Session Progress Matrix Completion for Content Ranking

Solution Overview

Problem

Existing content rating systems face challenges such as low user participation due to perceived burdensomeness and susceptibility to manipulation, leading to a disconnect between ratings and consumer preferences, which affects content selection and recommendation accuracy.

Innovation Solution

Implementing a session progress-based ranking system that uses implicit feedback from content consumption data, analyzed through matrix completion techniques and collaborative filtering, to predict consumption duration without requiring explicit user input, thereby enhancing accuracy and reducing manipulation risks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If explicit user rating systems are implemented, then content popularity can be measured, but user participation is low due to perceived burdensomeness

Engineering Contradiction:

Improvecontent rating accuracyVSAvoiduser participation

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system automatically generates content ratings by analyzing user consumption behavior data (session progress, duration, interactions) without requiring users to manually provide ratings. The system serves itself by extracting meaningful signals from passive observation of user actions, eliminating the need for user effort while maintaining measurement capability

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system implements implicit feedback mechanisms by continuously monitoring and analyzing user consumption patterns. Instead of requiring explicit user feedback through ratings, the system captures indirect feedback through behavioral data such as session progress, duration, and interaction patterns, which are then processed to generate content rankings

Inventive Principle:

Principle #23Feedback

2Measurement precision

If explicit user rating systems are implemented, then content popularity can be measured, but ratings are susceptible to manipulation

Engineering Contradiction:

Improvecontent rating accuracyVSAvoidrating authenticity

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The system automatically generates content ratings by analyzing user consumption behavior data (session progress, duration, interactions) without requiring users to manually provide ratings. The system serves itself by extracting meaningful signals from passive observation of user actions, eliminating the need for user effort while maintaining measurement capability

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system introduces behavioral data as an intermediary between user preferences and content ratings. Instead of directly using user-submitted ratings that can be manipulated, the system uses objective consumption behavior data (session progress, duration, interactions) as a mediator to infer true content popularity, making manipulation difficult

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If personalized rating systems with user authentication are implemented, then individual preferences can be captured, but the rating process becomes more burdensome

Engineering Contradiction:

Improvepersonalization capabilityVSAvoidrating process simplicity

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system automatically generates content ratings by analyzing user consumption behavior data (session progress, duration, interactions) without requiring users to manually provide ratings. The system serves itself by extracting meaningful signals from passive observation of user actions, eliminating the need for user effort while maintaining measurement capability

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically adapts content recommendations based on ongoing analysis of user consumption patterns. Rather than requiring users to update their profiles or preferences manually, the system continuously learns and adapts to individual user preferences by processing behavioral data in real-time, making personalization transparent and effortless

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10567838B2Digital content consumption analysis

Publication Date: 2020.02.18 ADOBE INC

AI summary

Content consumption session progress is predicted based on historical observations of how users have interacted with a repository of digital content. This is approached as a matrix completion problem. Information extracted from tracking logs maintained by one or more content providers is used to estimate the extent to which various content items are consumed. The extracted session progress data is used to populate a session progress matrix in which each matrix element represents a session progress for a particular user consuming a particular content item. This matrix, which in principle will be highly (≳95%) sparse, can be completed using a collaborative filtering matrix completion technique. The values obtained as a result of completing the session progress matrix represent predictions with respect to how much of a given content item will be consumed by a given user.