专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

91. 发明专利

DE102018222156A1 Verfahren, Anordnung und Verwendung zum Erzeugen einer Antwortausgabe in Reaktion auf eine Spracheingabeinformation 未知
公开(公告)号：DE102018222156A1
公开(公告)日：2020-06-18
申请号：DE102018222156
申请日：2018-12-18
申请人： VOLKSWAGEN AG
发明人： GALETZKA FABIAN , ROSE JEWGENI , KWOCZEK SIMON
IPC分类号： G06F16/332 , G06F3/16 , G06F40/40 , G06N3/02 , G10L15/22
摘要： Die Erfindung betrifft ein Verfahren zum Erzeugen einer Antwortausgabe (A) in Reaktion auf eine Spracheingabeinformation (E), mit:a) Ermitteln, ob eine Spracheingabeinformation (E) wenigstens eine vordefinierte Entität umfasst; und wenn dies der Fall ist:b) Erzeugen eines Sprachinformationssatzes auf Basis der Spracheingabeinformation (E), in dem die Entität nicht enthalten ist;c) Erzeugen eines Informationssatzes, wobei der Informationssatz der Entität zugeordnete Informationen aus einer Wissensdatenbank (14) umfasst;d) mit einer ersten Analyseeinheit (22): Ermitteln wenigstens einer Zielinformation aus dem Informationssatz, die hinsichtlich der Entität relevant ist;e) mit einer zweiten Analyseeinheit (24): Erzeugen einer Antwortteilausgabe auf Basis der ermittelten Zielinformation; undf) Erzeugen einer Antwortausgabe (A) auf Basis der Antwortteilausgabe.Ferner betrifft die Erfindung ein System (10) und die Verwendung eines solchen Systems (10).

92. 发明专利

CA3089037A1 DETECTING DUPLICATED QUESTIONS USING REVERSE GRADIENT ADVERSARIAL DOMAIN ADAPTATION 未知
公开(公告)号：CA3089037A1
公开(公告)日：2020-06-04
申请号：CA3089037
申请日：2019-07-26
申请人： INTUIT INC
发明人： CARVALHO VITOR R , KAMATH ANUSHA
IPC分类号： G06F16/242 , G06F16/2452 , G06F16/2453 , G06F16/33 , G06F16/332 , G06F16/35
摘要： Detect duplicated questions using reverse gradient adversarial domain adaptation includes applying a general network to multiple general question pairs to obtain a first set of losses, A target domain network is applied to multiple domain specific network pairs to obtain a second set of losses. Further, a domain distinguishing network is applied to a set of domain specific questions and a set of general questions to obtain a third set of losses. A set of accumulated gradients is calculated from the first set of losses, the second set of losses, and the third set of losses. Multiple features are updated according to the set of accumulated gradients to train the target domain network.

93. 发明专利

GB2574382A Method, apparatus and computer program for information retrieval using query expansion 未知
公开(公告)号：GB2574382A
公开(公告)日：2019-12-11
申请号：GB201808875
申请日：2018-05-31
申请人： UXLABS LTD
发明人： TONY RUSSELL-ROSE , PHIL GOOCH
IPC分类号： G06F16/242 , G06F16/2453 , G06F16/332 , G06F16/832 , G06F16/9032 , G06F16/9532
摘要： Methods and apparatus for information retrieval are disclosed. The method starts by determining candidate query expansion terms S102 for expanding a received query expression having a search term, by searching information sources basing on a number of the words included in the search term, to identify one or more terms potentially related to the query expression. The query expression is then modified S105 to include the candidate query expansion terms, and the modified query expression is then submitted to a search engine S106. With the received search results S107 from the search engine being presented to the user. The candidate terms may be output S103 for the user to select the terms to be S104 included in the search query. The candidate query expansion terms may be determined by searching a statistical model if the search term only comprises a single word, or by searching a curated database if the search term comprises a plurality of words. The related terms may hypernyms, hyponyms, synonyms or term defintions.

