Character Input Device Misinput Learning via Delete Key Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional character input technologies do not learn input characters until they are finalized, leading to inefficiencies when misinputs occur, as users must delete and reinput correct characters, failing to register appropriate conversion candidates for repeated mistakes.
Innovation Solution
A character input device with an input key determination unit, input character processing unit, and conversion candidate processing unit that learns conversion candidates by setting the character prior to deletion as the reading of the finalized input character, storing this in a history for repeated misinputs, and associating it with the correct character for conversion learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional learning technologies are used to store conversion candidates only after input finalization, then the conversion learning database remains clean and accurate, but learning cannot occur when misinputs are made and users must manually delete and reinput characters
Solution Approach 1:
The system performs preliminary learning actions by storing misinput characters and their corrections in the conversion learning database during the input process itself, rather than waiting for finalization. This allows the learning database to be populated with correction data in advance, enabling faster future conversions without waiting for manual reinput
Solution Approach 2:
The system implements feedback by detecting delete key inputs and using them to identify misinputs. When a delete operation is detected, the system retrieves the preceding character, associates it with the intended correction, and stores this correction pair in the conversion learning database, creating a self-learning feedback loop that improves future input accuracy
2Productivity
If the system learns every character input including misinputs, then learning opportunities increase, but the conversion learning database becomes polluted with incorrect character associations
Solution Approach 1:
The system uses delete key detection as a feedback mechanism to identify genuine misinputs that require learning. By triggering learning only when a delete operation occurs (indicating the user corrected an error), the system ensures that only meaningful correction pairs are stored, maintaining database accuracy while capturing learning opportunities
Solution Approach 2:
The system performs self-correction learning by automatically detecting misinputs through delete key patterns and generating appropriate conversion candidate associations without user intervention. The system serves itself by identifying its own errors and learning from them, improving accuracy without manual database management
Data Source
AI summary
A character input device is provided with an input key determination unit, an input character processing unit and a conversion candidate processing unit. The input key determination unit determines the type of a key that is input. The input character processing unit executes an operation on an input character and finalization of the input character that is based on the input key. The conversion candidate processing unit executes learning of a conversion candidate associated with the input character. When it is detected that there was input of a delete key before finalization of the input character, the conversion candidate processing unit, as learning of the conversion candidate, sets an input character existing prior to input of the delete key as a reading of the finalized input character.