Event Relationship Determination via Multi-Granularity Vector Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurately determining the relationship between events is challenging due to the complexity of causal and other logical connections, which is crucial for forecasting and risk control applications.
Innovation Solution
A method involving the generation of word sequence vectors with different granularities for each event, followed by fusion vector creation and relationship determination using these vectors, employing techniques like word embedding models and deep neural networks to capture more generalized features of events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple granularities and semantic representations are used to improve relationship determination accuracy, then the determining complexity and computational cost increase
Solution Approach 1:
The event representation is segmented into multiple granularities (word level, phrase level, sentence level) allowing the system to capture details at different levels of abstraction. This segmentation enables accurate relationship determination by comparing events at appropriate granularities while managing complexity through hierarchical processing.
Solution Approach 2:
The patent adds a granularity dimension to event representation, transforming a single-level comparison problem into a multi-level comparison problem. By representing events at word, phrase, and sentence granularities simultaneously, the system determines relationships across multiple dimensions, improving accuracy while organizing complexity structurally.
2Measurement precision
If multiple granularities and semantic representations are used to improve relationship determination accuracy, then the computational resources and time required increase
Solution Approach 1:
The patent performs preliminary action by pre-processing events into multiple granularity representations before relationship determination. Event extraction, word segmentation, and phrase identification are completed in advance, creating ready-to-use multi-level representations that can be efficiently compared during relationship determination without repeated processing.
Solution Approach 2:
The patent merges multiple granularity representations (word sequences, phrase sequences, sentence sequences) into a unified event structure. This combining allows the system to leverage information from all levels simultaneously during relationship determination, achieving high accuracy while avoiding redundant computations by integrating rather than separately processing each granularity.
Data Source
AI summary
A method for generating a relationship of events includes: obtaining a statement of a first event and a statement of a second event; generating a word sequence vector with first granularity and a word sequence vector with second granularity based on the statement of the first event; generating a word sequence vector with third granularity and a word sequence vector with fourth granularity based on the statement of the second event; generating a first fusion vector based on the word sequence vector with first granularity and the word sequence vector with second granularity; generating a second fusion vector based on the word sequence vector with third granularity and the word sequence vector with fourth granularity; and determining a relationship between the first event and the second event based on the first fusion vector and the second fusion vector.