Error Revision via Correlation Scoring for Character Strings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional error revision methods on mobile devices and PCs are inefficient, as they either delete correct characters along with errors or require cumbersome cursor movement, and existing automatic revising functions lack accuracy due to limited libraries and failure to consider user typing habits and partial string correlations.
Innovation Solution
A method that calculates a correlation score between the intended correction and previously inputted character strings, using a weighted scoring system to accurately identify and replace erroneous characters, considering user-specific typing habits and partial string correlations, allowing for precise error correction without re-inputting entire words.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a delete function key like Backspace is used to revise an error, then the erroneous character can be deleted, but all characters between the erroneous character and the current cursor position are also deleted
Solution Approach 1:
The patent segments the text editing operation into two independent parts: (1) selecting the erroneous character through cursor movement to its position, and (2) deleting only that specific character. This segmentation allows the user to precisely target the error without affecting other characters, resolving the contradiction between error correction precision and preservation of correct characters.
2Measurement precision
If a mouse is used to move the cursor to an erroneous character, then the cursor can be positioned accurately, but the user must take their hand off the keyboard and perform multiple actions
Solution Approach 1:
The patent implements a self-service mechanism where the system automatically detects and highlights the erroneous character based on user input patterns and context analysis. Instead of requiring the user to manually position the cursor with a mouse, the system serves itself by identifying the error location and preparing it for correction, thereby maintaining positioning precision while greatly improving operational convenience.
Solution Approach 2:
The patent replaces the mechanical cursor movement operation (using mouse to move cursor) with an automated information processing system that detects errors through text analysis and pattern recognition. This substitution eliminates the need for manual cursor positioning while maintaining accurate error location identification.
3Ease of operation
If a finger touch is used to move the cursor on a small-sized touch screen, then the cursor can be moved, but it is difficult and troublesome to move the cursor to a precise position
Solution Approach 1:
The patent implements a self-service mechanism where the system automatically detects and highlights the erroneous character based on user input patterns and context analysis. Instead of requiring the user to manually position the cursor with a mouse, the system serves itself by identifying the error location and preparing it for correction, thereby maintaining positioning precision while greatly improving operational convenience.
4Extent of automation
If an automatic revising function is used with a limited library, then word revision can be automated, but the accuracy is low and wrong words may be inputted
Solution Approach 1:
The patent implements a feedback mechanism where the system presents multiple revision candidates ranked by correlation score and waits for user confirmation before applying the correction. This feedback loop allows the user to verify the intended revision, preventing wrong word inputting while maintaining automation. The system learns from user corrections to improve future automatic revision accuracy.
Solution Approach 2:
The patent changes the parameter of library completeness by dynamically expanding the correction library through user feedback and learning. Instead of relying on a static limited library, the system accumulates user-specific correction patterns and terminology, thereby improving revision accuracy over time while maintaining automatic operation.
5Extent of automation
If the conventional automatic revising function presumes what the user might have inputted, then revision can be automated, but if the presumption accuracy is low, a wrong word that the user never intended to input is actually inputted
Solution Approach 1:
The patent implements a feedback mechanism where the system presents multiple revision candidates ranked by correlation score and waits for user confirmation before applying the correction. This feedback loop allows the user to verify the intended revision, preventing wrong word inputting while maintaining automation. The system learns from user corrections to improve future automatic revision accuracy.
Data Source
AI summary
The present invention relates to a method for revising an error, wherein a user inputs a desired character string for revision so as to calculate a correlation between the desired character string for revision and previously inputted character strings, so that a character string with a high correlation is replaced with the desired character string for revision. A characteristic configuration of the present invention is in a step for deciding a correlation. According to the present invention, in the step for deciding a correlation, a plurality of error-revising operations are defined, an error-revising operation score is given to each of the plurality of error-revising operations, a total of scores are calculated in such a way that scores are summed up for each of the error-revising operations required for revising a previously inputted character string into a desired character string for revision, wherein, if the number of cases of error-revising operations selectable for revision are plural, a score of a case in which a total of scores become a minimum is used for deciding a correlation. A total of scores calculated in such a way are compared with a predetermined threshold, thereby implementing a correlation decision.