| No. | Publication Number | Title | Publication/Patent Number Publication/Patent Number |
Publication Date
Publication Date
|
Application Number Application Number |
Filing Date
Filing Date
|
Inventor Inventor | Assignee Assignee |
IPC
IPC
|
|||||
|
1 | US2021011934A1 |
EFFICIENT CORPUS SEARCH AND ANNOTATION MANAGEMENT FOR A QUESTION ANSWERING SYSTEM
|
Publication/Patent Number: US2021011934A1 | Publication Date: 2021-01-14 | Application Number: 16/509,937 | Filing Date: 2019-07-12 | Inventor： Boxwell, Stephen A. Frost, Keith G. Vernier, Stanley J. Brake, Kyle M. | Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION | IPC: G06F16/33 | Abstract: A computer converts a question received in a natural language format into a string of text elements. The computer searches a corpus comprising unstructured passages with the string of the text elements as search terms to identify a selection of unstructured passages from the corpus relevant to the text elements. The computer annotates the selection of relevant unstructured passages with one or more annotations according to at least one natural language annotation type to generate an annotated selection knowledge base. The computer modifies the string of text elements by annotating at least one of the text elements according to the at least one natural language annotation type. The computer searches the annotated selection knowledge base using the modified string of text elements to generate a selection of ranked passages. The computer identifies an answer to the question based on the selection of ranked passages. | |||
|
|
2 | US2021011973A1 |
Multi-Lingual Action Identification
|
Publication/Patent Number: US2021011973A1 | Publication Date: 2021-01-14 | Application Number: 16/507,143 | Filing Date: 2019-07-10 | Inventor： Motahari, Nezhad Hamid Reza Mahajan, Pravar Dilip | Assignee: International Business Machines Corporation | IPC: G06F17/27 | Abstract: Embodiments relate to an intelligent computer platform to identify and process communications across multiple languages. An originating communication is identified, including identification of language, action tokens, and linguistic features. A first map of the identified action tokens and linguistic features from the originating language to a second format is created and populated into identifying machine learning model (MLM). A second communication is identified, including the originating language, action tokens, and linguistic features, and a second map is created of the identified action tokens and linguistic features of the second communication. The second map and the MLM are leveraged to identify and return a predicted action token class of the identified action tokens in the second communication. | |||
|
|
3 | US10892193B2 |
Controlling active fin height of FinFET device
|
Publication/Patent Number: US10892193B2 | Publication Date: 2021-01-12 | Application Number: 16/585,798 | Filing Date: 2019-09-27 | Inventor： Song, Yi Baskar, Veeraraghavan S. Strane, Jay W. De, Silva Ekmini Anuja | Assignee: International Business Machines Corporation | IPC: H01L21/8238 | Abstract: Semiconductor devices and methods are provided to fabricate fin field-effect transistor (FinFET) devices having uniform fin height profiles. For example, uniformity of fin height profiles for FinFET devices is obtained by implementing a gate oxide removal process which is designed to prevent etching of an isolation layer (e.g., a shallow trench isolation layer) formed of an oxide material during removal of, e.g., sacrificial gate oxide layers of dummy gate structures during a replacement metal gate process. | |||
|
|
4 | US10892049B2 |
Triage of patient medical condition based on cognitive classification of medical images
|
Publication/Patent Number: US10892049B2 | Publication Date: 2021-01-12 | Application Number: 15/844,254 | Filing Date: 2017-12-15 | Inventor： Stoval, Iii William Murray Sati, Marwan Azabagic, Andjela Covell, Grant | Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION | IPC: G06K9/00 | Abstract: Methods and systems for automatically triaging an image study of a patient generated as part of a medical imaging procedure. One system includes a computing device including an electronic processor. The electronic processor is configured to submit at least a portion of the image study to a cognitive system, the cognitive system configured to analyze the image study using a model developed using machine learning, receive, from the cognitive system, a BI-RADS classification assigned to the image study using the model, and automatically triage the image study based on the classification assigned to the image study by the cognitive system. | |||
|
|
5 | US10892408B2 |
Multivalent oxide cap for analog switching resistive memory
|
Publication/Patent Number: US10892408B2 | Publication Date: 2021-01-12 | Application Number: 15/852,290 | Filing Date: 2017-12-22 | Inventor： Ando, Takashi Khater, Marwan H. Kim, Seyoung Miyazoe, Hiroyuki | Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION | IPC: H01L45/00 | Abstract: A resistive random access memory (RRAM), including a first electrode, a base oxide being connected to the first electrode, and a multivalent oxide being connected to the base oxide layer. The multivalent oxide switches oxidative states. | |||
|
|
6 | US10891391B2 |
Remote file storage with multiple access levels
|
