Media Recommendation Prediction Engine for Style-Based Curation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current media distribution platforms face challenges in scalability, accuracy, and efficiency in recommending media content for specific styles of use, as they rely heavily on human curation, which is impractical for large catalogs and not personalized to individual user tastes.

Innovation Solution

A prediction engine trained with metadata, behavioral, acoustic, and cultural data is used to calculate the likelihood of media objects being suitable for a designated style of use, combined with a factorization machine that generates personalized recommendations based on user preferences, enabling automated and scalable content curation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If human curation is used to tag media content for styles of use, then recommendation accuracy for designated styles is improved, but scalability to large catalogs and new styles deteriorates

Engineering Contradiction:

Improverecommendation accuracyVSAvoidscalability to large catalogs and new styles

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent replaces the mechanical human curation system with an automated machine learning system. The prediction engine uses supervised learning trained on example media lists to automatically generate style tags for media content, eliminating the need for manual human tagging while maintaining recommendation accuracy and enabling scalability to large catalogs and new styles of use.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the prediction engine to automatically tag media content without human intervention. The engine learns from training data and independently applies style tags to new media content, making the curation process autonomous and scalable to handle expanding catalogs and emerging styles of use.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If human curation is used to create media lists for styles of use, then personalized recommendation capability is improved, but productivity and efficiency deteriorate

Engineering Contradiction:

Improvepersonalization accuracyVSAvoidcuration efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent substitutes the manual human curation process with an automated prediction engine that processes media content and generates style tags at scale. This machine-based system maintains the ability to provide personalized recommendations while dramatically increasing curation productivity and efficiency compared to manual human processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the operational parameters from manual human processing to automated computational processing. The prediction engine processes media content using algorithmic analysis of acoustic, behavioral, and cultural features, enabling high-speed tag generation that maintains personalization accuracy while achieving industrial-scale productivity.

Inventive Principle:

Principle #35Parameter changes

3Quantity of substance

If comprehensive human tagging of media content is performed, then coverage of media catalog is improved, but loss of time and computational resources deteriorates

Engineering Contradiction:

Improvemedia catalog coverageVSAvoidtagging time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-training the prediction engine on a training set of example media lists before deployment. This pre-training phase enables the engine to rapidly tag new media content without requiring time-consuming manual human review, achieving comprehensive catalog coverage while minimizing time loss through automated processing.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces time-consuming manual human tagging with automated machine learning processing. The prediction engine efficiently processes large volumes of media content to generate style tags, achieving comprehensive catalog coverage without the time loss associated with manual human curation of each individual media item.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10516906B2Systems, methods, and computer products for recommending media suitable for a designated style of use

Publication Date: 2019.12.24 SPOTIFY

AI summary

Media suitable for a designated style of use are recommended from among a database of media objects. A prediction engine is trained using a plurality of media object lists, each of the media object lists containing metadata associated with a plurality of media objects, the media object lists corresponding to the designated style of use, and the prediction engine being trained to calculate the likelihood that a media object is suitable for the designated style of use. The prediction engine includes a binomial classification model trained with feature vectors that include behavioral data, acoustic data and cultural data for the plurality of media objects. The trained prediction engine is applied to media objects in the database of media objects so as to calculate likelihoods that the media objects are suitable for the designated style of use. One or more media objects are recommended using the calculated likelihoods.