公开(公告)号：US10990759B1

公开(公告)日：2021-04-27

申请号：US16050488

申请日：2018-07-31

Applicant: Amazon Technologies, Inc.

Inventor： Mikhail Sosonkin ,  Michael Egan Butler ,  Ken F Eggers

IPC: G06F40/30 ,  G06Q10/06 ,  H04L9/32 ,  G06N20/00 ,  G06F16/2457

Abstract: This disclosure describes techniques implemented partly by a cloud-based service provider for providing a sentiment-analysis service to determine whether sentiments, or attitudes, of users have significantly decreased. The sentiment-analysis service is composed of various services provided by the cloud-based service in order to determine significant negative deviations in sentiment of users towards an entity. Further, the sentiment-analysis service utilizes various techniques, such as encryption, to privatize the process such that private user information is protected. In some examples, an entity may not know the content of communications or the reasons the user's sentiment has dropped, but may be alerted of a general negative change in sentiment towards the entity by the user.

公开(公告)号：US10740151B1

公开(公告)日：2020-08-11

申请号：US16113629

申请日：2018-08-27

Applicant: Amazon Technologies, Inc.

Inventor： Ryan Washington ,  Joe W. Pate ,  David Walker ,  Scott Conrad ,  Mikhail Sosonkin ,  Matthew Evans ,  Nathan Kevin McCarthy ,  Hugo Gabignon ,  Victor Chin ,  Joel Naomi Cornett ,  Joshua Stephen Du Lac

IPC: G06F9/46 ,  G06F9/50 ,  H04L12/58 ,  G06F9/54 ,  G06F16/14

Abstract: Systems, methods, and computer-readable media are disclosed for systems and methods for parallelized forensic analysis using cloud-based servers. Example methods may include generating a first request for one or more notifications in a notification queue, where the one or more notifications include a first notification indicative of a first data input at a datastore, determining a first data type of the first data input, and generating a second notification indicative of the first data type. Some example methods may include determining that a first software component is subscribed to notifications for the first data type, sending the second notification to the first software component, determining a first output of the first software component, where the first output comprises a set of extracted data from the first data input, and sending the set of extracted data to the datastore.

公开(公告)号：US10740565B1

公开(公告)日：2020-08-11

申请号：US16050488

申请日：2018-07-31

Applicant: Amazon Technologies, Inc.

Inventor： Mikhail Sosonkin ,  Michael Egan Butler ,  Ken F Eggers

IPC: G06F40/30 ,  G06Q10/06 ,  H04L9/32 ,  G06N20/00 ,  G06F16/2457

Abstract: This disclosure describes techniques implemented partly by a cloud-based service provider for providing a sentiment-analysis service to determine whether sentiments, or attitudes, of users have significantly decreased. The sentiment-analysis service is composed of various services provided by the cloud-based service in order to determine significant negative deviations in sentiment of users towards an entity. Further, the sentiment-analysis service utilizes various techniques, such as encryption, to privatize the process such that private user information is protected. In some examples, an entity may not know the content of communications or the reasons the user's sentiment has dropped, but may be alerted of a general negative change in sentiment towards the entity by the user.