公开(公告)号：US12197846B2

公开(公告)日：2025-01-14

申请号：US16688084

申请日：2019-11-19

Applicant: International Business Machines Corporation

Inventor： Lalit Agarwalla ,  Gandhi Sivakumar ,  Maharaj Mukherjee ,  Rashida A. Hodge

IPC: G06F17/00 ,  G06F18/21 ,  G06F18/214 ,  G06F40/151 ,  G06F40/169 ,  G06F40/56 ,  G06N20/00

Abstract: Provided is a method, a computer program product, and a system for associating mathematical functions to numerical text in a natural language sample. The method includes inputting a natural language sample from a text dataset and identifying a numerical text within the natural language sample. The method further includes displaying a mathematical function corresponding to the numerical text to be selected. The mathematical function can be selected via graphical user interface displayed on a computing device. The method also includes receiving and inserting the mathematical function as a feature into a feature vector of the natural language sample and selecting an output label for the natural language sample. The output label relates to the mathematical function selected for the numerical text. The method further includes exporting the natural language sample into a labeled dataset which can be used to train a machine learning model.

公开(公告)号：US12147768B2

公开(公告)日：2024-11-19

申请号：US17324043

申请日：2021-05-18

Applicant: International Business Machines Corporation

Inventor： Gandhi Sivakumar ,  Naeem Altaf ,  Wesley M Devine ,  Rizwan Dudekula

IPC: G06F40/30 ,  G06F16/9032 ,  G06N20/00

Abstract: A method, apparatus and computer program for detecting natural language (NL) bias by a conversational system is described. Embodiments of the invention determine an NL bias in a set of training questions used to train a machine learning model used by the conversational system to select a user intent. Other embodiments of the invention determine an NL bias in a user question received by the conversation system as compared to the set of training questions. The NL bias causes the machine learning to preferentially associate user queries to a user intent. In respective embodiments, the system takes a corrective action to adjust the NL bias of the training questions or the user question.

公开(公告)号：US20240362380A1

公开(公告)日：2024-10-31

申请号：US18306300

申请日：2023-04-25

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Jacob Dixon ,  Steven Chandler Borrillo ,  Andrea Giovannini ,  Gandhi Sivakumar

IPC: G06F30/27 ,  G06Q30/018

CPC classification number: G06F30/27 ,  G06Q30/018

Abstract: A computer-implemented method, a computer system and a computer program product build a product model that indicates an electrostatic discharge (ESD) sensitivity level. The method includes identifying a sensitive component in a model of an electronics device and mapping the sensitive component to a location in the electronics device. The method also includes obtaining ESD metadata for the sensitive component. In addition, the method includes determining an ESD voltage threshold for the sensitive component based on the ESD metadata for the sensitive component. Lastly, the method includes generating the product model of the electronics device, wherein the product model includes a plurality of sensitive components mapped to a plurality of locations in the electronics device and an indication of the ESD voltage threshold for each sensitive component at a respective mapped location in the electronic device.

公开(公告)号：US20240338627A1

公开(公告)日：2024-10-10

申请号：US18297800

申请日：2023-04-10

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Anbumunee Ponniah ,  Gandhi Sivakumar ,  Yuhong Yin ,  Shweta Shandilya

IPC: G06Q10/0633 ,  G06Q10/0637 ,  G06Q10/0639

CPC classification number: G06Q10/0633 ,  G06Q10/06375 ,  G06Q10/06393

Abstract: Computer implemented method, systems, and computer program products include program code executing on a processor(s) mapping, based on analyzing two or more nodes over which tasks comprising the workflow are distributed, each node to one or more parameters utilized to evaluate efficacy of the workflow. The processor(s) determines, based on the mapping, costs associated with each node and costs associated with changes to each node. The processor(s) identifies tasks in the workflow that can be changed to counter disruptions impacting at least one parameter of the one or more parameters utilized to evaluate the efficacy of the workflow.

公开(公告)号：US20240313856A1

公开(公告)日：2024-09-19

申请号：US18183996

申请日：2023-03-15

Applicant: International Business Machines Corporation

Inventor： Gandhi Sivakumar ,  Kushal S. Patel ,  Sarvesh S. Patel

IPC: H04B7/185

CPC classification number: H04B7/18593 ,  H04B7/18586 ,  H04B7/18589

Abstract: Computer-implemented methods for securing data stored in a low earth orbit (LEO) data center are provided. Aspects include obtaining physical characteristics of a low earth orbit (LEO) data center and obtaining physical characteristics of one or more celestial objects. Aspects also include comparing the physical characteristics of the one or more celestial objects and the LEO data center. Based on a determination that the one or more celestial objects pose a risk to the LEO data center, aspects include backing up one or more data items stored on the LEO data center to another LEO data center in communication with the LEO data center.

