Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 26 Next »

When a partner submits analysis requests for specific video URLs, Verity applies machine learning techniques to the video audio track, sampled video frames, video frames where text appears, and video metadata. Verity returns IAB v2 contextual categories, brand safety scoring, keyword and sentiment evaluation data.

The Verity video classification report is organized into these sections:

  • IAB Categories
    Identifies and scores contextual targeting categories. Includes standard IAB v2.0 categories.

  • Keywords
    Extracts and scores keywords according to prominence.

  • Brand Safety
    Identifies and scores brand safety threats according to GumGum's Threat classification taxonomy (GGThreat), which is in compliance with The 4A’s Advertising Assurance Brand Safety Floor

  • Sentiments  
    Analyzes the overall opinions expressed within the content of a video, and classifies the percentage of positive, negative, and neutral sentiments.

Video Analysis Categories and Keywords

Verity video analysis includes classification and keywords detected in the video’s audio track, sampled video frames, and text appearing in video frames.

Brand Safety (Threat) Categories 

Contextual
Categories

Keywords

Violence
Illegal
Drugs
Sexual
Obscene
Hate
Disaster
Malware
Illness

IAB v2.0
698 categories

Sentiment
Positive, Neutral, Negative categories

Contextual targeting keywords
Named Entities  


Video Analysis Language Support

Verity video keywords and categories are supported in the languages as specified in the following table:

Language

Keywords

Categories

Threat

IAB V2

Sentiment

English

Japanese

Spanish

French

German

Verity Response Format

The Verity API response format for both Page and Video analysis is JavaScript Object Notation (JSON)  Content-Type: application/json; charset=UTF-8.

  • No labels