Mobile Character Input Candidate Reordering via Context Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing character input systems face challenges in dynamically adjusting the display order of candidate character strings based on input states, leading to inefficiencies in operability and memory usage, particularly when handling long documents or varying input conditions.
Innovation Solution
A character input program that utilizes a conversion dictionary with attributes corresponding to input states, allowing for priority adjustments of candidate character strings based on the current input state, enabling dynamic reordering of candidates according to the input timing, mode, and expression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the score is calculated through a calculation using weight based on the frequency of various words contained in the document being created and the relative position of the words in the document for every type of field, then the candidate display can be optimized according to document context, but the load of the calculation becomes enormous
Solution Approach 1:
The patent extracts only the essential attributes needed for candidate ranking (input timing, input mode, input expression) from the complex document context, rather than calculating scores based on all words and their positions. This extraction approach maintains adaptability while dramatically reducing calculation load.
Solution Approach 2:
The patent performs preliminary classification of input states into discrete attributes (timing, mode, expression) before candidate extraction. This preliminary structuring of input state information enables efficient priority adjustment without requiring complex real-time calculations during candidate ranking.
2Adaptability or versatility
If a word dictionary table of a normal configuration is provided other than being registered with a plurality of date dictionary tables, then all words can be registered, but the memory capacity for storing the dictionary data increases
Solution Approach 1:
The patent segments the dictionary into multiple specialized dictionary tables (date dictionary tables) organized by input timing attributes, rather than using a single comprehensive dictionary. This segmentation allows the system to load only relevant dictionary portions based on current input state, reducing memory capacity while maintaining comprehensive word coverage across different contexts.
Solution Approach 2:
The patent implements dynamic dictionary selection based on input timing attributes, where the appropriate dictionary table is selected according to the current input state. This dynamic approach ensures all words are accessible when needed while minimizing the amount of dictionary data held in memory at any given time.
3Adaptability or versatility
If both the search on the date dictionary table corresponding to the date data obtained from the clock and the search on the word dictionary table need to be executed, then comprehensive word search is achieved, but the speed of the process decreases
Solution Approach 1:
The patent dynamically selects a single appropriate dictionary table based on the input timing attribute, eliminating the need to execute searches on multiple dictionary tables simultaneously. This dynamic selection maintains comprehensive search capability while significantly improving process speed by avoiding redundant searches.
Solution Approach 2:
The patent extracts and uses only the relevant timing attribute to select the appropriate dictionary table, rather than searching all available dictionary tables. This extraction approach maintains search comprehensiveness for the current context while eliminating unnecessary searches that would slow down the process.
4Productivity
If the selected results of the most recent candidate and the candidate having a high frequency selected in the past are generally displayed at a higher level, then recent and frequent candidates are prioritized, but the character string intended by the user may not be displayed at the higher level
Solution Approach 1:
The patent adds new dimensions (input timing, input mode, input expression) to the candidate ranking criteria beyond just recency and frequency. This multi-dimensional approach allows the system to prioritize candidates that match the current input context, improving user intent accuracy while maintaining conversion efficiency through comprehensive attribute matching.
Solution Approach 2:
The patent applies different ranking priorities to different candidates based on their specific attribute matches with the current input state. Rather than uniformly prioritizing recent or frequent candidates, the system adjusts priorities locally for each candidate based on how well its attributes match the current input context, ensuring the intended character string is displayed at the appropriate level.
Data Source
AI summary
This invention easily realizes a process of changing the display order of the candidates according to various input states. In a character input device of a mobile telephone, a wide variety of attributes representing an input state are set, and a conversion dictionary in which the attributes are registered in dictionary data of a corresponding word is set. An adjustment value setting unit determines the input state at a relative time point according to the start of input of the read character or the confirmation of the candidate, and sets an adjustment value to each attribute according to the determination content. A priority adjustment unit adjusts the priority with the adjustment value set to the attribute of the candidate for the candidate set with an attribute of each candidate extracted from the conversion dictionary by the read character string. A candidate display processing unit determines the display order of each candidate based on the priority after the adjustment process, and causes the display unit to display the candidate in the determined order.