Abstract: Techniques for temporal localization of mature content in long-form videos using only video-level labels are described.

Abstract: Devices, systems, and methods are provided for context-based abusive language detection and responses. A method may include identifying text associated with first video content, and determining that a first word in the text matches a first keyword indicative of abusive language. The method may include determining a first label associated with the first word, the first label indicating that the first word is ambiguous. The method may include identifying a first sentence of the text, the first sentence including the first word. The method may include determining first and second context of the first word and the first sentence. The method may include determining, based on the first and second context, using a machine learning model, a second label associated with the first sentence, the second label indicating a probability that the first sentence includes abusive language. The method may include generating second video content for presentation.

Abstract: Methods and apparatus are relating to the use of machine learning techniques to identify unrated content that will be appealing to particular demographics. Using explicit feedback (e.g., star ratings) and implicit feedback (e.g., viewing behavior) for a given demographic, feature sets are extracted from the rated titles and then used to recognize similar titles in unrated content.

Abstract: Methods and apparatus are described that relate to the use of machine learning techniques to accurately predict ratings for various types of content across multiple mature themes.

Abstract: Techniques for providing video content are described, where portions of the video content may be determined based an artificial intelligence model. In an example, a computer system may input, to a neural network, a vector representation of a frame. The neural network may be configured based on an error of associating the frame with an incorrect category. The error may be based on timing of the frame and on an error time margin relative to a start time or an end time of the incorrect category. The computer system may receive back an association between the vector representation and a category and may generate, based on the association, an identifier of a portion of the video content. The identifier may associate the portion of the video content with the category and identify a start time or an end time of the portion of the video content.

Abstract: Techniques are described for a content rating system that allows for automatic assignment of maturity ratings for media content.

Abstract: An security system for an enterprise network and data automates the revision, deployment, enforcement, auditing and control of security policies on mobile devices connected to said enterprise network, through automated communication between a security policy server and the mobile device. Control of the security system is centralized through administrative control of security policies stored on the security policy server. Automation of deployment of security policies to mobile devices occurs through transparent background communication and transfer of updated policies either triggered by a change in a security policy within the central repository of security policies or upon the expiration of a certain time period during which no policies were downloaded to the mobile device. When the mobile device is not in compliance with a security policy, a software security agent operating thereon limits access to said enterprise network and enterprise data.