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

Weakly supervised semantic entity recognition using general and target domain knowledge

A target field and semantic technology, applied in semantic analysis, semantic tool creation, special data processing applications, etc., can solve performance dependencies and other problems

Pending Publication Date: 2021-10-19
ROBERT BOSCH GMBH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the performance of these methods is highly dependent on large amounts of manually labeled data
In many scenarios, there is insufficient or even no manually labeled data for training domain-specific semantic entity recognition systems, making it challenging to apply existing methods to recognize domain-specific entities

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
  • Weakly supervised semantic entity recognition using general and target domain knowledge
  • Weakly supervised semantic entity recognition using general and target domain knowledge
  • Weakly supervised semantic entity recognition using general and target domain knowledge

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0030] The following examples illustrate the example systems and methods described herein. Example 1: A system for performing semantic entity recognition, the system comprising: a general knowledge data repository; a domain-specific expert knowledge data repository; and an electronic processor configured to: access documents stored in a memory; Based on the specified target domain, select the target domain information from the general knowledge data repository; based on the selected target domain information and expert knowledge from the domain-specific expert knowledge data repository, generate multiple weak annotators for the document; combine all The plurality of weak annotators are applied to the document to generate a plurality of weak labels; selecting at least one weak label from the plurality of weak labels as training data; and using the training data to train a semantic entity prediction model.

example 2

[0031] Example 2: The system of Example 1, wherein the electronic processor is further configured to preprocess the document to generate the unlabeled dataset.

example 3

[0032] Example 3: The system of Example 2, where multiple latent semantic entities are generated using an unlabeled dataset.

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

Provided is weakly supervised semantic entity recognition using general and target domain knowledge. Methods and systems for performing semantic entity recognition. The method includes accessing a document stored in a memory and selecting from a general knowledge data repository, target domain information based on a specified target domain. The method also includes generating a plurality of weak annotators for the document based upon the selected target domain information and expert knowledge from a domain-specific expert knowledge data repository and applying the plurality of weak annotators to the document to generate a plurality of weak labels. The method further includes selecting at least one weak label from the plurality of weak labels as training data and training a semantic entity prediction model using the training data.

Description

technical field [0001] Embodiments relate to semantic entity recognition using both general knowledge and target domain knowledge. Background technique [0002] One purpose of semantic entity recognition is to identify entities, concepts or terms in documents, such as function names or signal names. Identifying these semantic entities is an important step towards extracting structured information from unstructured text data. [0003] In the general field, there are many approaches that exploit named entity recognition (eg, people, places, and organizations). However, the performance of these methods is highly dependent on a large amount of manually labeled data. In many scenarios, there is insufficient or even no manually labeled data for training domain-specific semantic entity recognition systems, making it challenging to apply existing methods to recognize domain-specific entities. Contents of the invention [0004] Thus, among other goals, one goal of some embodimen...

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
IPC IPC(8): G06F40/30G06F40/295G06F16/36
CPCG06F40/30G06F40/295G06F16/367G06F40/279
Inventor 赵心言丁海波冯哲
Owner ROBERT BOSCH GMBH