The Verity product provides content-level contextual targeting and brand safety analysis. Publishers, DSPs, and advertisers use Verity to generate content-specific analysis, for example, Verity can detect violent content for the purposes of brand safety, and automobile-related content for the purposes of contextual targeting.
Verity works by applying machine learning (ML) techniques to digital content (web pages, video) and assigning contextual categories, detecting keywords, and calculating a brand safety score based on all of the supported elements.
Verity provides APIs to analyze the following web content:
Page – Verity scans and analyzes text and prominent imagery on a web page and returns a JSON response with IAB Contextual Categories v1.0 and v2.0 and Event targeting categories. For brand safety, Verity classifies unsafe content into multiple threat categories defined by GumGum, including drugs, weapons, and sexual content. The response includes keyword and sentiment evaluation data.
Video – Verity transcribes and analyzes the video’s audio track, the video metadata, and text identified in video frames via Optical Character Recognition (OCR). Verity returns a JSON response with IAB Contextual Categories v2.0 targeting categories, brand safety, and keyword and sentiment evaluation data.
Verity operates as a paid third-party service in the Cloud.