Browse Prior Art Database

SCAM - Method for Detection of Confidential Communication Disclosure Number: IPCOM000249021D
Publication Date: 2017-Jan-26
Document File: 2 page(s) / 61K

Publishing Venue

The Prior Art Database


Disclosed is a method to automatically classify and identify materials as confidential and then alert the senders and/or recipients to the presence of confidential content so that appropriate action can be taken.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.


SCAM - Method for Detection of Confidential Communication

People often do not realize which part of a communication is confidential and which is not. Confidentiality is not a one-size-fits-all process. Information that was confidential once may be made open later and vice-versa. Sometimes information may not be inherently confidential, but may be disseminated only on a need-to-know basis. It is not difficult for a person to receive some written document (e.g., email, chart, paper, etc.) that is marked as confidential and then carelessly or inadvertently disseminate that information in a spoke conversation. At other times, information can be declared confidential after people shared and disseminated it.

A method is needed to help make people aware of confidential parts of communications so that senders and/or recipients can take appropriate action (e.g., securing it in a device that is commensurate with the required protection needed for the level of confidentiality of the material or not disseminating it). A system is needed that can warn people when the content of a communication method (e.g., written message, voice message, etc.) includes or might include confidential material.

The novel contribution is method for detecting confidential information in communications. Through the following modules, the method:

· Applies Natural Language Processing (NLP) as well as processing of technical and specialized material to learn to identify confidential material and context that can be deemed confidential

· Creates a repository of areas/subjects and contexts identified as confidential. It also creates links to the original documents that it has used to learn the identified confidential material and context that can be deemed confidential.

· Updates the repository when certain additional confidential...