公开(公告)号：US11223663B1

公开(公告)日：2022-01-11

申请号：US16918531

申请日：2020-07-01

Applicant: Adobe Inc.

Inventor： Niranjan Shivanand Kumbi ,  Varinder Kumar ,  Uddhab Pant ,  Aditya Bindal ,  Amit Gupta ,  Lakshay Tanwar ,  Reddy Sreekanth ,  Ajay Awatramani

IPC: H04L29/06 ,  G06F3/048 ,  H04L29/08

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for initiating electronic chats based on conversation workflows identified in response to detected user actions in connection with an embedded document container displaying a PDF file. In particular, in one or more embodiments, the disclosed systems detect user interactions with a PDF file displayed by a document container embedded in a webpage. The disclosed systems can determine whether the detected user interactions include or indicate a conversation workflow trigger associated with a conversation workflow. The disclosed systems can further generate electronic messages based on the conversation workflow and provide the generated electronic messages to the user in connection with the webpage where the document container is embedded.

公开(公告)号：US20220006846A1

公开(公告)日：2022-01-06

申请号：US16918531

申请日：2020-07-01

Applicant: Adobe Inc.

Inventor： Niranjan Shivanand Kumbi ,  Varinder Kumar ,  Uddhab Pant ,  Aditya Bindal ,  Amit Gupta ,  Lakshay Tanwar ,  Reddy Sreekanth ,  Ajay Awatramani

IPC: H04L29/06 ,  H04L29/08 ,  G06F3/048

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for initiating electronic chats based on conversation workflows identified in response to detected user actions in connection with an embedded document container displaying a PDF file. In particular, in one or more embodiments, the disclosed systems detect user interactions with a PDF file displayed by a document container embedded in a webpage. The disclosed systems can determine whether the detected user interactions include or indicate a conversation workflow trigger associated with a conversation workflow. The disclosed systems can further generate electronic messages based on the conversation workflow and provide the generated electronic messages to the user in connection with the webpage where the document container is embedded.

公开(公告)号：US11025713B2

公开(公告)日：2021-06-01

申请号：US16384646

申请日：2019-04-15

Applicant: ADOBE INC.

Inventor： Niranjan Shivanand Kumbi ,  Ajay Awatramani

IPC: G06F15/173 ,  H04L29/08 ,  G06Q30/02 ,  G06F9/50 ,  G06N20/00

Abstract: An improved marketing automation system can optimize governance of server resources by managing the execution of campaigns. The marketing automation system can develop intelligence around a given customer's inflow of incoming campaigns, the execution time of the campaigns, and general resource utilization over time. The marketing automation system can learn to predict an expected number and type of campaigns for a pre-defined window of time. This intelligence can be leveraged to ensure that one or more executors remain available to execute predicted high priority campaigns upon placement into an execution queue. Further, this intelligence can be applied such that predicted dormant executors can be used to execute low priority tasks. In this way, the marketing automation system minimizes queue time until execution for high priority campaigns while optimizing use of server resources.

公开(公告)号：US20200320381A1

公开(公告)日：2020-10-08

申请号：US16375037

申请日：2019-04-04

Applicant: ADOBE INC.

Inventor： Vaidyanathan Venkatraman ,  Rajan Madhavan ,  Omar Rahman ,  Niranjan Shivanand Kumbi ,  Brajendra Kumar Bhujabal ,  Ajay Awatramani

IPC: G06N3/08 ,  G06Q10/06

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media for providing factors that explain the generated results of a deep neural network (DNN). In embodiments, multiple machine learning models and a DNN are trained on a training dataset. A preliminary set of trained machine learning models with similar results to the trained DNN are selected for further evaluation. The preliminary set of machine learning models may be evaluated using a distribution analysis to select a reduced set of machine learning models. Results produced by the reduced set of machine learning models are compared, point-by-point, to the results produced by the DNN. The best performing machine learning model with generated results that performs closest to the DNN generated results may be selected. One or more factors used by the selected machine learning model are determined. Those one or more factors from the selected best performing machine learning model may be provided to explain the results of the DNN and increase confidence in the understanding and accuracy of the results generated by the DNN.