Unlock instant, AI-driven research and patent intelligence for your innovation.

Feature fusion method for entities and user portraits

A feature fusion and user-friendly technology, applied in neural learning methods, character and pattern recognition, instruments, etc., can solve problems such as insufficient and reasonable use, avoid excessive feature dimensions, improve accuracy, and reduce noise features the effect of interference

Pending Publication Date: 2019-09-17
OCEAN UNIV OF CHINA
View PDF0 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, directly using or fine-tuning a single-level feature of the pre-trained network does not fully and reasonably use this pre-trained feature

Method used

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
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Feature fusion method for entities and user portraits

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Brief steps of the present invention see figure 1 ,include:

[0036] (1) Obtain the user's label information;

[0037] Obtain all kinds of data in the Internet, and integrate the obtained Internet data to form a knowledge base;

[0038] Obtain the user's online log;

[0039] Matching the online log with the knowledge base to form user tag information.

[0040] Use distributed crawlers to crawl to obtain various data from the Internet;

[0041] Subdividing the various types of data obtained from the Internet into categories, and then automatically merging labels and unifying the categories.

[0042] Specifically, the tag information of a user may be an inherent attribute of the user, may also be a dynamic attribute of the user, or may be a combination of the two, and different tag information may be obtained according to different business scenarios. Among them, the inherent attributes include attributes such as the user's age, gender, occupation, income level, and ...

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

PUM

No PUM Login to View More

Abstract

The invention relates to a feature fusion method for entities and user portraits. The method comprises the following steps: obtaining label information of a user; establishing a user portrait according to the label information; jointly extracting entity and user portrait information in the network through an end-to-end model based on the neural network; based on the portrait of the user, screening out entity contents which the user may like or be interested in; and carrying out feature fusion on the entites and the user portraits. The multi-layer features in the pre-training network are used as the multi-layer total pre-training features of the entities, and the multi-layer perceptron is used for supervised fusion and dimensionality reduction of the multi-layer total pre-training features of the entity under the guidance of the learning target that the entity is matched with the user portrait, so that the fused entity features are generated. Therefore, more useful pre-training features of different levels can be fully utilized, features useful for entity and user portrait matching tasks are concluded from the features, useless features are removed, and interference of noise features is reduced.

Description

technical field [0001] The invention belongs to the field of feature fusion of big data, and in particular relates to a feature fusion method oriented to entities and user portraits. Background technique [0002] With the development and progress of society, the construction of user portraits is becoming more and more important. User portraits can use the multi-dimensional view of data to objectively and truly reflect the user's behavior trajectory, habit characteristics and service needs, etc., and provide service capabilities in various fields. The mining of data analysis provides the necessary technical support. In the field of government big data fusion and cognition, with the task of matching entities and user portraits in recent years, it has gradually become popular in the fields of artificial intelligence and machine learning. We can now build an entity-to-persona matching system that matches appropriate personas based on entity content, and vice versa. This elimina...

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

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/253
Inventor 王晓东丁香乾王清
Owner OCEAN UNIV OF CHINA