to make predictions.
The process of building a random forest:
Suppose N represents the number of training cases (samples), M represents the number of features, and the construction process of random forests is as follows:
1) Enter the number of features m to determine the decision result of a node on the decision tree; where m should be much smaller than M.
2) Sampling N training cases (samples) in the form of put back sampling, forming a training set, and using undrawn use cases (samples) for prediction, and evaluating the error.
3) For each node, m features are randomly selected, and the decision of each node on the decision tree is determined based on these features. According to m characteristics, calculate the best split mode.
4) Each tree will grow intact without pruning, which may be used after building a normal tree classifier.
5) Repeat the above steps to build another decision tree until a predetermined number of decision trees are reached, that is, a random forest is constructed.
Among them, the number of pre-selected variables (m) and the number of trees in the random forest are important parameters, which is very critical for system tuning. These parameters also play a crucial role in regulating the accuracy of random forest models. The scientific use of these indicators will significantly improve the efficiency of the random forest model.
FT06 Series Fused Terminal Connectors
Plug Terminal Block,Fuse Connector,H Type Terminal Block,Lighting Connector
Jiangmen Krealux Electrical Appliances Co.,Ltd. , https://www.krealux-online.com