公开(公告)号：US20240305362A1

公开(公告)日：2024-09-12

申请号：US18120032

申请日：2023-03-10

Applicant: International Business Machines Corporation

Inventor： Kushal S. Patel ,  Gandhi Sivakumar ,  Sarvesh S. Patel

IPC: H04B7/185

CPC classification number: H04B7/18513

Abstract: A computer-implemented method, according to one embodiment, includes obtaining satellite information associated with a plurality of satellites, and using the satellite information to determine physical parameters associated with the satellites as the satellites pass over geographical areas. The method further includes determining, from data associated with a plurality of internet based resources, a future gathering event at a first geographical area in which a plurality of connection requests will be output from user devices located at a first geographical area. Furthermore, the method includes generating, based on the physical parameters and the data, assignment instructions that specify which of the satellites is to accept the plurality of connection requests from the first geographical area during the gathering event. The assignment instructions are output for preventing the connection requests from going unfulfilled during the gathering event.

公开(公告)号：US20240235980A9

公开(公告)日：2024-07-11

申请号：US18048126

申请日：2022-10-20

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Kushal S. Patel ,  Gandhi Sivakumar ,  Sarvesh S. Patel

IPC: H04L45/02 ,  H04L41/0806

CPC classification number: H04L45/02 ,  H04L41/0806

Abstract: A computer-implemented method for processing data connection requests is provided. The computer-implemented method includes receiving a connection request from an edge device and responding to the edge device such that the edge device recognizes that a connection is established and that data cannot be sent. The method further includes receiving a response from the edge device indicative of a connection parameter, determining that the connection is fiber-allocatable in accordance with the connection parameter being of a first connection parameter type, selecting one fiber from a plurality of fibers to have the connection allocated thereto based on one or more rules of fiber selection responsive to the determining that the connection is fiber-allocatable and allocating the connection to the one fiber.

公开(公告)号：US11977540B2

公开(公告)日：2024-05-07

申请号：US17532277

申请日：2021-11-22

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Faried Abrahams ,  Gandhi Sivakumar ,  Vinod A. Valecha ,  Neal Fishman

IPC: G06F16/242 ,  G06F16/248 ,  G06F40/20

CPC classification number: G06F16/2425 ,  G06F16/248 ,  G06F40/20

Abstract: A method includes: creating, by a computing device, a virtual consolidation layer; replicating, by the computing device, snippets that are results of a query; copying, by the computing device, the replicated snippets to the virtual consolidation layer; performing, by the computing device and by a machine learning model, matching of particular ones of the snippets in the virtual consolidation layer; and producing, by the computing device, a search result comprising the matched snippets.

公开(公告)号：US20240143624A1

公开(公告)日：2024-05-02

申请号：US17976690

申请日：2022-10-28

Applicant: International Business Machines Corporation

Inventor： Kushal S. Patel ,  Gandhi Sivakumar ,  Sarvesh S. Patel

IPC: G06F16/27 ,  H04B7/185

CPC classification number: G06F16/275 ,  H04B7/185

Abstract: In one general embodiment, a computer-implemented method includes determining when two satellites in low Earth orbit (LEO) and having duplicate instances of data stored thereon will be within a predefined distance of one another. A third satellite is identified to store a replication copy of the instance of data for at least a duration that the two satellites will be within the predefined distance of one another. A copy of the instance of data is caused to be transmitted to the third satellite for storage thereon.

公开(公告)号：US11799794B2

公开(公告)日：2023-10-24

申请号：US17463446

申请日：2021-08-31

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Craig M. Trim ,  Faried Abrahams ,  Gandhi Sivakumar ,  Kushal S. Patel ,  Sarvesh S. Patel

IPC: H04L47/38 ,  H04W28/02

CPC classification number: H04L47/38 ,  H04W28/0263 ,  H04W28/0278 ,  H04W28/0289

Abstract: Selective compression of packet payload data in a 5G network includes receiving a congestion signal by user equipment (UE) connected to a broadband cellular network based on network traffic congestion in hardware of the network and indicating the network traffic congestion, determining current processor utilization of the UE, based on the received congestion signal, the determined current processor utilization, and known times for data compression and decompression, determining whether to automatically enable selective packet payload data compression on the UE, and based on determining to automatically enable selective packet payload data compression, compressing payload data of data packet(s) generated by the UE, and forwarding the data packet(s) with the compressed payload data for transmission on the broadband cellular network.