Context-Aware Text Prediction Using Dynamic User Dictionaries

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional text prediction technologies rely on static dictionaries, limiting their effectiveness due to storage constraints and lack of context sensitivity, which hampers the accuracy and relevance of suggested text in mobile devices.

Innovation Solution

A computer-implemented method using a personalized text suggestion dictionary that associates text with context attributes based on historical user data, allowing for context-aware predicted-text suggestions by analyzing past communications and metadata to provide more accurate and relevant options.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional static dictionaries are used for text prediction, then storage resources are consumed, but text prediction accuracy and context sensitivity are limited

Engineering Contradiction:
Improvetext prediction accuracyVSAvoidstorage resources
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent transforms the static dictionary into a dynamic system that adapts to user behavior and context. The text prediction system continuously learns from user communications, updating predictions based on recurring text patterns and context attributes, thereby improving accuracy without relying on large pre-installed static dictionaries

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs preliminary analysis of user communications to identify recurring text and context attributes before generating predictions. By pre-processing communication data and storing context associations, the system prepares prediction models in advance, enabling accurate context-aware suggestions without requiring extensive storage of raw communication data

Inventive Principle:
Principle #10Preliminary action

2Adaptability or versatility

If large preinstalled dictionaries are used, then text prediction coverage is improved, but device storage capacity is exceeded

Engineering Contradiction:
Improvetext prediction coverageVSAvoiddevice storage capacity
Core Design Contradiction:
Adaptability or versatilityVSVolume of stationary object

Solution Approach 1:

Instead of using a single large universal dictionary, the patent implements context-specific text suggestions tailored to local communication scenarios. The system analyzes context attributes (time, contact, application) and provides predictions specific to each local context, achieving versatile coverage across different communication situations without requiring a comprehensive large dictionary

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system creates lightweight copies of relevant text patterns from user communications rather than storing complete dictionaries. By extracting and storing only the recurring text snippets and their context associations that are specific to each user, the system achieves adaptable prediction coverage with minimal storage requirements

Inventive Principle:
Principle #26Copying

3Adaptability or versatility

If static dictionaries are used, then implementation simplicity is maintained, but context sensitivity and relevance of suggestions are reduced

Engineering Contradiction:
Improvecontext sensitivityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a feedback mechanism where the text prediction system continuously monitors user interactions and communication patterns. By analyzing recurring text and context attributes from user communications, the system learns and adapts its predictions, improving context sensitivity through feedback loops without requiring complex manual configuration

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs self-learning by automatically analyzing user communications and generating context-aware prediction models without external intervention. The text prediction engine autonomously identifies recurring patterns, associates them with context attributes, and updates its prediction capabilities, reducing the need for complex external management while enhancing context sensitivity

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2929414B1Communication context based predictive-text suggestion
Publication Date: 2020.07.01 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP2929414B1 patent drawingFigure 1
  • EP2929414B1 patent drawingFigure 2
  • EP2929414B1 patent drawingFigure 3

AI summary

Disclosed herein are representative embodiments of tools and techniques for determining predicted-text suggestions based on communication contexts. According to one exemplary technique, text that recurs in one or more past communications is determined. The one or more past communications being associated with at least one context attribute. Also, a text entry is stored in a text suggestion dictionary. The text entry comprising the text and metadata associating the text with the at least one context attribute. Additionally, using the text suggestion dictionary, at least one predicted-text suggestion that includes the text is determined for a current communication associated with a communication context.