Linear SVM: Find a line dividing the data set into two parts that minimizes the error

Nonlinear SVM: uses kernel function to map data to a higher dimensional space to make it linearly separable

Random forest: An algorithm that combines multiple decision trees to integrate the classification results of each tree and vote for the final category result.

Neural network: Similar to biological neurons, it constructs input layer, output layer, and hidden layer to implement model training from input to output and obtain corresponding results.

Desicion tree: Construct visual classification rules and generate an interpretable model

Logistic regression: Transform the dependent variable y of the linear function through a logistic function so that its probability distribution is between 0 and 1

bayesian classifier: Calculate the conditional probability of occurrence of each category to determine the classification result

GBDT: A technology for learning from mistakes that integrates a bunch of poor learning algorithms for learning

Random forest: An algorithm that combines multiple decision trees to integrate the classification results of each tree and vote for the final category result.

Neural network: Similar to biological neurons, it constructs input layer, output layer, and hidden layer to implement model training from input to output and obtain corresponding results.

Desicion tree: Construct visual classification rules and generate an interpretable model

Univariate linear regression: Construct a linear model to express the relationship between a dependent variable and an independent variable to minimize the error

Multiple linear regression: Construct a linear model to express the relationship between a dependent variable and multiple independent variables

GBDT: A technology for learning from mistakes that integrates a bunch of poor learning algorithms for learning

PCA: Convert a set of potentially correlated variables into a set of linearly uncorrelated variables through orthogonal transformation

LDA: Project the data set onto a straight line so that projection points of the same type are as close as possible and projection points of different types are as far away as possible

SVD: A complex matrix is ​​represented by the multiplication of several smaller and simpler submatrices, and these small matrices express the important characteristics of the matrix.

A density clustering algorithm suitable for convex and non-convex sample sets, assuming that the category of the sample can be determined by the tightness of the sample distribution

K-MEAN: For K categories in a given data set, the center of each category is obtained by calculating the mean value of all values ​​in the category, and distance is used as the similarity index to perform clustering.

Hierarchical clustering: Calculate the similarity between nodes through a certain degree of similarity, and form a tree diagram hierarchy according to the network nodes to perform hierarchical splitting