Intelligent Keyboard Context-Aware Phrase Suggestions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current computing devices do not suggest entire phrases to start conversations based on the category of conversation and intent of the user, despite offering word suggestions and auto-correction features.

Innovation Solution

An intelligent keyboard system that provides conversation suggestions by utilizing user-generated content, usage history, profile data, and artificial intelligence to rank and tag content for relevance and context, allowing users to select phrases appropriate for specific conversation types and contexts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If word suggestions and auto-correction are provided, then typing efficiency is improved, but the ability to suggest entire conversation-starting phrases is lost

Engineering Contradiction:
Improvetyping efficiencyVSAvoidconversation phrase suggestion capability
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent combines traditional word suggestion technology with conversation analysis and categorization systems to create a unified keyboard that can suggest both individual words and entire conversation-starting phrases based on contextual understanding

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system pre-categorizes and pre-organizes conversation phrases by topic, intent, and context before the user needs them, allowing rapid retrieval and suggestion of appropriate phrases without requiring the user to think about what to say next

Inventive Principle:
Principle #10Preliminary action

2Ease of manufacture

If generic word suggestions are provided, then the system is simple to implement, but contextually relevant phrase suggestions cannot be provided

Engineering Contradiction:
Improvesystem implementation simplicityVSAvoidcontextual relevance
Core Design Contradiction:
Ease of manufactureVSLoss of information

Solution Approach 1:

The system applies different levels of processing and personalization to different aspects of keyboard functionality, maintaining simple word prediction while adding sophisticated contextual analysis specifically for conversation phrase suggestions

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system continuously learns from user interactions, analyzing which phrases are selected and how they are used to improve future suggestions, creating a feedback loop that enhances contextual relevance without requiring complete system redesign

Inventive Principle:
Principle #23Feedback

3Measurement precision

If comprehensive user data is collected for personalization, then conversation suggestion accuracy is improved, but system complexity increases

Engineering Contradiction:
Improvesuggestion accuracyVSAvoiddata processing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system divides user data into distinct categories (typing patterns, conversation topics, personal preferences, usage timing) and processes each segment separately through specialized algorithms, making the overall complex system more manageable and efficient

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11726656B2Intelligent keyboard
Publication Date: 2023.08.15 KEYS INC
  • US11726656B2 patent drawing
  • US11726656B2 patent drawing
  • US11726656B2 patent drawing

AI summary

An intelligent keyboard for mobile devices and other computer devices. This intelligent keyboard provides the user with suggestions of relevant words or phrases that can be used to start or continue a conversation on text message, email and/or various web applications. The intelligent keyboard provides conversation suggestions that are appropriate for given application contexts, categories, and conversation types. The intelligent keyboard uses user generated content from application users, usage history, profile data, dialogue data, platform generated content from the system managers/owners, content collected from various websites/integrations and natural language content generated by artificial intelligence. Content is ranked by preference, contextual suitability, and performance. Content is further tagged for application context. User behavior, user data and artificial intelligence models continuously update the system so that the relevance and performance of keyboard content is optimized.