Multi-word Autocorrect Contextual Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autocorrect systems on touch-sensitive devices often yield poor results when correcting text input, as they lack consideration of contextual information beyond individual words, leading to suboptimal corrections.
Innovation Solution
Implementing a multi-word autocorrect method that revisits initial corrections based on subsequently typed words, utilizing a dictionary, grammar rules, and context information to select more appropriate corrections, such as replacing 'new york' with 'New York' or 'tech ology' with 'technology', and handling mistakenly typed delimiter characters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional single-word autocorrect is used, then the system is simple and fast, but the correction accuracy is poor due to lack of contextual information
Solution Approach 1:
The patent transitions from single-word correction to multi-word phrase correction by adding the dimension of contextual information. The system now operates on sequences of words rather than isolated words, utilizing n-gram models to capture contextual relationships between adjacent words. This dimensional expansion enables the system to leverage contextual information for improved correction accuracy while managing complexity through efficient probabilistic modeling.
Solution Approach 2:
The system pre-computes and stores n-gram probabilities and contextual information in advance before actual correction is needed. By preparing language models, phrase databases, and contextual patterns beforehand, the system reduces real-time computational complexity while maintaining high correction accuracy. This preliminary action allows the correction algorithm to make rapid decisions based on pre-analyzed contextual patterns.
2Measurement precision
If multi-word autocorrect with contextual analysis is implemented, then correction accuracy improves, but processing time increases
Solution Approach 1:
The system applies partial contextual analysis by focusing on n-gram sequences of specific lengths (bigrams, trigrams, etc.) rather than analyzing entire sentences or unlimited contextual windows. This selective approach to contextual analysis provides sufficient correction accuracy while limiting processing time by avoiding excessive computational overhead from analyzing too much contextual information.
Solution Approach 2:
Language models and n-gram probabilities are pre-computed and cached before runtime. The system prepares contextual patterns, phrase databases, and statistical models in advance, allowing the correction algorithm to query pre-analyzed data during actual text correction. This eliminates the need for expensive real-time computations while maintaining high correction accuracy.
3Reliability
If the system analyzes more contextual information, then it can make better correction decisions, but the computational load increases
Solution Approach 1:
The system analyzes contextual information at partial depth by using fixed-length n-gram sequences rather than performing deep semantic analysis or unlimited contextual exploration. This approach achieves sufficient correction reliability by capturing local contextual patterns while avoiding the excessive computational energy required for comprehensive linguistic analysis.
Solution Approach 2:
The system uses lightweight probabilistic models and simple n-gram calculations instead of computationally intensive deep learning or complex linguistic parsers. These simpler computational models consume less energy while providing adequate correction reliability for typical mobile text input scenarios, trading off some theoretical maximum accuracy for practical energy efficiency.
Data Source
AI summary
Methods and systems of multi-word automatic correction (“autocorrect”) are provided. Autocorrect generally can select a corrected word based on a typed word and a dictionary of correctly-spelled words. Multi-word autocorrect can add to this functionality by revisiting the selection of an initial corrected word if a subsequently-typed word indicates that it would be more appropriate to instead select an additional corrected word. In some cases, an autocorrect system can make a multi-word correction based on a multi-word phrase in a dictionary, such as replacing “new york” with “New York” as described above. In other cases, an autocorrect system can make a multi-word correction to correct a mistakenly-typed delimiter character. In other cases, an autocorrect system can use grammar rules to obtain additional context information with each subsequently-typed word and make multi-word corrections on that basis.