Rule Intent Mining via Dependency Tree Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data mining techniques face challenges in extracting and analyzing business rules from large, noisy documents written in natural language, particularly in identifying rule intents and creating their formal representations.
Innovation Solution
A method and system that analyze dependency trees of sentences using heuristic rules to extract rule intents, saving them in Semantics of Business Vocabulary and Rules (SBVR) format, utilizing the Stanford Dependency Parser and extracting arguments with value and role attributes to mine rule intents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing text mining techniques are used to extract information from documents, then information can be extracted and stored in databases, but the ability to eliminate noise and identify rule sentences in structured/unstructured documents remains challenging
Solution Approach 1:
The patent segments the document processing into distinct stages: sentence extraction, dependency tree construction, rule intent identification, and formal representation. This segmentation allows each stage to focus on specific challenges, improving both productivity and precision by handling noise and identifying rule sentences systematically rather than monolithically.
Solution Approach 2:
The patent introduces dependency trees as an intermediary structure between raw sentences and formal rule representations. This intermediary layer helps eliminate noise and identify rule sentences by providing a structured analysis of sentence relationships, enabling more accurate rule intent extraction while maintaining processing efficiency.
2Measurement precision
If business documents are processed to extract rules, then rules can be identified, but the documents contain large amounts of noise and are written in natural language making manual extraction difficult
Solution Approach 1:
The patent replaces manual mechanical extraction processes with automated computational methods. By using dependency parsing and heuristic rules, the system automatically identifies rule sentences and extracts intents from noisy natural language documents, eliminating the need for manual processing while achieving high precision in rule extraction.
Solution Approach 2:
The patent transforms natural language documents into structured representations by changing the parameter space from raw text to dependency trees and formal rule representations. This parameter transformation simplifies the extraction process and enables automated processing while maintaining high precision through structured analysis.
3Ease of operation
If extracted rules are expressed in natural language again, then the rules can be reviewed, but they are not always easy for humans to comprehend and analyze for inconsistencies
Solution Approach 1:
The patent transforms extracted rules from natural language into formal representations using standardized vocabularies and syntax. This parameter change makes the rules easier to comprehend and analyze for inconsistencies, as the formal structure provides clear, unambiguous representations that are more suitable for automated and human analysis compared to natural language expressions.
4Measurement precision
If existing extraction techniques focus on specific document classes with predefined templates, then structured information can be extracted, but the ability to handle structured and unstructured documents uniformly remains limited
Solution Approach 1:
The patent creates a universal processing framework that handles both structured and unstructured documents through the same dependency tree-based approach. The system uses general-purpose sentence extraction and dependency parsing that work across different document types, eliminating the need for separate processing pipelines for each document class while maintaining high extraction accuracy.
Data Source
AI summary
A system and method for method and a system for mining rule intents from documents is provided, wherein the rule intents are basic atomic facts present in a sentence. The proposed method and system for mining of rule intents from documents is performed in multiple stages that include extraction of rule sentences from input documents that are further parsed to create dependency trees. The created dependency trees are further analyzing based on a set of plurality of heuristic rules to mine rule intents. The mined rule intents are saved and further displayed in Semantics of Business Vocabulary and Rules (SBVR) format. The mined rule intents that are displayed automatically in the SBVR format can be easily analyzed as they SBVR is a comprehensive standard for business rule representation by Object Management Group (OMG) in accordance with set of a standard pre-defined vocabularies.