Word Recommendation System for Portable Text Input
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The size constraints of portable communication devices limit the efficiency of text input due to smaller keyboards, leading to increased typing errors and user dissatisfaction.
Innovation Solution
A method and system that generate word recommendations by receiving input characters, generating permutations with neighbor characters, identifying candidate words in a dictionary, scoring them, and presenting a subset based on predefined criteria to improve text input efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If the keyboard size is reduced to fit portable devices, then the device portability is improved, but the typing accuracy deteriorates
Solution Approach 1:
The patent introduces an intermediary system (autocorrect engine) that operates between the user and the keyboard. This intermediary receives the garbled input from the small keyboard, processes it through algorithms that consider keyboard layout adjacency and word frequency, and transforms it into the intended correct word, thereby compensating for the physical limitations of the small keyboard.
Solution Approach 2:
The patent replaces the mechanical correction process (physically pressing correct keys multiple times) with an automated computational system. Instead of relying on mechanical precision from the user, the system uses software-based correction algorithms that analyze input patterns and suggest corrections, eliminating the need for precise mechanical input.
2Volume of moving object
If the keyboard keys are made smaller to fit portable devices, then the device portability is improved, but the ease of operation deteriorates
Solution Approach 1:
The patent implements a self-service correction system where the autocorrect engine automatically detects and corrects typing errors without requiring user intervention. The system monitors keyboard input, identifies errors based on adjacency graphs and word frequency databases, and automatically replaces incorrect words with correct ones, making the device serve itself rather than requiring the user to manually correct each error.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system continuously monitors keyboard input, compares it against expected patterns using adjacency graphs and word frequency data, and provides automatic correction feedback. This closed-loop system adjusts the output based on the input patterns observed, creating an adaptive correction mechanism that improves ease of operation.
3Reliability
If more correction backtracking is required due to typing mistakes, then the typing accuracy can be maintained, but the productivity deteriorates
Solution Approach 1:
The patent performs preliminary action by automatically correcting errors as they are detected during typing, rather than waiting for the user to complete their thought and then correct errors. The autocorrect engine operates in real-time during the typing process, proactively identifying and correcting errors before they affect the overall input efficiency, thereby maintaining both accuracy and productivity.
Data Source
AI summary
Word recommendations are provided in response to text input. For a particular text input, possible word recommendations are identified based on the characters of the input and corresponding neighbor characters on a keyboard layout. The possible word recommendations are scored based on how closely they match the input word on a character-by-character basis, and a subset of the possible word recommendations are selected for presentation to the user.