Input Method Editor Dictionary Update via Search Query Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Languages using logographic scripts, such as Chinese, Japanese, and Korean, face challenges in inputting characters on standard keyboards due to the many-to-one mapping of characters to keys, leading to inefficiencies in character entry and selection.
Innovation Solution
A computer-implemented method and system that identifies candidate words in search queries by counting their occurrences as single words and with other words, and adds them to an input method editor dictionary based on a relationship between these counts, enhancing the dictionary dynamically with context signals like Chinese book title marks and search query logs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a standard keyboard is used to input logographic characters, then the input device structure remains simple, but the character input efficiency deteriorates due to many-to-one mapping
Solution Approach 1:
The patent introduces an input method editor (IME) as an intermediary software component between the standard keyboard and the logographic character set. The IME captures keyboard input, processes it through candidate word generation and selection interfaces, and outputs the final character. This mediator layer enables efficient logographic input using a simple standard keyboard by adding intelligence to the input process rather than changing the physical device.
Solution Approach 2:
The system performs preliminary action by pre-generating candidate words and characters based on the input sequence before the user needs to select the final character. The IME anticipates possible interpretations of the input and prepares multiple candidates in advance, allowing the user to quickly select from a limited set rather than searching through all possible characters.
2Reliability
If a static dictionary is used in the input method editor, then the device structure remains simple, but the dictionary relevance deteriorates over time
Solution Approach 1:
The patent implements feedback mechanisms where the IME analyzes user selection patterns, candidate word choices, and input sequences to generate statistics about actual language usage. This feedback loop allows the system to identify frequently used words and phrases that may be missing from traditional dictionaries, enabling continuous improvement of dictionary relevance based on real-world usage data.
Solution Approach 2:
The system performs self-service by automatically updating its own dictionary content based on analyzed usage patterns. The IME monitors its own operation, identifies frequently used candidate words, and adds them to the dictionary without requiring manual intervention or external updates, allowing the dictionary to evolve and improve relevance over time.
3Productivity
If manual dictionary updates are performed, then the update process remains simple, but the user input time increases due to frequent dictionary consultations
Solution Approach 1:
The patent enables continuous accumulation of usage data during normal typing operations. Every time the user inputs characters and selects candidates, the system continuously records this information and updates usage statistics in the background without interrupting the user's workflow. This continuous data collection process happens seamlessly during regular use, eliminating the need for separate dictionary update sessions.
Solution Approach 2:
The system skips the traditional time-consuming manual dictionary update process by rapidly automatically analyzing usage patterns and generating updates in the background. The IME processes usage data continuously and generates dictionary updates without requiring the user to pause their work or manually review and add words, rushing through the update process efficiently in the background.
Data Source
AI summary
Candidate words in search queries are identified, each candidate word including one or more consecutive characters. For each candidate word, a first count is determined, the first count representing a number of times that the candidate word is the only word in the search queries, and a second count is determined, the second count representing a number of times that the candidate word and one or more other words are included in each of the search queries. One or more of the candidate words are added to an input method editor dictionary based on a relationship between the first count and the second count.