Activity Duration Learning for Runtime-Matched Content

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems struggle to provide content during variable-duration activities without manual user input, often leading to inappropriate or undesired content consumption due to mismatched durations and user inattention.

Innovation Solution

A system learns the duration of a variable-duration activity by recognizing start and stop cues, allowing it to identify and provide content items with a total runtime similar to the activity duration, using an activity engine and content curation engine to tailor content to the user's preferences and activity length.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If content is provided without manual user input for activity duration, then ease of operation is improved, but measurement precision of activity duration deteriorates

Engineering Contradiction:
Improveease of operationVSAvoidmeasurement precision
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The system automatically detects activity duration by monitoring device usage patterns, screen on/off events, and contextual cues without requiring manual user input. The activity engine self-determines activity boundaries and duration based on observed behavior, eliminating the need for users to manually specify time parameters while maintaining accurate duration measurement.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system continuously monitors device state and user behavior to provide feedback about detected activity duration. This feedback loop allows the system to refine its timing accuracy by comparing detected activity boundaries with actual usage patterns, progressively improving measurement precision through iterative observation and adjustment.

Inventive Principle:
Principle #23Feedback

2Adaptability or versatility

If content runtime does not match activity duration, then adaptability of content delivery is improved, but loss of time deteriorates

Engineering Contradiction:
ImproveadaptabilityVSAvoidloss of time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The content delivery system dynamically adjusts content selection and duration based on the detected activity length. The content curation engine receives the activity duration as input and selects content items whose total runtime matches the activity length, creating a dynamic adaptation mechanism that prevents time loss while maintaining flexibility in content choices.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the parameter of content runtime to match the detected activity duration. By adjusting the total content playback time to correspond with the measured activity length, the system eliminates wasted time while preserving adaptability through varied content options that can be selected based on user preferences and activity context.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If content is constantly delivered without knowing activity end, then productivity of content delivery is improved, but object-affected harmful factors deteriorate

Engineering Contradiction:
ImproveproductivityVSAvoidharmful factors
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The system uses feedback from detecting activity termination cues to stop content delivery appropriately. When the activity engine detects that the activity has ended (through device unavailability, time elapsed, or explicit termination signals), it sends feedback to the content curation engine to cease content delivery, preventing harmful effects such as content playing after the activity concludes while maintaining high productivity during the active period.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4600805A1Learning activity duration for providing content during activity
Publication Date: 2025.08.13 ADEIA GUIDES INC
  • EP4600805A1 patent drawingFigure 1A~1B
  • EP4600805A1 patent drawingFigure 2A~2B
  • EP4600805A1 patent drawingFigure 3

AI summary

Methods and systems are described for recognizing an activity and providing content for consumption during the activity. The methods and systems use an activity engine to learn a duration of an activity by receiving input with a start cue indicating a start of an activity and receiving input with a stop cue indicating an end of the activity. The activity engine determines an average or estimated duration for the activity based on the time difference between the start cue and the stop cue so that when the activity engine receives a third input and identifies the start cue, a content curation engine identifies one or more content items with a total runtime substantially similar to the average or estimated duration for the activity and provides the content for consumption.