Handheld Device Segment Learning Routine for Text Input

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for inputting text in languages like Standard Mandarin using Latin keyboards, such as pinyin Chinese, face challenges in accurately determining the intended character when multiple characters correspond to a single pin, leading to difficulties in generating the correct character interpretation.

Innovation Solution

A handheld electronic device with a processor and memory that employs a segment learning routine, using a combination of generic and learned segments, along with algorithms like the Maximum Matching Algorithm, to improve text input by learning and storing user preferences, thereby providing a more customized character interpretation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If the simple maximum matching algorithm is used to generate character interpretation, then the processing speed is improved, but the accuracy of character interpretation deteriorates when multiple characters correspond to a single pin

Engineering Contradiction:

Improveprocessing speedVSAvoidcharacter interpretation accuracy

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The patent segments the character interpretation process into multiple stages: initial rapid matching using simple maximum matching algorithm, followed by selective refinement using learned segments and user feedback. This segmentation allows the system to maintain speed for common cases while improving accuracy when needed, resolving the contradiction between processing speed and interpretation accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements feedback mechanisms where users can correct misinterpreted characters, and the system learns from these corrections to improve future interpretations. The learned segments are stored and applied to subsequent input, gradually improving accuracy without significantly impacting processing speed for most common cases.

Inventive Principle:

Principle #23Feedback

2Measurement precision

If more algorithms and linguistic data are employed to improve character interpretation accuracy, then the accuracy is improved, but the device complexity increases

Engineering Contradiction:

Improvecharacter interpretation accuracyVSAvoidalgorithm complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies partial action by implementing multiple algorithms and linguistic data structures only when necessary. The simple maximum matching algorithm handles the majority of cases efficiently, while more complex algorithms and user feedback mechanisms are activated only when the simple algorithm produces ambiguous or incorrect results, thus avoiding unnecessary complexity for common operations.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system performs preliminary action by pre-loading linguistic data and generic segments into memory before text input occurs. This allows the complex algorithms to operate more efficiently during actual text input, reducing the computational burden and perceived complexity during user interaction.

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If a Latin keyboard is used for inputting non-Latin languages, then the ease of operation is improved, but the difficulty of determining the intended character increases

Engineering Contradiction:

Improvetext input easeVSAvoidcharacter determination difficulty

Core Design Contradiction:

Ease of operationVSDifficulty of detecting and measuring

Solution Approach 1:

The patent introduces an intermediary layer of learned segments and linguistic data that mediates between the simple Latin keyboard input and the complex Chinese character output. This intermediary layer processes the ambiguous pin inputs through multiple algorithms and user feedback to resolve the difficulty of determining intended characters, maintaining the ease of Latin keyboard operation while addressing the character determination challenge.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS8296679B2Method of learning character segments from received text, and associated handheld electronic device

Publication Date: 2012.10.23 MALIKIE INNOVATIONS LTD

AI summary

An improved method of learning character segments from received text enables facilitated text input on an improved handheld electronic device. In receiving text on the handheld electronic device, the characters of the text are converted into the inputs with which the characters correspond. Then, segments and other objects are analyzed to generate a proposed character interpretation of the series of inputs. Responsive to detecting that at least a portion of the character interpretation differs from a corresponding portion of the received text, a character learning string comprising the differing characters of the received text are stored as a candidate. In response to receiving additional text on the handheld electronic device, the characters of the additional text are converted into the inputs with which the characters correspond. Then, segments and other objects are analyzed to generate another proposed character interpretation of the series of additional inputs. Responsive to detecting that at least a portion of the another character interpretation differs from a corresponding portion of the additional received text, another character learning string comprising the differing characters of the additional received text are compared with the candidate. If a set of characters in the another character learning string match characters in the candidate, the set of characters are stored as a segment.