Japanese Input Assistance Device Using Kana-Reading Probability Index

Solution Overview

Problem

Existing auto-complete systems for Japanese language input, such as those using kana-reading estimation, suffer from low accuracy in suggesting character strings due to the probability of word usage being different for each word in a character string, leading to incorrect suggestions.

Innovation Solution

An input assistance device and method that uses an index structure recording words with their kana-readings and indicating the probability of each word being inputted first, to suggest character strings based on the inputted kana character string, ensuring only probable words are suggested.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If kana-reading estimation is used to suggest character strings, then the auto-complete function can be applied to Japanese language input, but the accuracy of suggestions becomes low due to different word usage probabilities

Engineering Contradiction:

Improveapplicability of auto-complete to Japanese languageVSAvoidaccuracy of character string suggestions

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by differentiating the treatment of individual words within a character string. Each word is assigned a specific probability value representing its likelihood of being input first, allowing the system to selectively weight different parts of the character string based on their local input characteristics rather than treating all words uniformly.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent changes the parameter of word selection by introducing probability values for each word being input first. This parameter change allows the system to transition from simple kana-reading matching to probabilistic word ordering consideration, thereby improving suggestion accuracy while maintaining adaptability to Japanese input methods.

Inventive Principle:

Principle #35Parameter changes

2Device complexity

If all words in a character string are treated equally for kana-reading estimation, then the processing is simple, but the suggestion accuracy deteriorates due to ignoring word usage probability

Engineering Contradiction:

Improvecomplexity of kana-reading estimation processVSAvoidaccuracy of character string suggestions

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by differentiating the treatment of individual words within a character string. Each word is assigned a specific probability value representing its likelihood of being input first, allowing the system to selectively weight different parts of the character string based on their local input characteristics rather than treating all words uniformly.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent applies preliminary action by pre-calculating and storing probability values for each word in the character string before the actual input assistance is provided. This preliminary preparation of word probability data enables the system to quickly retrieve and use accurate probability information during input, improving suggestion accuracy without adding significant complexity to the real-time processing.

Inventive Principle:

Principle #10Preliminary action

3Speed

If the system suggests character strings based on simple kana matching, then the response speed is fast, but the relevance of suggestions to user intent deteriorates

Engineering Contradiction:

Improveresponse speed of suggestionVSAvoidrelevance of suggested character strings

Core Design Contradiction:

SpeedVSReliability

Solution Approach 1:

The patent changes the parameter of word selection by introducing probability values for each word being input first. This parameter change allows the system to transition from simple kana-reading matching to probabilistic word ordering consideration, thereby improving suggestion accuracy while maintaining adaptability to Japanese input methods.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent substitutes the mechanical system of simple character-by-character kana matching with a more intelligent system that incorporates probability-based word ordering. This substitution replaces the straightforward mechanical comparison with a probabilistic evaluation mechanism that better reflects actual Japanese input behavior, improving relevance while maintaining reasonable response speed.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS9996508B2Input assistance device, input assistance method and storage medium

Publication Date: 2018.06.12 NEC SOLUTION INNOVATORS LTD

AI summary

To provide an input assistance device, an input assistance method and a storage medium which can present a character string to be suggested to a user in consideration of a probability of being inputted first in a character string. An input assistance device is for assisting a user to input a character string, the device includes: a character string determining unit that determines a character string to be suggested relating to the inputted kana character string, in the case that a user inputs a kana character string by use of index structure, the index structure recording therein a word together with its kana-reading and the index structure indicating whether or not it is probable that the word is inputted first at a time of inputting a character string; and a suggested candidate presenting unit that presents the character string to be suggested, which is determined, as a suggested candidate.