Ontology-Based Media Content Marking via Text Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current media content management is inefficient due to reliance on manual operations and inaccurate image identification, lacking a structured ontology library for media content management, which hinders effective retrieval and editing of multimedia information.

Innovation Solution

A system and method utilizing a text classifier to classify subtitle information, mark specific time segments of media content, and match them with ontology library concepts for efficient media content management, enabling semantic retrieval and editing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual operations are used to mark and manage media content, then flexibility and adaptability are maintained, but time consumption and labor effort increase significantly

Engineering Contradiction:

Improvemanual marking flexibilityVSAvoidmedia content management efficiency

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The system enables automatic self-service marking by extracting text from media content, classifying it through ontology-based text classification, and automatically generating time-stamped markers without requiring manual intervention for each media file

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical marking operations with an automated information processing system that uses text extraction, classification algorithms, and ontology matching to perform marking tasks automatically

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Extent of automation

If image identification technology is used to mark media contents, then automation is improved, but accuracy and speed remain insufficient for practical application

Engineering Contradiction:

Improveautomatic content identificationVSAvoidcontent identification accuracy

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The patent replaces inaccurate image identification technology with text-based extraction and classification systems that leverage ontology libraries to achieve higher accuracy in content identification and marking

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system introduces text extraction and ontology-based classification as intermediary steps between raw media content and final marking, improving accuracy by processing content through multiple refinement stages

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If a perfectly structured ontology library is created for media content management, then retrieval and search capabilities are improved, but the complexity of creating and maintaining the library increases

Engineering Contradiction:

Improvecontent retrieval accuracyVSAvoidontology library structure

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The ontology library is segmented into hierarchical levels and modular components that can be independently developed, maintained, and updated, reducing overall system complexity while maintaining comprehensive coverage

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The ontology library is designed with universal structures and relationships that can accommodate multiple media types and content domains, reducing the need for separate specialized libraries for different content kinds

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Productivity

If media contents are described through text information with ontology correlation, then search and retrieval efficiency are improved, but the workload for text extraction and classification increases

Engineering Contradiction:

Improvesearch efficiencyVSAvoidtext processing time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary text extraction, classification, and ontology matching during the initial processing stage, so that retrieval operations can directly query pre-processed and pre-classified content without repeating analysis

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements continuous processing pipelines where text extraction, classification, and marking operations flow continuously without interruption, maximizing system throughput and minimizing idle time

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS8200597B2System and method for classifiying text and managing media contents using subtitles, start times, end times, and an ontology library

Publication Date: 2012.06.12 HUAWEI TECH CO LTD

AI summary

A system and method for managing media contents are disclosed. The system includes: a text classifier, adapted to classify subtitle information according to defined subjects, and obtain content clips of different subjects; and a media content marking unit, adapted to mark the time of playing each content clip of a different subject after contents are classified by the text classifier, obtain the content clips which have specific time information and different subjects, match the content clips with concepts in an ontology library, and mark the content clips through terms defined in the ontology library. Therefore, the media contents are described through standard terms, which is conducive to unification of the content description information and makes it possible to retrieve the media contents.