Publication/Patent Number: US10891391B2 | Publication Date: 2021-01-12 | Application Number: 16/115,834 | Filing Date: 2018-08-29 | Inventor： Gordon, Itai Nizri, Miriam Prager, Ilan D. | Assignee: International Business Machines Corporation | IPC: G06F21/62 | Abstract: Disclosed embodiments provide techniques for accessing a document from a cloud storage system and controlling the display of sensitive data within the document based on user permissions. One or more restricted information segments are identified within a document to be stored on the cloud storage system. Restricted information segments can include anything within an electronic file for which it is desired to provide multiple levels of access. In some embodiments, the restricted information segments are automatically identified via computer-implemented natural language processing (NLP) techniques. For each restricted information segment, one or more alternative data sequences are generated. The alternative data sequences are encrypted using various keys residing on a client device associated with a user. The keys can be used to decrypt data stored within a multiple-value encrypted field structure. In this way, multiple levels of information can be disseminated to an audience with various permission/privilege levels. | |||
|
|
7 | US10891950B2 |
Graph based prediction for next action in conversation flow
|
Publication/Patent Number: US10891950B2 | Publication Date: 2021-01-12 | Application Number: 16/144,977 | Filing Date: 2018-09-27 | Inventor： Huang, Lei Moore, Robert J. Ren, Guangjie Jiang, Shun | Assignee: International Business Machines Corporation | IPC: G10L15/22 | Abstract: One embodiment provides a method for predicting a next action in a conversation system that includes obtaining, by a processor, information from conversation logs and a conversation design. The processor further creates a dialog graph based on the conversation design. Weights and attributes for edges in the dialog graph are determined based on the information from the conversation logs. An unrecognized user input the conversation system is detected. The unrecognized user input is analyzed and a next action is predicted based on dialog nodes in the dialog graph and historical paths. A guiding conversation response is generated based on the predicted next action. | |||
|
|
8 | US10892056B2 |
Artificial intelligence based alert system
|
Publication/Patent Number: US10892056B2 | Publication Date: 2021-01-12 | Application Number: 16/193,396 | Filing Date: 2018-11-16 | Inventor： Xie, Yiting Graf, Ben Sitek, Arkadiusz | Assignee: International Business Machines Corporation | IPC: G06K9/00 | Abstract: A mechanism is provided to implement an artificial intelligence (AI) based alert mechanism system for alerting a medical professional of potential inaccuracies in medical image analysis. Responsive to receiving a medical image of a patient and a radiology report associated with the medical image, the AI based alert mechanism analyzes the radiology report to identify medical findings detected by the medical professional and analyzes the medical image to detect one or more medical findings associated with the medical image. Responsive to the AI based alert mechanism identifying one or more medical findings, the AI based alert mechanism compares the identified medical findings to those medical findings identified in the radiology report. Responsive to the AI based alert mechanism identifying a discrepancy between the identified medical findings to those in the radiology report, the AI based alert mechanism generates an alert to the medical professional who generated the radiology report. | |||
|
|
9 | US10892050B2 |
Deep image classification of medical images
|
Publication/Patent Number: US10892050B2 | Publication Date: 2021-01-12 | Application Number: 15/952,361 | Filing Date: 2018-04-13 | Inventor： Zhang, Xin Xie, Guo Tong Zhang, Xiaolu Li, Xiu Li Gao, Peng | Assignee: International Business Machines Corporation | IPC: G06K9/62 | Abstract: Methods, systems, and storage components for utilizing a deep neural network(s) for classifying at least one medical image. A deep neural network (DNN) can be configured by an image processing component to contain at least one revived neuron, where the revived neuron has an adjusted value based on a focus-learning function that is transferred to the DNN by the image processing component, where the focus-learning function provides the adjustment by updating the DNN with data that contains a corrected classification with respect to at least one normal image sample derived from a medical image, and where the correction is based on the focus-learning function comparing an annotation associated with an abnormal image sample derived from the medical image to another annotation associated with the at least one normal image. | |||
|
|
10 | US10892181B2 |
Semiconductor device with mitigated local layout effects
|
Publication/Patent Number: US10892181B2 | Publication Date: 2021-01-12 | Application Number: 16/807,716 | Filing Date: 2020-03-03 | Inventor： Zhou, Huimei Tsutsui, Gen Basker, Veeraraghavan S. Greene, Andrew M. Guo, Dechao Bu, Huiming Vega, Reinaldo | Assignee: International Business Machines Corporation | IPC: H01L27/088 | Abstract: Integrated chips include a semiconductor fin that has a first active region and a second active region that are electrically separated by an oxide region that completely penetrates the semiconductor fin. A first semiconductor device is formed on the first active region. A second semiconductor device formed on the second active region. | |||
|
|
11 | US10892403B2 |
Structured bottom electrode for MTJ containing devices
