Media Content Prediction System Using Contextual Feature Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users face difficulties in locating media content of interest due to an overabundance or scarcity of available content, which can lead to abandonment of search efforts, posing a challenge for content providers who seek to engage users for revenue generation.

Innovation Solution

A system and method that employs machine learning and statistical processes to predict user preferences by analyzing textual, visual, and social context features of media content, such as images, to classify content as of interest or not, using classifiers trained with features like spatial, temporal, and aesthetic characteristics, and social interactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If media content is made readily accessible to users, then user engagement and content consumption increase, but users become overwhelmed by the overabundance of content and abandon search efforts

Engineering Contradiction:

Improvecontent consumptionVSAvoidsearch efficiency

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The system segments the vast media content library into personalized subsets based on user preferences, history, and behavior patterns. The recommendation engine divides content into categories and prioritizes them according to individual user interests, transforming the overwhelming 'all content' view into manageable, relevant sections.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary recommendation system that acts as a mediator between users and media content. This intermediary layer processes user interactions, analyzes patterns, and presents curated content recommendations, filtering out irrelevant content while maintaining access to diverse media options.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If content providers offer diverse media content to satisfy user interests, then content availability increases, but the difficulty of locating desired content increases due to the overabundance of choices

Engineering Contradiction:

Improvecontent diversityVSAvoidcontent location difficulty

Core Design Contradiction:

Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The system applies local quality by tailoring content presentation to individual user contexts. Rather than uniformly presenting all diverse content, the system adjusts content organization, prioritization, and discovery mechanisms based on each user's specific interests, viewing history, and interaction patterns, making diverse content locally relevant and easier to navigate.

Inventive Principle:

Principle #3Local quality

3Quantity of substance

If users are presented with many content choices to ensure interest, then content availability improves, but user attention span decreases as they abandon search efforts

Engineering Contradiction:

Improvecontent availabilityVSAvoidsearch time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-processing and pre-organizing content based on predicted user interests before the user actively searches. The recommendation engine continuously learns from user behavior and pre-ranks content, so when users browse or search, the most relevant content is already positioned prominently, reducing the time needed to locate desired media.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9137574B2Method or system to predict media content preferences

Publication Date: 2015.09.15 R2 SOLUTIONS LLC

AI summary

Briefly, embodiments of a method or system of predicting media content preferences are disclosed.