Fault Knowledge Graph Construction for Non-Standard Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for constructing fault knowledge graphs in telecommunication networks struggle with recognizing non-standard signaling in natural language fault determining texts, leading to low success rates in converting these texts into logical expressions.
Innovation Solution
A method is provided to search for corresponding protocol signaling and atomic procedures based on signaling character strings in fault determining texts, using an atomic procedure graph to standardize the process, improve recognition success rates, and convert the texts into logical expressions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If natural language fault determining texts are processed using entity dictionary and syntax analysis, then the construction process can be automated, but the recognition success rate of non-standard signaling is low
Solution Approach 1:
The patent introduces an atomic procedure graph as an intermediary structure between the natural language fault determining texts and the entity dictionary-based processing system. The graph contains standardized protocol signaling and atomic procedures that serve as a bridge, enabling the system to recognize non-standard signaling by matching against the standardized atomic procedures in the graph rather than relying solely on entity dictionary matching.
Solution Approach 2:
The patent transforms the processing approach by changing the parameters of matching: instead of matching natural language entities directly against entity dictionaries, the system extracts signaling character strings and matches them against standardized signaling in the atomic procedure graph. This parameter change from entity-based matching to signaling-based matching improves recognition accuracy.
2Ease of manufacture
If entity dictionary and syntax analysis are used for processing fault determining texts, then the method is simple to implement, but a large quantity of fault determining texts cannot be converted into logical expressions
Solution Approach 1:
The patent segments the fault determining text processing into distinct stages: extracting signaling character strings, matching against atomic procedure graph, identifying atomic procedures, and generating logical expressions. This segmentation allows each stage to be optimized independently, improving overall conversion success rate while maintaining implementability.
Solution Approach 2:
The patent performs preliminary action by pre-building the atomic procedure graph with standardized protocol signaling and atomic procedures before processing fault determining texts. This pre-prepared knowledge structure enables more effective matching and conversion of texts into logical expressions, increasing productivity.
Data Source
Figure 1~2
Figure 3
Figure 4~6
AI summary
A method for constructing a fault knowledge graph is provided. The method includes: after a to-be-processed sentence is obtained based on a fault determining text, determining protocol signaling and an atomic procedure of the to-be-processed sentence from an atomic procedure graph based on a signaling character string of the to-be-processed sentence; after an atomic procedure clause is generated based on the atomic procedure and the to-be-processed sentence, determining a logical expression corresponding to the atomic procedure clause; generating a logical expression of the to-be-processed sentence based on the logical expression of the atomic procedure clause; generating a logical expression of the fault determining text based on the logical expression of the to-be-processed sentence; and adding the logical expression of the fault determining text to a fault knowledge graph. According to the method, an atomic procedure graph can be searched for corresponding protocol signaling and a corresponding atomic procedure based on a signaling character string of a fault determining text, so that accuracy of recognizing signaling can be improved, and a success rate of converting the fault determining text into a logical expression is improved. This application further provides a computing apparatus that can implement the foregoing method.