Keyword-Pair Phrase Mapping for Richer Text Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing phrase extraction methods are limited in generating concise and logically coherent phrases for target items, often resulting in a restricted number of words and requiring further model training.
Innovation Solution
A text processing method that involves extracting keywords from a text database, establishing a keyword table with mapping relationships, determining keyword pairs with association relationships, and creating a relationship chart to expand the number of applicable phrases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If phrase extraction is performed from existing longer relevant text, then phrase corpus resources are expanded, but the number of extractable phrases is limited by the amount of existing text
Solution Approach 1:
The patent performs preliminary keyword extraction from existing texts and establishes keyword tables with mapping relationships before generating phrases. This preliminary processing enables the system to leverage existing text-phrase mapping relationships to generate additional phrases without requiring further text extraction, thus resolving the contradiction between expanding phrase corpus and limited extractable phrases
Solution Approach 2:
The patent introduces keyword tables as intermediary structures that map keywords to phrase elements. These keyword tables serve as mediators between existing texts and generated phrases, enabling the system to expand phrase corpus by establishing new mapping relationships through keyword association rather than direct text extraction
2Adaptability or versatility
If neural network model is trained to generate phrases, then phrase generation capability is improved, but model training is required and generated phrases may fail to adhere to language logic
Solution Approach 1:
The patent enables the system to self-expand its phrase corpus by automatically extracting keywords from existing texts, establishing keyword tables, and generating phrases based on keyword associations. This self-service mechanism eliminates the need for external model training while maintaining phrase generation capability and language logic adherence
Solution Approach 2:
The patent replaces the complex neural network model with a simpler keyword-based mechanical system. Instead of using deep learning mechanisms to generate phrases, the system uses keyword extraction, keyword table establishment, and keyword pair association to generate phrases, thereby reducing device complexity while maintaining functionality
Data Source
AI summary
A text processing method includes the steps described below. Multiple texts in a preset text database are acquired, and at least one keyword is extracted from each text. Based on each keyword which is extracted, a keyword table is established. A mapping relationship between each keyword and a phrase element in a phrase set corresponding to a text in which each keyword is located is the same as a mapping relationship between the text in which each keyword is located and the phrase element in the corresponding phrase set. A keyword pair having an association relationship is determined in the keyword table. A phrase element in a phrase set corresponding to each keyword in the keyword pair are updated. According to a mapping relationship between each keyword and phrase elements in an updated phrase set, a relationship chart between a keyword and a phrase is established.