Flow Diagram Recognition Using Semantic Node-Edge Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional machine learning models struggle with accurately recognizing flow diagrams due to unclear start and end points of edges and intersections, leading to misrecognition and incorrect inference of process flows.
Innovation Solution
A method using a dialogue system with a multimodal machine learning model that performs chain-of-thought reasoning, where the system first estimates nodes and their meanings, then estimates edges based on these meanings, and uses supplementary information like legend diagrams or shape classification to improve recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a conventional machine learning model is used to recognize flow diagrams, then the recognition process is simple and fast, but the accuracy is low due to unclear edge start/end points and intersections
Solution Approach 1:
The patent segments the flow diagram recognition into distinct components: node recognition, edge detection, and relationship inference. By dividing the recognition task into separate processing stages with dedicated machine learning models for each component, the system achieves higher overall accuracy while maintaining manageable complexity through modular architecture.
Solution Approach 2:
The patent applies preliminary action by first performing node detection and semantic understanding before edge recognition. The system pre-processes the diagram by identifying nodes, their meanings, and relationships, then uses this preliminary information to guide subsequent edge detection, thereby improving accuracy in resolving ambiguous intersections and connections.
2Reliability
If the system performs comprehensive semantic estimation and chain-of-thought reasoning, then the reliability of inference improves, but the processing time increases
Solution Approach 1:
The system performs preliminary semantic estimation of nodes and their meanings before conducting full chain-of-thought reasoning on relationships. This preliminary action prepares the data in advance, allowing the reasoning process to focus on relationship inference rather than basic element identification, thereby improving reliability while reducing overall processing time.
Solution Approach 2:
The patent segments the reasoning process into distinct phases: semantic understanding of individual elements, then relationship inference between elements. By dividing the comprehensive analysis into separate, specialized processing steps rather than a single monolithic process, the system maintains high inference reliability while optimizing processing efficiency through targeted computation at each stage.
Data Source
AI summary
An information processing apparatus inputs, to a dialogue system using a machine learning model, a first directive instructing estimation of a plurality of first information items based on data including the plurality of first information items and a plurality of second information items representing the relationships between the plurality of first information items. The information processing apparatus inputs, to the dialogue system, a second directive instructing semantic estimation of the first information items based on third information related to the meaning of the first information items. The information processing apparatus inputs, to the dialogue system, a third directive instructing estimation of the second information items based on the data. The information processing apparatus acquires, from the dialogue system, output information generated based on the second information items estimated in response to the third directive.