Utilizing machine learning models to identify context of content for policy compliance determination

US11275948B2 · kind B2 · utility

Assignee

• ACCENTURE GLOBAL SOLUTIONS LIMITED · IE

Inventors

• Srikanth G. Rao · Bengaluru, IN
• Tarun Singhal · Bulandshahr, IN
• Mathangi Sandilya · Kanchinakote, IN
• Vipin Bhasin · Pune, IN
• Bolaka Mukherjee · Jharia, IN
• J. Nooji Shekar · Bengaluru, IN
• Manish Phukan · Dudhnai, IN
• Swati Dubey · Bengaluru, IN
• Mamta Kanad Bhagwat · Mumbai, IN
• T Siva Rama Sarma · Secunderabad, IN

Key dates

• Filing date: Dec 10, 2019
• Grant date: Mar 15, 2022
• Expiry date: Sep 1, 2040

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N20/20
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

A device may receive video data associated with a video, and may identify image data associated with the video data and that corresponds to a frame of the video. The device may process the image data, with an image analysis model, to determine an image compliance score associated with a policy and with context of the image data, and may identify text data associated with the video data. The device may process the text data, with a text analysis model, to determine a text compliance score associated with the policy and with context of the text data, and may determine, based on the image compliance score and the text compliance score, an aggregate compliance score associated with the frame. The device may determine whether the aggregate compliance score satisfies a threshold associated with the policy, and may perform actions based on whether the aggregate compliance score satisfies the threshold.