|
Publication/Patent Number: US10892403B2 | Publication Date: 2021-01-12 | Application Number: 16/261,051 | Filing Date: 2019-01-29 | Inventor： Marchack, Nathan P. Doris, Bruce B. Hashemi, Pouya | Assignee: International Business Machines Corporation | IPC: H01L43/12 | Abstract: A bottom electrode structure for a magnetic tunnel junction (MTJ) containing device is provided. The bottom electrode structure includes a mesa portion that is laterally surrounded by a recessed region. The recessed region of the bottom electrode structure is laterally adjacent to a dielectric material, and a MTJ pillar is located on the mesa portion of the bottom electrode structure. Such a configuration shields the recessed region from impinging ions thus preventing deposition of resputtered conductive metal particles from the bottom electrode onto the MTJ pillar. | |||
|
|
12 | US2021012221A1 |
ESTIMATING FEASIBILITY AND EFFORT FOR A MACHINE LEARNING SOLUTION
|
Publication/Patent Number: US2021012221A1 | Publication Date: 2021-01-14 | Application Number: 16/508,698 | Filing Date: 2019-07-11 | Inventor： Perera, Pathirage Dinindu Sujan Udayanga Raz, Orna Routray, Ramani Farchi, Eitan Daniel | Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION | IPC: G06N5/04 | Abstract: A method, computer system, and a computer program product for assessing a likelihood of success associated with developing at least one machine learning (ML) solution is provided. The present invention may include generating a set of questions based on a set of raw training data. The present invention may also include computing a feasibility score based on an answer corresponding with each question from the generated set of questions. The present invention may then include, in response to determining that the computed feasibility score satisfies a threshold, computing a level of effort associated with developing the at least one ML solution to address a problem. The present invention may further include presenting, to a user, a plurality of results associated with assessing the likelihood of success of the at least one ML solution. | |||
|
|
13 | US2021012030A1 |
DE-IDENTIFYING SOURCE ENTITY DATA
|
Publication/Patent Number: US2021012030A1 | Publication Date: 2021-01-14 | Application Number: 16/505,023 | Filing Date: 2019-07-08 | Inventor： Soni, Kalp Singh, Prashant | Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION | IPC: G06F21/62 | Abstract: De-identification of source entity data is provided, in which a process obtains source entity data having an entity identifier and entity attribute data in respective formats for entity attributes of that entity, obtains a predefined library of fictitious entities with corresponding fictitious entity attribute data, for the entity attributes, that is in the respective formats, and generates a de-identified entity lookup list by applying a transformation to the entity identifier to produce a transformed entity identifier, selecting a fictitious entity from the predefined library of fictitious entities, and writing to the de-identified entity lookup list the entity identifier, the transformed entity identifier, and the fictitious entity attribute data, for the entity attributes, that corresponds to the selected fictitious entity. The process also processes entity file(s) using the de-identified entity lookup list to de-identify the entity and produces de-identified entity file(s). | |||
|
|
14 | US2021012496A1 |
VASCULAR DISSECTION DETECTION AND VISUALIZATION USING A SUPERIMPOSED IMAGE WITH AN IMPROVED ILLUSTRATION OF AN OUTERMOST VESSEL PERIPHERY
|
Publication/Patent Number: US2021012496A1 | Publication Date: 2021-01-14 | Application Number: 16/507,840 | Filing Date: 2019-07-10 | Inventor： Bronkalla, Mark D. Graf, Ben Sitek, Arkadiusz Xie, Yiting | Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION | IPC: G06T7/00 | Abstract: Methods and systems for detecting a dissection of an elongated structure in a three dimensional medical image. One system includes an electronic processor that receives the three dimensional medical image. The electronic processor determines a first periphery and a second periphery of the elongated structure, the first periphery and the second periphery associated with an enhancing part and a non-enhancing part, respectively, of the elongated structure. The electronic processor determines whether the first periphery or the second periphery best illustrates an outermost periphery of the elongated structure and generate a base image based on whether the first periphery or the second periphery best illustrates an outermost periphery of the elongated structure. The electronic processor superimposes a contrast image associated with the three dimensional image on top of the base image, detects a dissection in the elongated structure using the superimposed image, and outputs a medical report identifying the dissection. | |||
|
|
15 | US2021014652A1 |
Automatic Query-Based Communication System
|
Publication/Patent Number: US2021014652A1 | Publication Date: 2021-01-14 | Application Number: 16/508,438 | Filing Date: 2019-07-11 | Inventor： Shivashankaraiah, Nataraj | Assignee: International Business Machines Corporation | IPC: H04W4/18 | Abstract: An incoming message is responded to by a computer system. The incoming message is received by the computer system. Natural language processing is performed on the incoming message by the computer system to generate a query in response to an automatic response to the incoming message being enabled for an originator of the incoming message. Information responsive to the query and contextual information related to the query from information sources are identified by the computer system using the query generated from the natural language processing of the incoming message. An outgoing message is generated by the computer system using the information and the contextual information. The outgoing message is sent by the computer system to the originator of the incoming message. | |||
