Data sub-item quantification method and device and electronic equipment

Abstract

The invention relates to a data item quantization method. The method comprises the following steps: extracting a feature quantization parameter and a magnitude parameter from a database; establishing a regression model, and training the regression model; wherein in the training, a feature quantization parameter is used as an xn input regression model, and a magnitude parameter is used as a y input regression model; obtaining bn as a feature weight coefficient based on the trained regression model; and generating and sending sub-item quantization information to a user side, wherein the sub-item quantization information comprises magnitude difference data obtained based on the feature quantization parameter and the feature weight coefficient. The invention also relates to a method for establishing and training the regression model, and equipment and an electronic device for implementing the methods. According to the technical scheme, the influence of the feature data on the magnitude data can be visually presented, and meanwhile, result errors caused by overfitting are avoided.

Description

technical field [0001] The application belongs to the technical field of data analysis, and specifically relates to a method for sub-item quantification of data, and related devices and electronic equipment. Background technique [0002] When a user wants to know multiple products with different characteristics, such as different product information such as age, mileage, working conditions, and transaction times, the prior art usually provides the information of multiple products directly to the user, so that The user compares by himself, so that the user needs to spend a lot of time and energy to select the desired product, or directly provides the comparison results of multiple products to the user, and the user often does not know what the basis of these comparison results is; None of the technical methods can enable users to have a quantitative understanding of the impact of various characteristics of the product, and cannot meet the needs of users who want to understand...

AI Technical Summary

Problems solved by technology

[0002] When a user wants to know multiple products with different characteristics, such as different product information such as age, mileage, working conditions, and transaction times, the prior art usually provides the information of multiple products directly to the user, so that Users make comparisons by themselves, so that users need to spend a lot of time and energy to select the products they need, or directly provide users with the comparison results of multiple products, and users often do not know what the basis of these comparison results is; Technical methods cannot enable users to have a quantitative understanding of the impact of various characteristics of the product, and cannot meet the needs of users who want to understand the specific differences brought about by various characteristic factors.

Images