Context-Aware Candidate Sentence Generation for Text Input Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing input methods for text editing on smart devices are inefficient when users need to input longer sentences, as they require multiple inputs and selections to complete the sentence, leading to low input efficiency.
Innovation Solution
A method and electronic device that acquire text preceding and following the current input location, determine candidate sentences based on occurrence probabilities of possible conjunctions, and connect these conjunctions to form predicted candidate sentences, which are then displayed as candidate inputs, improving input efficiency by reducing the need for multiple inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional input methods provide candidate inputs based on single words or phrases, then the input method can provide simple completion options, but the user must perform multiple inputs and selections to complete longer sentences, resulting in low input efficiency
Solution Approach 1:
The system pre-calculates and stores multiple possible conjunctions (such as 'and', 'but', 'so') that may follow the current input position, along with their occurrence probabilities. When the user inputs text, the system immediately retrieves pre-computed candidate sentences formed by combining the current input with these possible conjunctions and subsequent text, rather than waiting for user selection after each word. This preliminary preparation of completion options significantly reduces the number of interactive steps required.
Solution Approach 2:
The candidate sentence generation process is dynamic and context-aware. The system adjusts the set of possible conjunctions and candidate sentences based on the current input position, preceding text, and following text. The occurrence probabilities of different conjunctions are dynamically calculated and used to rank candidate sentences, allowing the system to adapt to different contexts and user intentions in real-time during the input process.
2Adaptability or versatility
If the input method provides candidate options for each word input, then the user can select from available options, but the candidate options do not include contextual phrases like 'outside' for 'outside now', requiring additional manual input
Solution Approach 1:
The system transitions from traditional single-word candidate completion to multi-dimensional sentence-level completion. By incorporating preceding text and following text context, the system generates candidate sentences that span multiple dimensions of linguistic context rather than just completing the current word. This dimensional expansion allows the system to provide contextual phrases like 'outside' when the user inputs 'outside now', as the candidate sentence is formed by combining the current input with pre-stored conjunctions and subsequent text from the context.
Solution Approach 2:
The system introduces an intermediary processing layer that bridges the gap between simple word completion and full sentence input. This intermediary layer analyzes the current input position, retrieves relevant preceding and following text, calculates occurrence probabilities of various conjunctions, and generates contextual candidate sentences that naturally extend the user's input. This intermediary process enables the system to understand and complete contextual phrases without requiring the user to manually type each word.
3Adaptability or versatility
If the system provides candidate inputs based on common sentences and user input records, then the system can offer personalized completion, but it fails to provide complete sentence structures for longer text input
Solution Approach 1:
The system merges three key elements: (1) common sentences from training data, (2) user's personal input records and preferences, and (3) contextual information from preceding and following text. By combining these elements, the system generates candidate sentences that are both personalized to the user and complete in structure. The merging process integrates the user's input with pre-stored conjunctions and contextual text to form coherent, contextually-appropriate candidate sentences that maintain grammatical completeness and personalization.
Data Source
AI summary
A method for determining a candidate input is provided. Text that precedes and/or follows a current text input location in an input interface is acquired. A candidate sentence is acquired based on the text which is contained in a current sentence that corresponds to the current text input location. A preset number of possible conjunctions for the candidate sentence are determined based on occurrence probabilities of the possible conjunctions that are determined for the candidate sentence. The preset number of possible conjunctions are connected to the candidate sentence to obtain predicted candidate sentences. Occurrence probabilities of the predicted candidate sentences are calculated according to the text that precedes and/or follows the current text input location. Further, a preset number of the predicted candidate sentences are provided based on the calculated occurrence probabilities of the predicted candidate sentences as candidate inputs.