Task-Oriented Content Delivery via Activity Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing digital content delivery methods struggle to provide relevant content to users based on their current tasks, as they rely on limited understanding of user activities and lack the ability to automatically deliver remote content.

Innovation Solution

A task-oriented user activity system that predicts the user's current task by analyzing event records and metadata, communicating this information to a digital content service provider to deliver highly relevant digital content, which is then filtered and presented to the user based on their task-related metadata.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If search engine queries are used to filter and prioritize digital content, then content delivery can be implemented, but the relevance to user's current task is limited due to restricted frame of reference

Engineering Contradiction:

Improvecontent relevanceVSAvoidunderstanding of user task

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent introduces an intermediary component (the task monitoring system) that sits between the user's computer activities and the digital content delivery system. This intermediary captures event records from multiple executing programs, predicts current tasks, and translates them into meaningful context for content selection, thereby bridging the gap between limited search engine understanding and actual user intent

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical search engine query system with an automated task prediction and monitoring system. Instead of relying on users to manually input queries or the system to interpret limited signals, the new system automatically monitors program events, predicts tasks using machine learning algorithms, and delivers contextually relevant content without requiring explicit user input

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Speed

If local user resources are used for task assistance, then quick access is achieved, but remote relevant content cannot be delivered

Engineering Contradiction:

Improvecontent access speedVSAvoidcontent delivery scope

Core Design Contradiction:

SpeedVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal content delivery system that can handle both local and remote resources through a single integrated architecture. The system predicts user tasks and delivers relevant content from any source (local or remote) based on task requirements, making the content delivery mechanism versatile rather than limited to specific resource locations

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The task prediction system acts as an intermediary that determines whether content should be delivered from local or remote sources based on the predicted task context. This intermediary layer enables seamless integration of local quick access with remote content delivery, selecting the appropriate source dynamically

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If machine learning algorithms predict current task from stored evidence, then task understanding is improved, but automated delivery of remote content is not achieved

Engineering Contradiction:

Improvetask prediction accuracyVSAvoidremote content delivery

Core Design Contradiction:

Measurement precisionVSExtent of automation

Solution Approach 1:

The patent merges the task prediction functionality with the content delivery system into an integrated automated workflow. The machine learning task prediction and the remote content delivery are combined into a single system that automatically flows from task prediction to content selection and delivery without manual intervention at any stage

Inventive Principle:

Principle #5Merging (Combining)

4Productivity

If digital content is delivered based on limited user activity understanding, then content delivery can occur, but effectiveness is reduced due to lack of task context

Engineering Contradiction:

Improvecontent utilization effectivenessVSAvoiduser task context

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The patent performs preliminary action by predicting the user's current task before delivering content. The system continuously monitors event records, predicts tasks using machine learning algorithms, and prepares relevant content in advance based on the predicted task context, ensuring content is delivered proactively rather than reactively

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9524355B2Methods for delivering task-related digital content based on task-oriented user activity

Publication Date: 2016.12.20 EMC IP HLDG CO LLC

AI summary

A method for automatically presenting digital content to a user of a computer computes task-related metadata from data which may include i) a most recent event record, ii) a most recent specification received from the user of a task being performed by the user, iii) an automatic prediction of the current task being executed by the user, iii) past event records and associated task identifiers stored in a database and/or iv) content in resources associated with a given task. The task-related metadata is communicated to a digital content service provider. Digital content relevant to the user based on the task-related metadata is then selected, sent to the computer, and presented to the user. Rules and filters control the metadata going out and the content coming in, allowing automatic adaptation based on the current task, characteristics of the content, and other factors.