Abstract: A method, computer program product, and computer system for redacting information in a video stream. The method includes accessing content of an original video stream and classifying sensitive data within the original video stream as classified data components of one or more of multiple permission categories. The method provides sub-streams of the classified data components for one or more permission categories. The method redacts the classified data components in the original video stream and combines the redacted video stream with one or more of the sub-streams to output a combined video stream depending on the access permissions of an endpoint.

Abstract: Exposure risk quantification is provided. The method comprises a first device recording audio data responsive to a wireless proximity contact event with a second device and performing a first Fourier transform on the recorded audio data. The first device sends the first Fourier transform to the second device and receives a second Fourier transform of audio data recorded concurrently by the second device in response to the wireless proximity contact event. The first device, computes a cross correlation of the Fourier transforms and finding a maximum of the cross correlation that constitutes an exposure score. The first device compares the exposure score to a specified threshold that determines whether or not users of the first and second devices have had unacceptable exposure to each other and outputs the exposure score and a binary threshold result.

Abstract: A processor may analyze one or more unread messages. The one or more unread messages may be from one or more respective conversations. The processor may generate, from the analyzing, an ambiguous strings model. The processor may summarize, via utilization of the ambiguous strings model, the one or more unread messages. The processor may genericize the ambiguous strings model.

Abstract: A processor may analyze one or more unread messages. The one or more unread messages may be from one or more respective conversations. The processor may generate, from the analyzing, an ambiguous strings model. The processor may summarize, via utilization of the ambiguous strings model, the one or more unread messages. The processor may genericize the ambiguous strings model.