Ordered Tree Structure for Efficient Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional natural language analysis processing methods face inefficiencies in computational efficiency and memory usage, particularly when handling text information with regular expressions, due to the need for backtracking and increased memory usage during pattern matching.
Innovation Solution
An information processing apparatus and method that generates an ordered tree structure with additional transition links defined by set relations between transition conditions, allowing efficient pattern matching by identifying indeterminate constraints and minimizing memory usage through a deterministic finite automaton configuration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If NFA to DFA conversion is applied to improve computational efficiency, then pattern matching speed increases, but memory usage increases significantly
Solution Approach 1:
The patent segments the pattern matching process into two phases: pre-processing phase where patterns are compiled into a trie structure, and execution phase where a simplified automaton processes the input text. This segmentation allows the system to use less memory during execution while maintaining efficient processing speed.
Solution Approach 2:
The patent performs preliminary compilation of patterns into a trie structure during a pre-processing phase, storing transition information in a compact format. This preliminary action enables the execution phase to use a simplified automaton that requires less memory while maintaining the computational efficiency of DFA-like processing.
2Quantity of substance
If backtracking algorithm is used for pattern matching, then memory usage is reduced, but computational efficiency deteriorates
Solution Approach 1:
The patent employs a dynamic automaton that combines features of both NFA and DFA. The automaton uses trie structure for pattern storage and implements transitions that avoid full backtracking while maintaining memory efficiency. The system dynamically adjusts the matching process based on the input text and pattern structure.
Solution Approach 2:
The patent introduces an intermediate automaton structure that mediates between the trie pattern storage and the input text processing. This intermediate structure enables efficient pattern matching without requiring full backtracking, achieving a balance between memory usage and computational efficiency.
3Reliability
If constraint array pattern is used for syntactic dependency extraction, then semantic information extraction is achieved, but processing complexity increases
Solution Approach 1:
The patent creates a universal pattern matching framework that handles multiple types of constraints (word strings, regular expressions, part-of-speech tags) within a single automaton structure. This multi-functionality reduces processing complexity by unifying different constraint types into one consistent processing mechanism.
Solution Approach 2:
The patent changes the representation parameters of constraints by converting them into transition conditions of the automaton. Word strings, regular expressions, and part-of-speech tags are all transformed into standardized transition conditions that can be processed uniformly, reducing the complexity of handling different constraint types.
Data Source
AI summary
An information processing apparatus according to the present invention comprises a generation section configured to generate an ordered tree structure by defining transition edges between nodes using, as transition conditions, respective constraints from one or more constraint pattern each including plural constraints, a search section configured to search for a second substructure K similar to a first substructure P from a root node by determining a set relation between transition conditions of respective transition edges, and a transition defining section configured to define an additional transition link from a tail node sKn of the second substructure K to a child node sPn+1 of the first substructure P, the additional transition link adding a constraint to be met by an indeterminant identified from the set relation, wherein a data structure for pattern matching is defined.