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

A device, comprising:

• one or more memories;
• and
• one or more processors, communicatively coupled to the one or more memories, configured to:
• receive video data associated with a video;
• identify image data associated with the video data,
• wherein the image data corresponds to one or more frames of the video;
• process the image data, with an image analysis model, to determine an image compliance score associated with the image data,
• wherein the image analysis model is trained based on associations between a plurality of objects included in historical image data, and
• wherein the image compliance score is associated with a policy and with context of the image data;
• identify text data associated with the video data;
• process the text data, with a text analysis model, to determine a text compliance score associated with the text data,
• wherein the text analysis model is trained based on associations between a plurality of words included in historical text data, and
• wherein the text compliance score is associated with the policy and with context of the text data;
• determine, based on the image compliance score and the text compliance score, an aggregate compliance score associated with the one or more frames of the video;
• determine whether the aggregate compliance score satisfies a threshold score associated with the policy;
• and
• perform one or more actions associated with the video based on whether the aggregate compliance score satisfies the threshold score.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors, cause the one or more processors to:
• train an image analysis model based on associations between a plurality of objects included in historical image data;
• train a text analysis model based on associations between a plurality of words included in historical text data;
• receive video data associated with a video;
• identify image data associated with the video data,
• wherein the image data corresponds to a frame of the video;
• process the image data, with the image analysis model, to determine an image compliance score associated with the image data,
• wherein the image compliance score is associated with context of the image data and with a policy that dictates conditions for finding video compliant or non-compliant;
• identify text data associated with the video data;
• process the text data, with the text analysis model, to determine a text compliance score associated with the text data,
• wherein the text compliance score is associated with the policy and with context of the text data;
• determine, based on the image compliance score and the text compliance score, an aggregate compliance score associated with the frame of the video;
• determine whether the aggregate compliance score satisfies a threshold score associated with the policy;
• and
• perform one or more actions associated with the video based on whether the aggregate compliance score satisfies the threshold score.