Knowledge Graph Construction for News Event Information Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the era of information explosion, readers often receive incomplete or incorrect information about news events due to varying levels of expertise among interviewees, leading to inconsistent reporting across different media sources.
Innovation Solution
A method and system for building a knowledge graph that classifies news articles into main events and sub-events, extracts event summaries, and identifies commenter identities, allowing for the creation of a structured graph that can quickly provide relevant information when queried by users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If news articles are read directly to obtain information, then readers can receive various information from different media, but the information may be incomplete or incorrect due to varying expertise levels of interviewees
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary between news articles and readers. The knowledge graph extracts, verifies, and structures information from multiple news sources, creating a reliable intermediate representation that filters out inaccurate or incomplete information while preserving the quantity of information available across different media sources.
Solution Approach 2:
The patent segments information into structured components (main events, sub-events, entities, relationships) within the knowledge graph. This segmentation allows systematic verification of information accuracy by breaking down complex news content into verifiable factual elements, while maintaining the comprehensive quantity of information from multiple sources.
2Reliability
If readers search through multiple news articles to find accurate information, then they can access various perspectives, but they spend excessive time on information retrieval
Solution Approach 1:
The system performs preliminary action by pre-processing news articles to extract and verify information, building the knowledge graph in advance. This preliminary extraction and verification of facts, entities, and relationships eliminates the need for readers to manually search through multiple articles, significantly reducing information retrieval time while ensuring accuracy through pre-verified structured data.
Solution Approach 2:
The patent creates a copied and structured version of information from multiple news articles in the knowledge graph. This copied structured representation contains verified factual elements that can be quickly queried, allowing readers to obtain accurate information without physically reading through the original time-consuming news articles.
3Reliability
If a knowledge graph is built from multiple news articles to verify information, then information accuracy improves, but the system complexity increases
Solution Approach 1:
The patent segments the information processing system into distinct functional modules: news article parsing, entity extraction, event extraction, relationship identification, and knowledge graph construction. This segmentation manages system complexity by organizing complex processing tasks into manageable, independent components that can be developed and maintained separately while working together to improve information accuracy.
Solution Approach 2:
The knowledge graph itself serves as an intermediary data structure that simplifies the complexity of processing multiple news articles. By transforming unstructured article data into a standardized graph structure with verified entities and relationships, the system manages complexity through a unified intermediate representation that can be efficiently queried and updated.
Data Source
AI summary
A method of building a knowledge graph, performed by a processing device, includes: classifying news articles to a main event associated with sub events, using the main event as a first node of the knowledge graph, using the sub events as second nodes of the knowledge graph respectively, connecting the second nodes to the first node, extracting event summaries from the news articles respectively according to a template, using the event summaries as third nodes of the knowledge graph respectively, and connecting each of the third nodes to one of the second nodes according to association between the event summaries and the sub events, extracting commenter identities from the event summaries, and using the commenter identities as fourth nodes of the knowledge graph, and connecting each of the fourth nodes to one of the third nodes.