公开(公告)号：US20240012837A1

公开(公告)日：2024-01-11

申请号：US18168358

申请日：2023-02-13

Applicant: Oracle International Corporation

Inventor： Qian Rui Chow ,  Donald Creig Humes ,  Kaarthik Balasubrahmanian ,  Sridhar Tadepalli ,  Saravanan Anandan ,  Kartik Raghavan

IPC: G06F16/31 ,  G06F40/30 ,  G06F16/383 ,  G06N20/00

CPC classification number: G06F16/313 ,  G06F40/30 ,  G06F16/383 ,  G06N20/00

Abstract: Techniques for initiating system actions based on text content are disclosed. A system applies a semantic analysis model at run-time to human-understandable text to identify actionable content within the human-understandable text. The system analyzes metadata associated with the text to identify a mapping between one or more data objects associated and the semantic content in the text. The system identifies one or more contact lists associated, respectively, with the one or more data objects. Using a database schema, the system analyzes the contact lists to determine whether to modify the contact lists.

公开(公告)号：US20220197906A1

公开(公告)日：2022-06-23

申请号：US17130099

申请日：2020-12-22

Applicant: Oracle International Corporation

Inventor： Pradeep Krishnappa Ajanahalli ,  Sanjay Anant Pusegaonkar ,  Atanu Majumdar ,  Kartik Raghavan

IPC: G06F16/2453 ,  G06F16/22 ,  G06F16/901

Abstract: A system filters a data set for the efficient execution of queries by removing at least some irrelevant records to generate a reduced data set. The records are determined to be irrelevant based on the values of the records for the particular attribute(s) corresponding to the filtering operation. The system filters out records within a data set that do not include any of the values defined for the particular attribute by a set of queries that are to be executed on the filtered set of records.

公开(公告)号：US20240273556A1

公开(公告)日：2024-08-15

申请号：US18168842

申请日：2023-02-14

Applicant: Oracle International Corporation

Inventor： Vivek Kumar ,  Kartik Raghavan

IPC: G06Q30/0201 ,  G06N20/00

CPC classification number: G06Q30/0201 ,  G06N20/00

Abstract: Techniques for enterprise data aggregation and collective insights generation are disclosed, including: receiving multiple sets of enterprise-specific customer relationship management (CRM) data, respectively, from enterprise-specific CRMs; aggregating the sets of enterprise-specific CRM data, to obtain collective CRM data; receiving a request to generate a collective insight that is applicable to a particular subset of the collective CRM data; responsive to the request, selecting a particular machine learning model from multiple machine learning models, wherein each machine learning model is configured to generate collective insights for a respective subset of the collective CRM data; generating the collective insight that is applicable to the particular subset of the collective CRM data, using the particular machine learning model.

公开(公告)号：US11823667B2

公开(公告)日：2023-11-21

申请号：US17670891

申请日：2022-02-14

Applicant: Oracle International Corporation

Inventor： Raja V. Penneru ,  Kaarthik Balasubrahmanian ,  Donald Creig Humes ,  Sridhar Tadepalli ,  Saravanan Anandan ,  Kartik Raghavan

IPC: G10L15/18 ,  G10L15/22 ,  G06F3/16

CPC classification number: G10L15/1815 ,  G06F3/167 ,  G10L15/22

Abstract: Techniques for modifying rules for completing requests using a conversational interface are disclosed. A conversational interface receives a user request to perform a particular function. A system extracts semantic content from the request and obtains contextual metadata associated with the request. Using the semantic content and contextual metadata, the system generates a data set of attribute values. The system determines whether the data set matches a data set of required data attributes associated with the request. If the provided data set is different than the required data set, the system determines whether to modify a set of rules associated with the request. The set of rules specifies which attribute values are required to complete the request. When the system modifies the rules associated with the request, the system also determines whether to apply the rule modification to subsequent requests by the same user and other users.

公开(公告)号：US20230065223A1

公开(公告)日：2023-03-02

申请号：US17670891

申请日：2022-02-14

Applicant: Oracle International Corporation

Inventor： Raja V. Penneru ,  Kaarthik Balasubrahmanian ,  Donald Creig Humes ,  Sridhar Tadepalli ,  Saravanan Anandan ,  Kartik Raghavan

IPC: G10L15/18 ,  G10L15/22 ,  G06F3/16

Abstract: Techniques for modifying rules for completing requests using a conversational interface are disclosed. A conversational interface receives a user request to perform a particular function. A system extracts semantic content from the request and obtains contextual metadata associated with the request. Using the semantic content and contextual metadata, the system generates a data set of attribute values. The system determines whether the data set matches a data set of required data attributes associated with the request. If the provided data set is different than the required data set, the system determines whether to modify a set of rules associated with the request. The set of rules specifies which attribute values are required to complete the request. When the system modifies the rules associated with the request, the system also determines whether to apply the rule modification to subsequent requests by the same user and other users.

公开(公告)号：US11537611B2

公开(公告)日：2022-12-27

申请号：US17130099

申请日：2020-12-22

Applicant: Oracle International Corporation

Inventor： Pradeep Krishnappa Ajanahalli ,  Sanjay Anant Pusegaonkar ,  Atanu Majumdar ,  Kartik Raghavan

IPC: G06F16/24 ,  G06F16/2453 ,  G06F16/901 ,  G06F16/22

Abstract: A system filters a data set for the efficient execution of queries by removing at least some irrelevant records to generate a reduced data set. The records are determined to be irrelevant based on the values of the records for the particular attribute(s) corresponding to the filtering operation. The system filters out records within a data set that do not include any of the values defined for the particular attribute by a set of queries that are to be executed on the filtered set of records.