Feature construction method and device, equipment and storage medium

A feature construction and feature set technology, applied in special data processing applications, unstructured text data retrieval, resources, etc., can solve problems such as high-dimensional sparseness, features cannot be effectively used, and achieve the effect of increasing feature density
CN113240304APending Publication Date: 2021-08-10BEIJING BAIDU NETCOM SCI & TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
BEIJING BAIDU NETCOM SCI & TECH CO LTD
Publication Date
2021-08-10

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention provides a feature construction method and device, equipment, a storage medium and a program product, and relates to the field of artificial intelligence, in particular to the technical field of natural language processing and deep learning. A specific embodiment of the method comprises the following steps: acquiring an original user behavior feature set; determining semantic information of original user behavior characteristics in the original user behavior characteristic set; according to the semantic information, performing feature merging on original user behavior features in the original user behavior feature set, and generating a target user behavior feature set, wherein feature dimensions contained in the target user behavior feature set are smaller than feature dimensions contained in the original user behavior feature set. According to the embodiment, the feature dimension contained in the user behavior feature set is reduced, the feature intensity contained in the user behavior feature set is increased, and the information of the feature semantic level is reserved.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present disclosure relates to the field of artificial intelligence, further relates to the technical fields of natural language processing and deep learning, and specifically relates to a feature construction method, device, equipment, storage medium and program product. Background technique

[0002] In the risk control scenario, using the user's historical behavior data can predict whether the user will default. Therefore, it is often necessary to process user behavior data to construct user behavior characteristics for risk control modeling. Traditional methods usually directly digitize user behavior data, such as OneHot (one-hot) encoding, calculating mean, variance, etc.

[0003] However, the user behavior data in the Internet scene has the characteristics of high dimensionality and sparseness, and the sample data in the wind control scene has the characteristics of small sample imbalance. In this case, the tree model commonly used to predict ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More