Predictive Text Scoring Using Typing Speed and String Length
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Character input on touchscreens, particularly in electronic devices like smartphones and tablets, is cumbersome due to the small screen size, especially when entering long messages, as existing predictive text solutions do not effectively account for typing speed and context to enhance input efficiency.
Innovation Solution
A method that uses a processor to generate predicted strings based on input, considering typing speed and context, and assigns scores to these predictions for display, allowing users to quickly and accurately input text by suggesting completions or corrections, thereby improving text input speed and reducing processing cycles while saving power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If predictive text solutions are implemented on touchscreens, then text input speed is improved, but the small touchscreen area limits the effectiveness of predictions
Solution Approach 1:
The patent changes the parameter of prediction scoring by incorporating typing speed and predicted string length into the scoring function. This allows the system to adapt predictions to the user's typing behavior, providing more accurate predictions that account for individual typing patterns and preferences, thereby improving text input efficiency on small touchscreens.
Solution Approach 2:
The system performs preliminary analysis of typing speed and contextual information before generating predictions. By pre-calculating these parameters and using them to weight prediction scores, the system prepares prediction candidates in advance, enabling faster and more accurate text completion on limited touchscreen real estate.
2Ease of operation
If existing predictive text solutions are used, then some text input assistance is provided, but they do not account for typing speed and context effectively
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors typing speed and uses this information to adjust prediction scoring in real-time. The typing speed metric serves as feedback that refines the prediction model, allowing the system to adapt to the user's current typing state and provide more accurate predictions.
Solution Approach 2:
The prediction system transitions from a static model to a dynamic one by incorporating typing speed as a variable factor. The scoring function dynamically adjusts based on the user's typing velocity, making the prediction system adaptive and responsive to changing input conditions, thereby improving prediction accuracy.
3Measurement precision
If more comprehensive prediction analysis is performed, then prediction accuracy is improved, but processing cycles increase
Solution Approach 1:
The patent extracts only the most relevant features (typing speed and predicted string length) from the available data to inform prediction scoring. By selecting and weighting only these key parameters rather than analyzing all possible contextual factors, the system achieves good prediction accuracy while minimizing processing overhead and cycle time.
4Productivity
If predictive strings are displayed based on traditional scoring, then predictions are provided, but typing speed and prediction length are not considered
Solution Approach 1:
The patent modifies the scoring function parameters to include typing speed and predicted string length as weighting factors. This parameter change enhances the sophistication of the prediction system, enabling it to account for user-specific typing characteristics and provide more personalized, accurate predictions without requiring a complete redesign of the underlying architecture.
Data Source
AI summary
A method that includes receiving an input, determining, by the processor, a likelihood that a predicted string associated with the received input matches an intended input string, where the determination is a function of at least one of a length of the predicted string and a typing speed associated with the received input, and displaying the predicted string.