AI Content Sectioning for Rapid Information Retrieval

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing content providing methods are inefficient in quickly locating desired information within long content reproductions, such as videos or audio, and lack robust security measures during content reprocessing and sharing.

Innovation Solution

A content providing method that divides content into sections based on changes in speakers, subjects, or background images, and uses artificial intelligence to extract keywords for each section, allowing for quick retrieval and secure sharing of desired content.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If content is searched by moving reproduction time point and checking thumbnail images, then user can find desired information, but it requires considerable amount of time especially in long content reproduction

Engineering Contradiction:

Improveprecision of finding desired informationVSAvoidtime to find desired time point

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides long content into multiple sections based on changes in speakers, subjects, or background images. Each section is assigned preview information including representative time points and keywords extracted through AI analysis. This segmentation allows users to quickly navigate to relevant sections without manually scrolling through entire content or repeatedly using time-based navigation functions.

Inventive Principle:

Principle #1Segmentation

2Productivity

If content is divided into sections with AI-generated keywords and preview information, then user can quickly locate desired information, but system complexity increases

Engineering Contradiction:

Improvespeed of finding desired informationVSAvoidcomplexity of content processing system

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary processing system that performs AI-based analysis to extract keywords and generate preview information for each content section. This intermediary layer between the original content and the user interface automates the complexity of content analysis, allowing users to benefit from intelligent sectioning and keyword extraction without directly managing the complex processing requirements.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Ease of operation

If user saves and edits content on personal PC for reprocessing and sharing, then user can obtain desired information, but content security is compromised

Engineering Contradiction:

Improveease of reprocessing and sharing contentVSAvoidsecurity of content

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent extracts and provides only the specific sections and information that users need, rather than allowing them to download and save entire content files to personal devices. By extracting precise content segments with associated preview information and enabling sharing through the platform's secure infrastructure, the system maintains content security while still allowing users to access and share desired information.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS12315536B2Content providing method and device for implementing the same

Publication Date: 2025.05.27 SAMSUNG SDS CO LTD

AI summary

A content providing method performed by a computing device is provided. The content providing method includes: obtaining preview information on a plurality of sections of target content, the preview information including a keyword group associated with content of a corresponding section, and providing the preview information on the plurality of sections together with the target content.