Information Extraction Using Node-Based Parsing and Rule Synthesis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in effectively extracting structured data from unstructured Web data, as most Web data is in an unstructured format, making it difficult for application programs to understand and utilize.
Innovation Solution
An information extraction method and system that parse unstructured text data into nodes based on a node format, using a rule database to generate structured data, where nodes are categorized as root or non-root nodes, allowing for the synthesis of structured information without the need for extensive tagging and model training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If unstructured text data is converted to structured data using traditional methods (extensive tagging and model training), then extraction accuracy is improved, but system cost and complexity increase significantly
Solution Approach 1:
The patent segments the text processing task into discrete node operations. Each node represents a specific text element with defined attributes, allowing systematic processing without requiring complex global models. The text is divided into extractable nodes based on simple syntactic rules rather than comprehensive semantic analysis.
Solution Approach 2:
The patent changes the approach from semantic parameter analysis (requiring extensive training) to syntactic parameter analysis. By focusing on structural parameters like node type, position, and basic grammatical relationships, the system achieves extraction with simpler, more cost-effective rules.
2Loss of information
If unstructured text data is processed using comprehensive tagging and extraction model training, then information completeness is improved, but processing time and resource consumption increase
Solution Approach 1:
The patent performs preliminary text preprocessing to identify and mark potential extractable nodes before detailed extraction. This preliminary action organizes the text into a node structure that facilitates efficient subsequent processing, reducing the time needed for comprehensive information extraction.
Solution Approach 2:
The patent creates node representations that copy essential structural and semantic information from the original text in a standardized format. These node copies can be processed independently and efficiently, maintaining information completeness while reducing processing complexity and time.
3Device complexity
If simple node-based parsing is used without extensive tagging, then system cost is reduced, but extraction precision may deteriorate
Solution Approach 1:
The patent creates a universal node structure that can handle multiple extraction tasks through a single standardized framework. Each node type is designed to be multi-functional, capable of representing different text elements (entities, events, relationships) with consistent attributes, maintaining precision across diverse extraction scenarios without requiring task-specific complex models.
Data Source
AI summary
The present disclosure provides an information extraction method. The method includes obtaining an unstructured text data; parsing the unstructured text data based on a node format; and generating a first node set that comprises nodes describing the unstructured text data. The method further includes obtaining a rule database, the rule database comprising a plurality of rules used for generating a node, each rule designating a role of a node generated by using the rule, and the role of the node being a root node or a non-root node; synthesizing the nodes in the first node set based on the rule database to generate a root node; and generating structured data based on the root node.