Reduced QWERTY Text Disambiguation Using Learned Context
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Handheld electronic devices with reduced keyboards face challenges in text entry due to ambiguous inputs, as multiple letters, symbols, and digits are assigned to a single key, requiring effective disambiguation methods to improve usability and mimic the QWERTY keyboard experience.
Innovation Solution
A handheld electronic device with a disambiguation function that uses contextual data and a reduced QWERTY keyboard layout, employing a processor, memory, and input apparatus to generate permutations of key actuations, prioritize word objects based on frequency, and learn user-defined words, providing a customizable and efficient text entry experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If a reduced keyboard is used to enable text entry in a compact device, then the device can be made smaller and more portable, but the input becomes ambiguous as multiple letters are assigned to each key
Solution Approach 1:
The system pre-generates all possible word permutations from ambiguous key inputs before user selection. When a user presses a sequence of keys, the processor immediately creates and displays multiple candidate words that could result from different interpretations of those key presses, allowing the user to select the intended word without resolving the ambiguity through additional keystrokes.
Solution Approach 2:
The disambiguation system provides feedback by displaying multiple candidate words to the user based on the ambiguous input sequence. The user's selection of one candidate word provides feedback that confirms the correct interpretation, and this feedback is used to update the contextual data for future disambiguation decisions, improving accuracy over time.
2Volume of moving object
If multiple letters are assigned to each key to reduce keyboard size, then the device becomes more compact, but the ease of operation decreases due to ambiguous inputs requiring disambiguation
Solution Approach 1:
The system pre-generates all possible word permutations from ambiguous key inputs before user selection. When a user presses a sequence of keys, the processor immediately creates and displays multiple candidate words that could result from different interpretations of those key presses, allowing the user to select the intended word without resolving the ambiguity through additional keystrokes.
Solution Approach 2:
The disambiguation system automatically generates candidate words and presents them to the user without requiring additional input or clarification. The system serves itself by using the ambiguous input sequence to generate meaningful word candidates, reducing the operational burden on the user compared to traditional multi-tap or chord-based systems.
3Measurement precision
If contextual data is used to improve disambiguation accuracy, then text entry accuracy improves, but the device complexity increases
Solution Approach 1:
Contextual data including word frequencies, bigram probabilities, and trigram probabilities are pre-calculated and stored in the device before operation. This preliminary preparation allows the disambiguation system to quickly consult contextual information without performing complex real-time calculations, reducing the computational burden and device complexity during actual text entry.
Solution Approach 2:
The system uses a hierarchical approach to contextual analysis, starting with bigram probabilities and only employing more complex trigram analysis when necessary. This partial use of computational resources balances accuracy with device complexity, applying the full extent of contextual analysis only when simpler methods are insufficient for disambiguation.
Data Source
AI summary
A handheld electronic device includes a reduced QWERTY keyboard and is enabled with disambiguation software that is operable to disambiguate text input. In addition to identifying and outputting representations of language objects that are stored in the memory and that correspond with a text input, the device is able to employ contextual data in certain circumstances to prioritize output and to learn new contextual data.