Adaptive Dictionary Generation for Predictive Text Input
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile devices face challenges in text input due to limited keypads, with existing dictionaries being incomplete, particularly lacking proper nouns and slang, requiring users to manually add common words, which is inefficient.
Innovation Solution
A method to combine personal dictionaries from multiple users into an adapted dictionary, using statistical analysis and weight calculations to include frequently used words, which can be transmitted to mobile devices for predictive text entry, reducing the need for manual word entry and improving dictionary management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If users manually add custom words to the dictionary, then the dictionary becomes more comprehensive, but user effort and time increase significantly
Solution Approach 1:
The system automatically collects words from multiple personal dictionaries and uses statistical analysis to identify frequently used words, eliminating the need for users to manually add words. The server performs self-service by autonomously updating the adapted dictionary based on aggregated data from multiple sources.
Solution Approach 2:
The system uses statistical analysis of word usage frequency from multiple personal dictionaries to automatically determine which words should be added to the adapted dictionary. This feedback mechanism ensures that commonly used words are prioritized without requiring manual user input.
2Reliability
If a comprehensive dictionary is provided, then predictive text accuracy improves, but device memory requirements increase
Solution Approach 1:
Instead of providing a single comprehensive dictionary to all devices, the system creates adapted dictionaries tailored to specific user groups based on their actual usage patterns. Each adapted dictionary contains only the words relevant to that particular user group, optimizing memory usage while maintaining predictive accuracy for local users.
Solution Approach 2:
The system segments the comprehensive dictionary into multiple adapted dictionaries, each targeted at specific user groups. This segmentation allows devices to store smaller, more relevant dictionary subsets rather than a single large comprehensive dictionary, reducing memory requirements while maintaining predictive text accuracy for each user group.
3Adaptability or versatility
If personal dictionaries from multiple users are combined, then language representation improves, but dictionary processing complexity increases
Solution Approach 1:
The server acts as an intermediary that handles the complex task of combining and processing multiple personal dictionaries. It performs statistical analysis to identify frequently used words and generates adapted dictionaries, relieving individual devices from the complexity of processing multiple source dictionaries while improving language representation.
Solution Approach 2:
Instead of having each device process and store multiple personal dictionaries, the system creates simplified copies (adapted dictionaries) that contain only the essential frequently used words. These copied dictionaries maintain the benefits of combined language representation while reducing processing complexity on individual devices.
Data Source
AI summary
In a system, a server controller is configured to receive at least part of at least one personal dictionary, and is further configured to generate an adapted dictionary using words of the at least part of at least one personal dictionary. The server controller is also configured to transmit at least part of the adapted dictionary to the apparatus configured to use the adapted dictionary during predictive text entry. The apparatus includes an apparatus controller that is configured to receive at least part of an adapted dictionary from the server. The apparatus controller is also configured to utilize at least part of an adapted dictionary when predictive text entry is performed.