Feature processing method and device

Abstract

The invention relates to a feature processing method and device. According to one embodiment of the present disclosure, the method comprises: allocating an object ID to an object; allocating a type IDto the type of the feature of the object; generating a hash value corresponding to the feature of the object by using an operator corresponding to the type; and generating a signature representing afeature of the object based on the object ID, the type ID, and the hash value of the object. The method and device disclosed by the invention at least have one of the following beneficial technical effects of enriching the expression capability of the feature vector and ensuring that the feature processing process has relatively good usability and expandability.

Description

technical field [0001] The present disclosure relates to computer technology, and in particular, to a feature processing method and device. Background technique [0002] Feature processing is an important part of the model training and service deployment process, which can be understood as the mapping process from user, context, item and other attribute values ​​(discrete or continuous attributes) to several feature vectors. A single feature is easier to express, but in the process of model construction, a large number of attribute values ​​need to be used, with various types, and the final vector dimension may be tens of thousands or even billions, which leads to the difficulty in managing features and processing features. There are certain difficulties. In the process of feature processing and service deployment, in order to strictly ensure the consistency of the data processing process, it is necessary to use strict and consistent logic modules to perform this operation,...

AI Technical Summary

Problems solved by technology

A single feature is easier to express, but in the process of model construction, a large number of attribute values ​​need to be used, with various types, and the final vector dimension may be tens of thousands or even billions, which leads to the difficulty in managing features and processing features. there are certain difficulties

In the process of feature processing and service deployment, in order to strictly ensure the consistency of the data processing process, it is necessary to use strict and consistent logic modules to perform this operation, such as using a common class library or the same set of codes. Therefore, when adding features Code intervention is often required, resulting in limited usability and scalability

Images