Input Method Cursor Context Candidate Queue
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing input methods fail to provide a reliable enter-on-screen candidate word queue when the input cursor moves, as they rely solely on the last entry on the screen and lack text information before and after the cursor.
Innovation Solution
The method acquires text information at the input cursor position, extracts keywords, and searches through associative candidate lexicons to determine language models, merging candidate words based on distance relationships and application properties to generate a reliable enter-on-screen candidate word queue.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the input method relies solely on the last entry on screen to provide enter-on-screen candidate words, then the system complexity is reduced, but the reliability of the candidate word queue deteriorates when the input cursor moves
Solution Approach 1:
The patent segments the text information into three parts relative to the input cursor: text before the cursor, text after the cursor, and the last entry on screen. This segmentation allows the system to selectively use different text segments based on cursor position, providing reliable candidate words even when the cursor moves, while maintaining manageable system complexity through structured information processing.
2Ease of operation
If the input method uses only the last entry on screen for association, then the ease of operation is improved, but the accuracy of understanding user input intent deteriorates
Solution Approach 1:
The patent merges multiple text sources (text before cursor, text after cursor, and last entry on screen) into a comprehensive text information set for association. This merging enables the system to accurately understand user input intent by considering the complete contextual environment, while maintaining ease of operation through automated multi-source integration without requiring additional user effort.
3Speed
If the input method does not consider text information before and after the cursor, then the processing speed is improved, but the comprehensiveness of the candidate word queue deteriorates
Solution Approach 1:
The patent performs preliminary action by pre-acquiring and storing text information from multiple sources (before cursor, after cursor, last entry) before the actual candidate word generation process. This preliminary preparation enables the system to quickly generate comprehensive candidate word queues without sacrificing processing speed, as the text information is already available when needed for association.
Data Source
AI summary
An input method, device and electronic apparatus are provided. The input method includes acquiring text information at an input cursor position, where the text information includes above text information located before the input cursor and/or below text information located after the input cursor; extracting keywords from the text information; searching through associative candidate lexicons of the keywords to obtain an enter-on-screen candidate word queue at the input cursor position; outputting the enter-on-screen candidate word queue. By acquiring the text information at the input cursor position and determining the enter-on-screen candidate word queue based on the keywords in the text information, embodiments of the present disclosure solve the issue in existing techniques that after the input cursor changes it position, no enter-on-screen candidate word may be provided by association because no reliable enter-on-screen entry is obtained.