|
|
16 | US2021013321A1 |
SPACER-CONFINED EPITAXIAL GROWTH
|
Publication/Patent Number: US2021013321A1 | Publication Date: 2021-01-14 | Application Number: 16/504,739 | Filing Date: 2019-07-08 | Inventor： Yogendra, Karthik Rahman, Ardasheir Robison, Robert Carr, Adra | Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION | IPC: H01L29/66 | Abstract: Semiconductor devices and methods of forming the same include forming a dummy gate on a stack of alternating channel layers and sacrificial layers. A spacer layer is formed over the dummy gate and the stack. Portions of the spacer layer on horizontal surfaces of the stack are etched away to form vertical spacers. Exposed portions of the stack are etched away. Semiconductor material is grown from exposed sidewalls of remaining channel layers to form source and drain structures that are constrained in lateral dimensions by the vertical spacers. | |||
|
|
17 | US10890460B2 |
Navigation and location validation for optimizing vehicle-based transit systems
|
Publication/Patent Number: US10890460B2 | Publication Date: 2021-01-12 | Application Number: 16/165,307 | Filing Date: 2018-10-19 | Inventor： Marzorati, Mauro Diamanti, Gary F. Jenkins, Jana H. Özses, Seda | Assignee: International Business Machines Corporation | IPC: G01C21/34 | Abstract: Methods and systems are provided for navigating to a pre-scheduled vehicle. A code is analyzed for a user including location information pertaining to the pre-scheduled vehicle, wherein the location information indicates a destination for the user. A location of the user is determined based on a GPS sensor. The user is directed to traverse a route between the location of the user and a location of the pre-scheduled vehicle. A destination of the pre-scheduled vehicle is verified as being consistent with the destination for the user indicated by the location information of the code. | |||
|
|
18 | US10891813B2 |
Cognitive vetting system for unsolicited residential visitors
|
Publication/Patent Number: US10891813B2 | Publication Date: 2021-01-12 | Application Number: 16/841,960 | Filing Date: 2020-04-07 | Inventor： Carbonell, Lee A. Cheng, Tsz Edgington, Jeff L. Mariadoss, Pandian | Assignee: International Business Machines Corporation | IPC: G07C9/00 | Abstract: A method, computer system, and a computer program product for cognitive vetting is provided. The present invention may include determining that an identified person is not expected. The present invention may include retrieving historical information relevant to a past interaction with the identified person. The present invention may then include enabling a present interaction with the identified person. The present invention may lastly include determining a recommendation based on an analysis of the retrieved historical information and the present interaction. | |||
|
|
19 | US10891181B2 |
Smart system dump
|
Publication/Patent Number: US10891181B2 | Publication Date: 2021-01-12 | Application Number: 16/171,082 | Filing Date: 2018-10-25 | Inventor： Paruthi, Meghna Duron, Mike C. Kobylak, Adriana Gurram, Nagendra K. Albert, Jason | Assignee: International Business Machines Corporation | IPC: G06F11/07 | Abstract: Techniques are disclosed for performing dump collection on a computing system. One technique includes detecting an error event on the computing system. After detecting the error event, a subset of hardware registers associated with the error event are determined. One or more hardware units within the computing system to capture data from are determined from a set of rules that specify an association between the one or more hardware units and the subset of hardware registers associated with the error event. Data is captured from each of the one or more hardware units. | |||
|
|
20 | US10891654B2 |
Location-based advertising using hybrid radio
|
Publication/Patent Number: US10891654B2 | Publication Date: 2021-01-12 | Application Number: 15/064,657 | Filing Date: 2016-03-09 | Inventor： Jhoney, Albee Meti, Mahantesh S. Sharma, Kalpesh | Assignee: International Business Machines Corporation | IPC: G06Q30/02 | Abstract: A method for generating location-based advertisement includes receiving a selection of a first radio station from a user of a mobile device, receiving an electronic program guide from the first radio station including a plurality of advertisement time slots, where each advertisement time slot in the plurality of advertisement time slots includes an advertisement sequence for a first advertisement content, receiving a second advertisement content from a second radio station located in a geographic location matching a real-time geographic location of the user as determined by a location-based service of the mobile device, substituting the first advertisement content in the advertisement sequence of one of the plurality of advertisement time slots with the second advertisement content from the second radio station, and broadcasting the second advertisement content to the user via the mobile device. |
3.Current Quota