Trouble Report Graph Processing for Cause Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately processing trouble reports due to fluctuations in expression and description among maintenance personnel, leading to low precision in identifying trouble causes in industrial products.
Innovation Solution
A trouble document processing device and method that utilizes a named entity extracting process, relationship extracting process, data matching process, and graph integrating process to create a trouble knowledge database by collating and updating information from trouble reports, using a hardware knowledge database and a language processing model to standardize expressions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data matching is performed based on co-occurrence relation of failure expression and procedure expression, then knowledge of failures can be created, but the number of combinations becomes enormous and desired result cannot be obtained
Solution Approach 1:
The patent extracts and prioritizes only the most relevant co-occurring expression pairs from the enormous combination space. By identifying and focusing on high-probability pairs between failure expressions and procedure expressions, the system filters out redundant combinations and maintains only essential knowledge relationships, thereby reducing processing complexity while preserving accuracy.
Solution Approach 2:
The patent performs preliminary classification and grouping of expressions before data matching. By pre-organizing failure expressions and procedure expressions into categories and hierarchies, the system reduces the search space and establishes a structured framework that guides subsequent matching operations, preventing combinatorial explosion.
2Adaptability or versatility
If language processing model is used to extract information from trouble reports, then variations in expression can be handled, but the model does not always learn knowledge regarding parts names and phenomena of industrial products
Solution Approach 1:
The patent creates a multi-functional expression mapping system that serves both general language understanding and specific industrial knowledge representation. The same processing framework handles diverse expression variations while simultaneously mapping them to standardized parts names and phenomena, enabling the system to be both versatile and precise.
Solution Approach 2:
The patent introduces an intermediary layer between general language processing and specific industrial knowledge. This intermediate mapping layer translates varied expressions into standardized internal representations, allowing the system to leverage general language models while achieving domain-specific precision through the intermediary knowledge base.
3Ease of manufacture
If trouble reports are processed using stereotypical rule base or word dictionary, then processing is simple, but expressions vary among maintenance personnels and processing is difficult
Solution Approach 1:
The patent transitions from static rule-based processing to dynamic expression mapping. Instead of fixed rules, the system dynamically adapts to various expressions by learning from examples and adjusting its mapping relationships, enabling it to handle diverse maintenance personnel expressions while maintaining systematic processing.
Solution Approach 2:
The patent changes the fundamental parameters of expression representation from fixed categorical labels to flexible semantic mappings. By representing expressions as mappable concepts rather than fixed categories, the system can accommodate variations in wording while maintaining processing consistency through parameter transformation rather than rule matching.
Data Source
AI summary
To create a trouble knowledge database which can be used for trouble cause diagnosis by extracting trouble information by a uniform expression from a trouble report sentence including fluctuations in expression.A trouble document processing device includes a trouble knowledge database storing information regarding a trouble, a hardware knowledge database storing a name of a part in a product, its synonym, and design information, an input/output unit inputting and outputting information, and a processor unit performing a predetermined computing process. The processor unit performs a named entity extracting process of extracting a term related to a component of a target product and a trouble of the component from a trouble report sentence, a relationship extracting process of extracting a relationship between named entities from the trouble report sentence, a named entity and relationship integrating process of generating first graph data from a result of the named entity extracting process and a result of the relationship extracting process, a data matching process of generating second graph data obtained by collating a named entity in the first graph data with the hardware knowledge database and correcting a synonymous element, a graph integrating process of generating third graph data by connecting the second graph data and the named entity in the trouble knowledge database, and a trouble knowledge database updating process of collating the third graph data and the trouble knowledge database and updating the trouble knowledge database.