Mobile Message Priority System Using Context-Aware Feature Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users of mobile communication devices face overwhelming numbers of messages and notifications, leading to inefficiencies as they often need to respond to non-priority messages, especially when busy or in specific contexts like driving.

Innovation Solution

A method that extracts features from messages and user context to determine priority levels, using models like linear regression or Support Vector Machines, and adjusts notifications based on user actions and context, such as location and time, to prioritize messages and manage notifications effectively.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If users receive notifications for all messages, then users are informed of all incoming messages, but users become overwhelmed and waste time responding to non-priority messages

Engineering Contradiction:
Improvemessage notification deliveryVSAvoidtime spent responding to messages
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system extracts and separates messages into different priority levels (high, medium, low) based on analyzed features such as sender contact, message content, timing, and user behavior patterns. This extraction allows the system to present only the most important messages prominently while suppressing less important ones, resolving the contradiction between informing users of all messages and preventing time waste on non-priority messages.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies different notification qualities to different messages based on their priority levels. High-priority messages receive prominent notifications with full details, medium-priority messages receive subdued notifications, and low-priority messages may be suppressed entirely or notified only under specific conditions. This local differentiation resolves the contradiction by ensuring important information is delivered while reducing distraction from less important messages.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If the system analyzes multiple features and user context to determine priority, then message prioritization accuracy improves, but system complexity increases

Engineering Contradiction:
Improvepriority determination accuracyVSAvoidfeature extraction and model processing
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary analysis of message features and user context in advance, building up a comprehensive profile before final priority determination. Features such as contact information, message content, sending time, and historical user behavior are pre-processed and stored. This preliminary action allows the complex priority determination model to operate more efficiently by working with preorganized data, thus achieving high accuracy without proportionally increasing real-time processing complexity.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediary priority prediction model that sits between the raw message data and the final notification delivery decision. This intermediary model processes multiple features and user context to generate priority predictions, which then guide the notification system. The model acts as a mediator that consolidates complex analysis into actionable priority levels, achieving accurate prioritization while managing system complexity through a dedicated processing layer.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2842274B1Systems and methods for prioritizing messages on a mobile device
Publication Date: 2019.06.26 SONY MOBILE COMM AB
  • EP2842274B1 patent drawingFigure 1
  • EP2842274B1 patent drawingFigure 2
  • EP2842274B1 patent drawingFigure 3

AI summary

A method, performed by a mobile communication device, may include receiving a message (1010); extracting one or more features associated with the message (1020); and determining a priority level for the message, based on the extracted one or more features, using a model that relates the extracted one or more features to an expected response time for the message (1030, 1040). The method may further include determining a context associated with the user of the mobile communication device (1050); modifying the priority level based on the determined context (1060); and providing the message to a list of messages based on the modified priority level (1090). Additionally, the user may be provided with a notification tone associated with the modified priority level.