Context Inference Engine for Mobile Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile device users often fail to recognize or remember the need for important location-based services when they are busy, as existing systems require overt action to initiate these services, leading to missed opportunities for useful information or assistance.
Innovation Solution
A method for automatically inferring a user-specific current context using real-time sensor inputs and stored user-specific information, applying an activity knowledge base with automated reasoning to generate timely suggestions, notifications, or actions without user intervention, and presenting them in a suitable form based on the context.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the system requires overt user action to initiate location-based services, then the system complexity is reduced, but the user productivity and service utilization are worsened due to missed opportunities when users are busy
Solution Approach 1:
The system performs preliminary actions by continuously monitoring sensor inputs and pre-processing location data in the background before any user request. The inference engine proactively analyzes sensor patterns and prepares context information ahead of time, so when a service is needed, it is immediately available without requiring user initiation.
Solution Approach 2:
The system serves itself by automatically inferring user context and generating service recommendations without human intervention. The automated reasoning engine continuously processes sensor data, determines user situations, and presents relevant information or actions autonomously, freeing users from the burden of manually requesting services.
2Reliability
If the system continuously monitors sensor inputs and applies automated reasoning, then the service relevance and timeliness are improved, but the energy consumption and processing load are worsened
Solution Approach 1:
The system applies partial monitoring by selectively processing sensor inputs based on their relevance to inferred user context. Rather than analyzing all sensor data uniformly, the inference engine focuses computational resources on specific sensor patterns that indicate meaningful situations, performing excessive analysis only when necessary to confirm context hypotheses.
Solution Approach 2:
The system uses periodic action by updating context inference at intervals rather than continuously processing all data streams. The automated reasoning engine periodically reassesses user context based on accumulated sensor patterns, reducing peak processing loads while maintaining reliable service relevance through rhythmic analysis cycles.
3Ease of operation
If the system presents multiple possible contexts and suggestions, then the user assistance quality is improved, but the information overload and user decision complexity are worsened
Solution Approach 1:
The system applies local quality by tailoring the presentation of context information to the specific user situation and inferred needs. Rather than presenting all possible contexts uniformly, the inference engine selectively highlights the most relevant suggestions based on the current user context, providing detailed information locally where needed while suppressing irrelevant data elsewhere.
Solution Approach 2:
The system uses partial action by presenting only a subset of possible contexts and suggestions rather than exhaustively listing all alternatives. The automated reasoning engine filters and prioritizes information to show the most probable and useful contexts first, providing sufficient detail for common situations while offering expanded options only when initial suggestions are insufficient.
Data Source
AI summary
A device, method and system for automatically inferring a mobile user's current context includes applying a user activity knowledge base to real-time inputs and stored user-specific information to determine a current situation. Automated reasoning is used to infer a user-specific context of the current situation. Automated candidate actions may be generated and performed in accordance with the current situation and user-specific context.