Event Knowledge Base Construction via Syntax Fragment Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional entity knowledge bases fail to provide accurate search results for in-depth knowledge queries, such as event-related information, leading to a reduced user search experience.
Innovation Solution
A method and device for constructing an event knowledge base by identifying text to obtain event mining candidate sentences, dividing them into syntax fragments, generating event knowledge instances based on preset event knowledge construction, and merging related events to create a structured knowledge base that supports inference and calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional entity knowledge base is used to store knowledge, then knowledge representation is refined with clear attributes, but event-related information and in-depth knowledge queries cannot be answered
Solution Approach 1:
The patent segments event knowledge into distinct components: event triggers (verbs), participants (entities with roles), and attributes (time, location, quantity). This segmentation allows the system to capture event-specific information that cannot be represented in traditional entity-centered knowledge bases, enabling answers to event-related queries while maintaining refined knowledge representation.
Solution Approach 2:
The patent introduces a new dimension to knowledge representation by adding event-type classifications and temporal-spatial attributes. This transforms the traditional flat entity-attribute structure into a multi-dimensional event knowledge structure that includes event triggers, participants with roles, time periods, locations, and quantities, thereby enabling versatile event-related queries.
2Device complexity
If entity-centered knowledge structure is adopted, then knowledge organization is simplified, but details of event relationships are lost
Solution Approach 1:
The patent segments event information into structured components (triggers, participants, attributes) that can be systematically organized. This segmentation maintains manageable complexity while preserving detailed event relationships, as each component can be independently processed and linked through standardized schemas.
Solution Approach 2:
The patent changes the organizational parameters from entity-centric to event-centric, introducing new parameters such as event triggers, participant roles, time periods, and locations. This parameter transformation enables detailed event relationship representation while maintaining systematic organization through standardized event schemas.
3Ease of operation
If traditional search methods are used, then simple factual queries can be answered, but in-depth knowledge search yields no results
Solution Approach 1:
The patent creates a universal event knowledge representation that serves multiple query types. The standardized event structure with triggers, participants, and attributes can handle both simple factual queries and complex event-related queries, making the search system multi-functional without sacrificing ease of operation for simple queries.
Solution Approach 2:
The patent adds event-type and attribute dimensions to the knowledge structure, enabling the system to handle in-depth knowledge searches. The enhanced structure preserves compatibility with simple queries while adding capability for complex event-related searches through additional organizational dimensions.
Data Source
AI summary
Proposed are a method and device for constructing an event knowledge. The method comprises: identifying text to obtain an event mining candidate sentence; dividing the event mining candidate sentence into syntax fragments; generating an event knowledge instance according to the syntax fragments and a preset event knowledge construction, in which the number of the event knowledge instances is equal to the number of verb-object fragments and subject-predicate fragments in the syntax fragments; obtaining an event mining target sentence according to the verb-object fragments and the subject-predicate fragments in the syntax fragments, dividing the event mining target sentence, and writing divided members into an event knowledge instance correspondingly, so as to accomplish construction of the event knowledge base.