Media Guidance Application Keyword Extraction for Recommendation Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional media guidance systems fail to utilize all available information from media assets to provide relevant recommendations, leading to inefficient content discovery for users.

Innovation Solution

A media guidance application that utilizes user actions and keywords from media asset content to recommend related media assets, including live streaming, by modifying user profiles based on viewing habits and content analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional recommendation systems use basic metadata for recommendations, then the system complexity is low, but the recommendation accuracy and relevance to user interests deteriorates

Engineering Contradiction:

Improverecommendation accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary action by extracting keywords from media asset content in advance and storing them in association with the media assets. This pre-processing of content into searchable keywords enables accurate recommendations without adding complexity during the actual recommendation generation process

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediary mechanism by using keyword matching as a bridge between user profile data and media asset content. This intermediary layer enables precise recommendations by comparing user interests (represented as keywords) with media content keywords, resolving the contradiction between accuracy and complexity

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If the system recommends media assets based on keyword matching, then the recommendation relevance to user interests improves, but the processing time and computational resources increase

Engineering Contradiction:

Improverecommendation relevanceVSAvoidprocessing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system extracts and stores keywords from media asset content in advance, creating a pre-processed index that enables rapid keyword matching during recommendation generation. This preliminary action significantly reduces processing time while maintaining high recommendation relevance

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates a simplified copy or representation of media asset content in the form of keywords, which can be quickly compared against user profiles without processing the full content. This copying approach maintains recommendation accuracy while minimizing computational overhead and processing time

Inventive Principle:

Principle #26Copying

Data Source

PatentUS9781479B2Methods and systems of recommending media assets to users based on content of other media assets

Publication Date: 2017.10.03 ADEIA GUIDES INC

AI summary

Methods and systems are described for a media guidance application that provides recommendations to a user viewing a media asset. For example, the media guidance application may provide a recommendation of a media asset based on a user's profile and may modify the user's profile based on the user's actions and keywords in the content of the media asset. For example, the media guidance application may determine whether the media asset is of interest to a user and in response may update the user's profile based on keywords in the media asset. If the media guidance application determines that the media asset is of interest to the user, it may add keywords in the content of the media asset to the user's profile and increase their corresponding weights. Otherwise, the media guidance application may decrease the weights in the user's profile corresponding to keywords in the media asset.