Dynamic Database Augmentation for Multi-Word Substitution

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing database systems for communication devices fail to learn from user activities and accurately handle groups of words, leading to incorrect spelling corrections and predictions.

Innovation Solution

A method is implemented where a communication device monitors a second database for added linguistic objects and updates its own database accordingly, allowing for real-time adaptation and accurate multi-word substitution during text input.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a database of default words is stored in a communication device, then spelling corrections and predictions can be performed, but the database cannot learn from user activities and remains static

Engineering Contradiction:
Improvespelling correction accuracyVSAvoiddatabase learning capability
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent transforms the static database into a dynamic system that automatically updates itself by monitoring user activities. The database evolves over time by incorporating new words and phrases from contact databases, message databases, and user input patterns, making it adaptive to individual user needs while maintaining its core spelling correction function.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The database performs self-updates by automatically monitoring user activities across different applications and databases within the device. It autonomously identifies new linguistic patterns and incorporates them without requiring external intervention or manual updates, enabling the system to improve its own performance continuously.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If single-word spelling correction is performed using a default database, then individual word accuracy can be improved, but multi-word phrases and groups of words are not recognized

Engineering Contradiction:
Improvesingle word spelling accuracyVSAvoidmulti-word phrase recognition
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent merges single-word correction capabilities with multi-word phrase recognition by integrating data from contact databases and message databases. This combination allows the system to recognize both individual words and their contextual relationships, enabling accurate correction of phrases like 'John Doe' as a unified linguistic unit rather than separate words.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent transitions from one-dimensional single-word correction to multi-dimensional phrase-level correction by incorporating contextual information from multiple sources. The system analyzes linguistic patterns across different dimensions (contact information, message history, user input) to recognize and correct multi-word phrases accurately.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP2613271B1Method and apparatus for database augmentation and multi-word substitution
Publication Date: 2020.12.02 BLACKBERRY LTD
  • EP2613271B1 patent drawingFigure 1
  • EP2613271B1 patent drawingFigure 2
  • EP2613271B1 patent drawingFigure 3

AI summary

A method and communication device are provided for database augmentation using linguistic data stored on a device, and utilizing a database stored on a device to perform multi-word substitution. A database may be augmented by monitoring other databases that contain linguistic data, such as contact databases containing linguistic data regarding entities that a device may communicate with, and updating the database with linguistic data in the other databases. The linguistic data in the augmented database may be compared with words received from an input apparatus to determine whether any of the received words should be replaced with linguistic data from the augmented database. The augmented database may contain one word entries and multi-word entries to allow for multi-word substitution.