Web table-oriented paired entity joint disambiguation method

An entity and table technology, applied in the field of knowledge graph, can solve the problem of not fully suitable for knowledge graph and web form, strong assumptions, etc., to achieve high-quality joint disambiguation and ensure reliability.

Pending Publication Date: 2021-09-07
SOUTHEAST UNIV
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The joint disambiguation method has achieved a good disambiguation effect, but it has the disadvantage of too strong assumptions, which is not completely suitable for knowledge graphs and Web tables in reality.

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
  • Web table-oriented paired entity joint disambiguation method
  • Web table-oriented paired entity joint disambiguation method
  • Web table-oriented paired entity joint disambiguation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The implementation process of the present invention will be described in detail below in conjunction with the embodiments and the accompanying drawings.

[0038] The present invention is designed to complete the table entity linking task with the entity joint disambiguation algorithm, which mainly includes the following steps:

[0039] 1) Column semantic consistency calculation.

[0040] According to the characteristics of the table, the cell contents in the same column have similar semantic characteristics. In the entity linking task, entities linked in the same column usually belong to a certain category, which makes these linked entities have similar vector representations to a certain extent. The present invention designs a column semantic consistency calculation method based on variance.

[0041] Given a column of data in a web form, first calculate the variance element-wise of the vector representation of the linked entities in the column, and obtain a variance v...

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 discloses a Web table-oriented paired entity joint disambiguation method, which is used for solving a Web table-oriented entity link task. The Web table-oriented entity linking task is to link the entity mentions in the Web table to the entities in the knowledge base in an unambiguous manner. The paired entity joint disambiguation method is designed according to the table characteristics, joint disambiguation is conducted on a pair of entity mentions with the highest confidence coefficient iteratively, and all disambiguation of the entity mentions in the whole table is achieved step by step. The confidence coefficient calculation method comprehensively considers various information, including similarity of entity mention and candidate entities, consistency between linked entities and semantic consistency of rows and columns in a table. In the algorithm iteration process, the linked entities have very high confidence, and effective auxiliary information can be provided for subsequent linking work, so that high-quality joint disambiguation is realized.

Description

technical field [0001] The invention relates to a paired entity joint disambiguation method oriented to a Web form, and belongs to the technical field of knowledge graphs. Background technique [0002] Web forms organize data in a structured form and provide high-quality and high-density information. It is estimated that the Web contains 14.1 billion tables, of which about 154 million are related tables. In order to utilize these valuable data, it is necessary for the computer to understand these tables from the semantic level. Entity linking of tables is an effective means to realize table understanding. [0003] Entity linking in tables requires associating entity mentions in table cells with corresponding entities in the knowledge graph. An effective table entity linking system should be able to unambiguously link entity mentions to corresponding entities in the knowledge graph according to the context information of the entity mentions in the table. Different from th...

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): G06F40/30G06F40/295G06K9/62
CPCG06F40/30G06F40/295G06F18/22
Inventor 吴天星李林漆桂林
Owner SOUTHEAST UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products