Subject-Specific Look-Up Tables for Named Entity Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data processing techniques face difficulties in resolving named entities from unstructured text data, particularly in mapping unstructured text to predefined structures or knowledge databases due to lack of metadata and ambiguity in entity recognition.
Innovation Solution
A method involving the use of subject-specific look-up tables to process webpage data, identifying matching strings using string matching techniques, and performing named entity resolution by linking text data to corresponding entities within a knowledge database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subject-specific look-up tables are used to improve named entity resolution accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the knowledge base into multiple subject-specific look-up tables, where each table contains entities and text data relevant to a particular subject. This segmentation allows the system to select and use only the appropriate look-up table for the given subject, improving resolution accuracy while managing complexity through organized division of data.
Solution Approach 2:
The patent performs preliminary organization of entity data into subject-specific look-up tables before the actual named entity resolution process. By pre-structuring the data according to subjects and their related entities, the system prepares the information in advance, enabling faster and more accurate resolution during text processing without requiring complex real-time computations.
2Adaptability or versatility
If multiple look-up tables are maintained for different subjects, then adaptability is improved, but loss of information increases
Solution Approach 1:
The patent creates a universal framework where each subject-specific look-up table follows the same structure and can be used for any subject. The look-up tables are designed to handle different types of entities and relationships in a consistent manner, allowing the system to adapt to various subjects while maintaining information completeness through a standardized approach.
Solution Approach 2:
The patent creates multiple copies of the look-up table structure for different subjects, where each copy contains subject-specific entities and text data. This copying approach allows the system to maintain subject-specific information separately while following a unified structure, preventing information loss by ensuring each subject's data is preserved in its dedicated table.
3Productivity
If string matching techniques are used to identify related entities, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The patent segments the text processing task by first using string matching to quickly locate potential entity mentions, then applying subject-specific context from the look-up tables to verify and refine the identification. This segmentation allows fast initial filtering followed by precise verification, maintaining both productivity and accuracy.
Solution Approach 2:
The patent introduces subject-specific context from the look-up tables as an intermediary between string matching and final entity identification. The look-up tables provide subject-relevant entities and text data that act as a bridge, helping to disambiguate string matches and improve identification accuracy without significantly slowing down the overall process.
Data Source
AI summary
A method comprising receiving subject data indicative of a subject entity and selecting, from a plurality of data sets, and based on the subject data, a subject entity data set which corresponds to the subject entity. The subject entity data set comprises first related entity data representative of a first related entity related to the subject entity and first text data representative of first text associated with the first related entity. Unstructured text data representative of unstructured text is received and processed, using the first text data, to identify a portion of the unstructured text data corresponding to the first text data. The first text data is used to identify, from the subject entity data set, the first related entity data and the portion of the unstructured text data is identified as corresponding to the first related entity data.