-
公开(公告)号：US11769007B2
公开(公告)日：2023-09-26
申请号：US17303349
申请日：2021-05-27
Applicant: International Business Machines Corporation
Inventor： Yousef El-Kurdi , Radu Florian , Hiroshi Kanayama , Efsun Kayi , Laura Chiticariu , Takuya Ohko , Robert Todd Ward
IPC: G06F40/205 , G06N3/04 , G06N3/088 , G06F40/47
CPC classification number: G06F40/205 , G06F40/47 , G06N3/04 , G06N3/088
Abstract: An approach for generating synthetic treebanks to be used in training a parser in a production system is provided. A processor receives a request to generate one or more synthetic treebanks from a production system, wherein the request indicates a language for the one or more synthetic treebanks. A processor retrieves at least one corpus of text in which the requested language is present. A processor provides the at least one corpus to a transformer enhanced parser neural network model. A processor generates at least one synthetic treebank associated with a string of text from the at least one corpus of text in which the requested language is present. A processor sends the at least one synthetic treebank to the production system, wherein the production system trains a parser utilized by the production system with the at least one synthetic treebank.
-
公开(公告)号：US11651270B2
公开(公告)日：2023-05-16
申请号：US15077303
申请日：2016-03-22
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor： Vittorio Castelli , Radu Florian , Taesun Moon , Avirup Sil
IPC: G06N20/00 , G06N20/20 , G06F16/332
CPC classification number: G06N20/00 , G06F16/3326 , G06N20/20
Abstract: A method and system are provided for combining models. The method includes forming, by a computer having a processor and a memory, model pairs from a model ensemble that includes a plurality of models. The method further includes comparing the model pairs based on sets of output results produced by the model pairs to provide comparison results. The method also includes constructing, by the computer, a combination model from at least one of the model pairs based on the comparison results. The comparing step is performed using user-generated set-based feedback.
-
公开(公告)号：US10698964B2
公开(公告)日：2020-06-30
申请号：US15419615
申请日：2017-01-30
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor： Vittorio Castelli , Radu Florian , Xiaoqiang Luo , Hema Raghavan
IPC: G06F40/295 , G06F16/28 , G06F16/9535 , G06F16/338 , G06F16/33 , G06F16/35 , G06F16/36
Abstract: A method for automatically extracting and organizing information by a processing device from a plurality of data sources is provided. A natural language processing information extraction pipeline that includes an automatic detection of entities is applied to the data sources. Information about detected entities is identified by analyzing products of the natural language processing pipeline. Identified information is grouped into equivalence classes containing equivalent information. At least one displayable representation of the equivalence classes is created. An order in which the at least one displayable representation is displayed is computed. A combined representation of the equivalence classes that respects the order in which the displayable representation is displayed is produced.
-
公开(公告)号：US10656602B2
公开(公告)日：2020-05-19
申请号：US16285380
申请日：2019-02-26
Applicant: International Business Machines Corporation
Inventor： Vittorio Castelli , Radu Florian , Gourab Kundu , Taesun Moon , Avirup Sil
Abstract: A power tool accident prevention system receiving images from a static camera of a setup of a power tool, the system comprising: a processor; and a memory, the memory storing instructions to cause the processor to: analyze the images to identify inherent dangers in the setup of the power tool; identify at least one potential cause of an accident based on the identified inherent dangers; and activate an emergency safety measure of the power tool to avoid the at least one potential cause of the accident.
-
公开(公告)号：US10051426B2
公开(公告)日：2018-08-14
申请号：US15220610
申请日：2016-07-27
Applicant: International Business Machines Corporation
Inventor： Vittorio Castelli , Georgiana Dinu , Radu Florian , Gourab Kundu , Taesun Moon , Avirup Sil
Abstract: Embodiments are directed to a computer implemented method and system of proactively caching content for a mobile electronic device. The method includes determining the location of a mobile electronic device, predicting the location of the mobile electronic device, including a prediction of when the mobile electronic device will be in a location of low or no network connectivity. The data that will be retrieved while the mobile electronic device is not connected to the network is also predicted. Thereafter, the predicted data that will be retrieved is downloaded prior to losing network connectivity so that it can be accessed from memory while the mobile electronic device has low or no network connectivity. Other embodiments are also described.
