Method and device for predicting target variable

Abstract

The invention discloses a method and device for predicting a target variable in the technical field of computers. The specific embodiment of the method comprises the steps of obtaining feature data and a model identifier; determining a prediction model which corresponds to the model identifier and is obtained by training a machine learning model; and inputting the feature data into the predictionmodel to predict a target variable. A model trained by machine learning is adopted to predict a target variable, so that the problem of inaccurate link function selection caused by irregular distribution of the target variable and poor mathematical distribution fitting when a generalized linear model is adopted is avoided. The target variable is directly modeled, the prediction model obtained through machine learning is more flexible in characteristic space division compared with a traditional generalized linear model, and the predicted target variable is more accurate.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method and device for predicting target variables. Background technique [0002] In the prior art, the generalized linear model GLM (Gerneralized Linear Model) is mostly used to predict the target variable, and the premise of the generalized linear model is that the target variable satisfies a certain mathematical distribution, and an appropriate link function is selected according to the mathematical distribution. For example, in the field of auto insurance, the generalized linear model is used to introduce the sub-vehicle factor and a reasonable risk distribution estimate to determine the appropriate link function to achieve a more reasonable auto insurance pricing. [0003] In the course of realizing the present invention, the inventor finds that there are at least the following problems in the prior art: [0004] 1. The premise of establishing a generalized linea...

AI Technical Summary

Problems solved by technology

[0004] 1. The premise of establishing a generalized linear model is to find a suitable link function, that is, to find the mathematical distribution that the target variable satisfies. However, the determinants of the data distribution are intricate and related to the amount of data. It is usually difficult to fully describe it with a certain distribution, which brings great difficulties to the modeling. to uncertainty

[0005] 2. With the continuous introduction of data features, the feature space of the model continues to increase, and the nonlinear characteristics become more and more significant. Even after dividing according to some significant categorical variables and then using the generalized linear model for modeling, the model of linear combination factors has already Incapable of such complex data characteristics, as the amount of data increases, the resolution of the feature space of the linear model will lag behind the resolution of the data in the feature space, resulting in a decline in its predictive ability

Images