Adaptive Identifier Detection via Dynamic Rule Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rule-based information extraction systems become nonviable due to changes in the structure of incoming messages, which can be sudden and unannounced, rendering obsolete once-useful rules without warning.
Innovation Solution
An identifier detection system is updated based on detected changes in incoming message structure over time, using shared identifiers to filter messages, calculate characterizing values, and determine whether to generate or modify extraction rules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If rule-based information extraction is used for message processing, then extraction accuracy is maintained for known message structures, but the system becomes nonviable when message structure changes occur
Solution Approach 1:
The patent implements dynamic rule generation by training language models on incoming message data to automatically adapt extraction rules when message structures change. Instead of using static predefined rules, the system continuously learns from new message formats, transforming the extraction system from rigid to adaptive while maintaining accuracy through model-based generation rather than manual rule updates
Solution Approach 2:
The system performs self-service by automatically detecting message structure changes and generating updated extraction rules without human intervention. The language model analyzes incoming messages, identifies structural changes, and autonomously creates new extraction rules, enabling the system to self-adapt to changing message formats while maintaining reliable extraction accuracy
2Adaptability or versatility
If manual rule updates are performed to adapt to message structure changes, then adaptability is improved, but processing time and operational complexity increase
Solution Approach 1:
The system eliminates manual rule update time by implementing self-service through automated language model training. The model continuously monitors incoming messages, detects structural changes, and automatically generates updated extraction rules without requiring human operators to manually analyze and update rules, thus adapting to changes in real-time without time loss
Solution Approach 2:
The system uses feedback from incoming message data to continuously improve extraction rules. By monitoring message structures and detecting changes through language model analysis, the system receives real-time feedback about structural variations and automatically adjusts rules accordingly, eliminating the time delay associated with manual detection and update cycles
3Adaptability or versatility
If language model training is performed to generate adaptive rules, then adaptability to message changes is improved, but computational resources and processing time increase
Solution Approach 1:
The system applies partial training actions by selectively training the language model only when message structure changes are detected, rather than continuously training on all incoming data. This approach uses computational resources proportionally to the need for adaptation, reducing overall energy consumption while maintaining adaptability by triggering training only when necessary based on change detection thresholds
Data Source
AI summary
A method and related system for generating new rules and patterns to detect identifiers includes operations to obtain a first set of messages comprising a set of identifiers and determining a labeled set of messages associated with a target identifier of the set of identifiers based on detected matches between the target identifier and text extracted from the first set of messages with a first construct. The method further includes determining a characterizing value based on the labeled set of messages and determining whether the set of characterizing values satisfies a set of criteria. The operations further include performing a search for the target identifier in a database of messages to retrieve a set of matching messages based on a result indicating that the set of characterizing values satisfies the set of criteria. The operations further include generating a second construct based on the set of matching messages.