-
Patent Agency Ranking
公开(公告)号：US09898654B2
公开(公告)日：2018-02-20
申请号：US15084035
申请日：2016-03-29
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor： Vittorio Castelli , Georgiana Dinu , Radu Florian , Gourab Kundu , Taesun Moon , Avirup Sil
CPC classification number: G06K9/00476 , G06F17/30011 , G06F17/30569 , G06F17/30657 , G06K9/00355
Abstract: A method and system are provided for assisting a user performing a procedure. The method includes capturing, by a camera, images of user activity while the user is performing the procedure. The method further includes converting, by computer processing system, the images of user activity into a text representation of user activity. The method also includes comparing, by the computer processing system, the textual representation of user activity to procedure documentation. The method additionally includes at least one of visually and audibly indicating, by a display and a speaker, a corrective action to the user responsive to a mismatch result from said comparing step.
-
公开(公告)号：US20180033320A1
公开(公告)日：2018-02-01
申请号：US15220166
申请日：2016-07-26
Applicant: International Business Machines Corporation
Inventor： Vittorio Castelli , Georgiana Dinu , Radu Florian , Avirup Sil
CPC classification number: G08G5/065 , B63G2008/005 , B64C39/024 , B64C2201/12 , B64D47/08 , B64F5/60 , G05D1/0083 , G08G5/0008 , G08G5/0021 , G08G5/0069 , G08G5/0078 , G08G5/045
Abstract: Techniques for drone device control are provided. In one example, a computer-implemented method comprises: meeting, by a drone device operatively coupled to a processor, an aircraft at a first location; and guiding, by the drone device, the aircraft to a second location along a ground movement path selected from a plurality of ground movement paths associated with an airport. The guiding can comprise providing a direction indication to the aircraft; and monitoring a defined region around the aircraft for one or more hazards. The guiding can also comprise, in response to identifying a hazard from the one or more hazards related to the defined region around the aircraft, providing a hazard indication to the aircraft.
-
8.发明申请公开(公告)号：US20170269930A1
公开(公告)日：2017-09-21
申请号：US15075262
申请日：2016-03-21
Applicant: International Business Machines Corporation
Inventor： Vittorio Castelli , Radu Florian , Taesun Moon , Avirup Sil
CPC classification number: G06F8/73 , G06F17/30011 , G06F17/30719 , H04L51/046 , H04L51/16
Abstract: A project documentation method, system, and non-transitory computer readable medium, include a matching circuit configured to match the multimodal communications between users stored in the database to the project, an identification circuit configured to associate a chat thread of the multimodal communications to a sub-project of the project, a relating circuit configured to relate words of the chat thread and words in text of the project, an extracting and creating circuit configured to extract text of the chat thread that is relevant to the text of the project and create a document including the relevant text of the chat thread, and a decision circuit configured to decide whether to update the document including the relevant text created by the extracting and creating circuit with newly extracted text based on a similarity between data of the document and the newly extracted text to avoid redundancies within the created document.
-
公开(公告)号：US20170140057A1
公开(公告)日：2017-05-18
申请号：US15419615
申请日：2017-01-30
Applicant: International Business Machines Corporation
Inventor： VITTORIO CASTELLI , Radu Florian , Xiaoqiang Luo , Hema Raghavan
IPC: G06F17/30
Abstract: A method for automatically extracting and organizing information by a processing device from a plurality of data sources is provided. A natural language processing information extraction pipeline that includes an automatic detection of entities is applied to the data sources. Information about detected entities is identified by analyzing products of the natural language processing pipeline. Identified information is grouped into equivalence classes containing equivalent information. At least one displayable representation of the equivalence classes is created. An order in which the at least one displayable representation is displayed is computed. A combined representation of the equivalence classes that respects the order in which the displayable representation is displayed is produced.
-
10.发明授权公开(公告)号：US09471559B2
公开(公告)日：2016-10-18
申请号：US13835924
申请日：2013-03-15
Applicant: International Business Machines Corporation
Inventor： Vittorio Castelli , Radu Florian , Xiaoqiang Luo , Sameer Maskey , Hema Raghavan
CPC classification number: G06F17/271 , G06F17/2715 , G06F17/278 , G06F17/2785 , G06F17/30707 , G06F17/30976
Abstract: Creating training data for a natural language processing system may comprise obtaining natural language input, the natural language input annotated with one or more important phrases; and generating training instances comprising a syntactic parse tree of nodes representing elements of the natural language input augmented with the annotated important phrases. In another aspect, a classifier may be trained based on the generated training instances. The classifier may be used to predict one or more potential important phrases in a query.
Abstract translation: 为自然语言处理系统创建训练数据可以包括获得自然语言输入，用一个或多个重要短语注释的自然语言输入; 以及生成包括表示所述自然语言输入的元素的节点的句法解析树的训练实例，所述节点用所述注释的重要短语增强。 在另一方面，可以基于生成的训练实例来训练分类器。 分类器可以用于预测查询中的一个或多个潜在的重要短语。
-
-
-
-
-
-
-
-
-
Search Results
48
records
(took 0.065 seconds)
