Method and device for obtaining target object characteristic quantity prediction model and prediction method and device thereof

Abstract

The invention discloses a method and device for obtaining a target object characteristic quantity prediction model and a prediction method and device thereof, and the method for obtaining the target object characteristic quantity prediction model comprises the steps: carrying out the characteristic engineering processing of the statistical data of a target object characteristic quantity under a target ring layer of a target geographic object according to the prediction demands of the target object characteristic quantity, obtaining a plurality of pieces of training data under a target ring layer of the target geographic object; training a plurality of preset regression models with different hyper-parameters through the plurality of pieces of training data, and obtaining a first target object characteristic quantity prediction model corresponding to each preset regression model; and screening at least one first target object characteristic quantity prediction model as a second target object characteristic quantity prediction model according to a preset screening condition based on the determinable coefficients of the plurality of first target object characteristic quantity prediction models and the average error rates corresponding to the plurality of first target object characteristic quantity prediction models.

Description

technical field [0001] The present application relates to the technical field of data analysis, and in particular to a method and device for obtaining a prediction model of a feature quantity of a target object and its prediction. Background technique [0002] With the rapid development of science and technology, it is necessary to predict the characteristic quantity of the target object in the field of real estate in order to control the relevant situation of the characteristic quantity of the target object. The traditional method of predicting the characteristic quantity of the target object refers to using the common excel line chart to compare the degree of influence of different dimensional data on the characteristic quantity of the target object, coupled with the research experience of the forecaster on the characteristic quantity of the target object, and deducing the trend of the characteristic quantity of the target object to predict target object features. [0003...

AI Technical Summary

Problems solved by technology

[0003] However, each circle of each geographical object has its own characteristics of market, customer reserve, competition, etc., and the feature quantity of the target object is clearly differentiated and different in different circles of different geographical objects; It is necessary to predict each circle layer of each geographic object, which consumes a lot of manpower; the degree of influence of different dimensional data on the characteristics of the target object cannot be accurately quantified, resulting in the inability to maximize the use of different dimensional data; and relying on forecasters to The research experience of the characteristic quantity of the target object causes the prediction result to be subjective and accidental

Images