Keyword-Based Open Information Extraction for Knowledge Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing open information extraction systems extract too much general information from text, making it difficult to extract relevant facts for specific tasks, and lack context sensitivity, leading to errors and inefficiencies in knowledge graph construction and automated decision-making.
Innovation Solution
A method for fact-relevant open information extraction using keyword queries expanded with aliases, combined with context, to generate a targeted knowledge graph that is computationally efficient and accurate, utilizing an OpenKG extractor trained on keywords and aliases, and employing pruning and classification techniques to refine the extracted information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If existing open information extraction systems extract all possible information from text, then the quantity of extracted information increases, but the relevance and accuracy for specific tasks decreases
Solution Approach 1:
The patent applies local quality by making the information extraction process task-specific and context-sensitive. Instead of uniformly extracting all possible information, the system adjusts extraction behavior based on the specific task requirements and contextual relevance, ensuring high-quality extraction only for relevant information while ignoring unrelated content.
Solution Approach 2:
The patent implements partial action by selectively extracting only the necessary portion of information required for specific tasks. The system uses task descriptions and contextual analysis to determine which information should be extracted, avoiding the excessive extraction of all possible information and focusing only on relevant facts.
2Measurement precision
If keyword queries are expanded with aliases and context, then the relevance and completeness of extracted information improves, but the complexity of the extraction process increases
Solution Approach 1:
The patent applies preliminary action by pre-expanding keyword queries with aliases and contextual information before the extraction process begins. This preparation step creates enriched query representations that guide the extraction process, reducing the need for complex real-time decision-making during extraction while improving relevance.
Solution Approach 2:
The patent uses task descriptions and contextual information as intermediaries between the keyword queries and the extraction process. These intermediaries bridge the gap by providing additional guidance and context, enabling more accurate extraction without requiring direct complex interactions between all system components.
3Reliability
If all information is extracted from text without filtering, then no relevant information is missed, but computational resources and processing time are wasted on unnecessary data
Solution Approach 1:
The patent applies the extraction principle by removing irrelevant information from the extraction process through task-specific filtering. The system extracts only the necessary facts required for specific tasks, leaving out unnecessary information that would consume computational resources without contributing to the task objective.
Solution Approach 2:
The patent implements partial action by performing extraction only on relevant portions of text identified through task descriptions and contextual analysis. This selective approach avoids the excessive processing of all text content, improving productivity while maintaining reliability for task-relevant information.
4Productivity
If general information extraction is performed without context sensitivity, then the extraction process is simpler and faster, but errors increase and task-specific accuracy decreases
Solution Approach 1:
The patent applies local quality by implementing context-sensitive extraction that adapts to task-specific requirements. The system adjusts its extraction behavior based on local contextual information and task descriptions, improving accuracy for specific tasks while maintaining reasonable processing speeds through targeted rather than exhaustive analysis.
Data Source
AI summary
A method for automated decision making in an artificial intelligence task by fact-relevant open information extraction and knowledge graph generation includes obtaining a keyword query for performing the fact-relevant open information extraction and expanding the keyword query using keyword alias and query generation. The fact-relevant open information extraction is performed to extract triples from a text which contains the keyword or the keyword alias. The knowledge graph is generated using the extracted triples and an open knowledge graph (OpenKG) extractor that has been trained using keywords and aliases. Supervised or unsupervised classification is performed using the generated knowledge graph to make the automated decision in the artificial intelligence task.