Dependency Parsing for Accurate Grammar-Based Sentence Search

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing sentence search technologies fail to consider grammatical structure and similarity, leading to inaccurate results in fields like language education and machine translation, while sentence segmentation methods are inefficient and user-uniform, failing to meet individual user needs.

Innovation Solution

A system utilizing dependency parsing for grammar structure analysis to provide accurate sentence search, modification relation analysis, and customizable sentence segmentation based on user-defined criteria, employing neural network-based methods for enhanced accuracy and flexibility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing sentence search technologies compare and search sentences based on words such as similarity of words included in the sentence or part of speech information, then the search process is simple and fast, but the search accuracy is insufficient because grammatical meaning or structure of the sentence cannot be considered

Engineering Contradiction:
Improvesearch accuracyVSAvoidanalysis complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments sentences into grammatical components using dependency parsing, breaking down the sentence structure into manageable units (headwords, modifiers, dependencies) that can be analyzed and compared systematically. This allows the system to consider grammatical structure without overwhelming complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the search parameters from simple word similarity to grammatical structure parameters including dependency relationships, part of speech tags, and modification relations. This transformation enables more accurate search results by comparing sentences based on their grammatical architecture rather than just lexical similarity.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If conventional modification relation analysis technologies display a modification relation centered on a main word from the beginning of a sentence, then the analysis is straightforward, but modifiers or prepositional phrases in the back may not be accurately analyzed

Engineering Contradiction:
Improvemodification relation accuracyVSAvoidprocessing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent inverts the traditional analysis approach by starting from the end of the sentence and working backwards to identify modification relations. This reverse engineering approach ensures that modifiers and prepositional phrases at the end of sentences are captured accurately, then connected to their headwords through dependency parsing.

Inventive Principle:
Principle #13The other way round (Inversion)

Solution Approach 2:

The patent introduces dependency parsing as an intermediary layer between the raw sentence and the modification relation analysis. This intermediary process identifies grammatical relationships and structures the data in a way that facilitates comprehensive modification relation detection throughout the entire sentence.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Adaptability or versatility

If constituency parsing mainly used rule-based and statistical approaches, then the method is well-established, but the rule-based approach has limitations in processing new sentence structures because it is difficult to comprehensively write rules, and the statistical approach takes too much time to process complex sentence structures

Engineering Contradiction:
Improveprocessing capabilityVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent replaces the mechanical rule-based system with a data-driven dependency parsing approach that automatically learns grammatical patterns from corpora. This substitution eliminates the need for manual rule creation while significantly reducing processing time compared to statistical methods, as the parsed structure can be efficiently stored and queried.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

4Adaptability or versatility

If segmentation results of an identical sentence are uniform according to conventional technologies, then the processing is consistent and simple, but the results may not meet the user's needs depending on user's segmentation purpose or important factors

Engineering Contradiction:
Improvecustomization flexibilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent makes the segmentation system dynamic by allowing users to specify important factors or focus areas. The dependency parsing results are then filtered and presented according to user preferences, enabling the same sentence to be segmented differently based on analytical needs while maintaining the underlying grammatical structure.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS20250348524A1System and method for analyzing sentence
Publication Date: 2025.11.13 AJOU UNIV IND ACADEMIC COOP FOUND
  • US20250348524A1 patent drawing
  • US20250348524A1 patent drawing
  • US20250348524A1 patent drawing

AI summary

A system includes: a communication interface connected to another computing device and receiving an input related to information about a grammar to be searched; and at least one processor configured to obtain a query including information about the grammar to be searched based on the received input, compare grammar structure information corresponding to the grammar included in the obtained query with sentence structure information of each sentence stored in a database, and obtain a search result including at least one sentence having the grammar structure information according to a comparison result, wherein the grammar structure information and the sentence structure information include structure information based on dependency parsing.