ML Page Categories

Overview

This dataset reflects periodic daily ML inferences where Sincera categorizes text assets from the publisher_text_assets dataset with the current available IAB model on that environment.

Dataset

Field	Type	Description
id	integer	Unique identifier that applies to the related inference run.
page_id	integer	Unique identifier for the specific publisher page asset being classified.
classification	jsonb	Metadata providing category weights based on IAB category.
model_id	integer	Corresponds to latest ML model version.
created_at	datetime	Date when the inference run was first seen by the Sincera platform.
updated_at	datetime	Date when the inference run was last updated by the Sincera platform.
page_url	string	URL to publisher page being inferenced upon.
article_title	string	Article being inferenced on the page.
reviewed	boolean	Indicates whether the inference run has been reviewed.
user_id	integer	The user ID running the inference.
en_text	boolean	Detects whether the text is English language.
truncated_text	string	The js script found on the page that instantiates the video player.
pretty_classification	jsonb	Pretty version of the `classification` metadata outputs.
publisher_id	integer	Maps to the `publisher_id` primary key in the publishers dataset.
traffic_rank	integer	Indicates the publisher’s traffic rank.
result_status	integer	Result ID.