94. 发明专利

BR102019007765A2 técnicas de aprendizagem profunda baseadas em agentes conversacionais multi-uso para o processamento de consultas em linguagem natural 未知
公开(公告)号：BR102019007765A2
公开(公告)日：2019-11-05
申请号：BR102019007765
申请日：2019-04-16
申请人： TATA CONSULTANCY SERVICES LTD
发明人： ASHISH CHAUDHARY , GAUTAM SHROFF , MAHESH PRASAD SINGH , MAYUR PATIDAR , PRATEEK SACHAN , PRERNA KHURANA , PUNEET AGARWAL , RACHIT BANSAL , ROHIT KUMAR , VIVEK BISHT
IPC分类号： G06F16/332
摘要： patente de invenção: técnicas de aprendizagem profunda baseadas em agentes conversacionais multi-uso para o processamento de consultas em linguagem natural" sistemas e métodos para técnicas de aprendizagem profunda baseadas em agentes conversacionais multiuso para o processamento de consultas em linguagem natural. os sistemas e métodos tradicionais forne-cem sistemas conversacionais para o processamento de consultas em lin-guagem natural, mas não empregam técnicas de aprendizagem profunda e, portanto, não conseguem processar um grande número de intenções. for-mas de concretização da presente divulgação proporcionam para agentes conversacionais multiuso multiuso para técnicas em linguagem natural baseadas na aprendizagem profunda para processar as consultas em lin-guagem natural pela definição e integração lógica de uma pluralidade de componentes compreendendo agentes conversacionais multiuso, pela identificação de um agente apropriado para processar uma ou mais consul-tas em linguagem natural por uma técnica de identificação de intenção de alto nível, pela previsão de uma intenção de usuário provável, classifica-ção da consulta e geração de um conjunto de respostas consultando ou atualizando um ou mais gráficos de conhecimento (322).

95. 发明专利

RU2691851C1 СИСТЕМА КОМПОЗИЦИИ ЗАПРОСОВ 未知
公开(公告)号：RU2691851C1
公开(公告)日：2019-06-18
申请号：RU2017137750
申请日：2016-09-27
申请人： GOOGLE INC
发明人： USZKOREIT JAKOB D
IPC分类号： G06F16/242 , G06F16/332 , G06F16/951 , G06F16/9537 , G06F17/18
摘要： Изобретениеотноситсяк средствамвыборкии представлениязапросов. Техническийрезультатзаключаетсяв уменьшениивременивыполнениязапроса. Осуществляютдоступпосредствомустройстваобработкиданныхк даннымзапросов, указывающимвводызапросов, принимаемыеотпользовательскихустройствмножествапользователей, причемданныезапросовтакжеуказываютконтекстввода, которыйописывает, длякаждоговводазапроса, контекствводадлявводазапроса, которыйотличаетсяотконтента, описываемогопосредствомвводазапроса. Группируютпосредствомустройстваобработкиданныхвводызапросовв контекстныегруппынаоснове, отчасти, контекставводадлякаждогоизвводовзапросови контента, описываемогокаждымвводомзапроса. Определяютпосредствомустройстваобработкиданных, длякаждойизконтекстныхгрупп, вероятностьконтекстнойгруппынаосновесоответствующихвероятностейпоступлениявводовзапросов, которыепринадлежатконтекстнойгруппе, причемвероятностьконтекстнойгруппыуказываетвероятностьтого, чтопоменьшеймереодинвводзапроса, которыйпринадлежитконтекстнойгруппеи предоставляетсядляконтекставводаконтекстнойгруппы, будетвыбранпользователем. 4 н. и 12 з.п. ф-лы, 6 ил.

96. 发明专利

CA2120447A1 Automatically Relaxable Query for Information Retrieval 未知
公开(公告)号：CA2120447A1
公开(公告)日：1995-10-01
申请号：CA2120447
申请日：1994-03-31
申请人： LIZEE ROBERT , LIZEE MARTIN
发明人： LIZEE ROBERT , LIZEE MARTIN
IPC分类号： G06F16/9032 , G06F16/242 , G06F16/332 , G06F17/00

97. 发明专利

GB2584042B Personal data management 未知
公开(公告)号：GB2584042B
公开(公告)日：2022-09-21
申请号：GB202011783
申请日：2018-12-06
申请人： TRISENT LTD
发明人： GORDON JOHNSTON ROBERTSON POVEY
IPC分类号： G06F16/242 , G06F16/332 , G06F16/9535
摘要： A method of searching comprising performing a first search; creating a search query for a second search based at least on the results of the first search, wherein one of the first search and second search is a search performed on a store of collected personal data and the other of the first search and second search is a search performed on at least one store of publicly available data.

98. 发明专利

AU2021219697A1 Reference-based document ranking system 未知
公开(公告)号：AU2021219697A1
公开(公告)日：2022-09-01
申请号：AU2021219697
申请日：2021-02-10
申请人： COPYRIGHT CLEARANCE CT INC
发明人： MARMANIS HARALAMBOS
IPC分类号： G06F16/38 , G06F16/31 , G06F16/33 , G06F16/332 , G06F16/35 , G06F16/383
摘要： A system for ranking electronic documents based on reference frequency includes a central controller in electronic communication with a document database. The central controller maintains a graphical model of the electronic documents that identifies all references between documents. A weight is automatically calculated and assigned to each reference within the graphical model in order to increase the significance of document references which are based on subject matter relevance and decrease the significance of document references which are based on interpersonal relationships or other meritless factors. Using the weighted graphical model, the central controller is able to automatically identify document clusters having similar subject matter, create a probability matrix for each cluster based on the weighted graphical model, and apply a power iteration to each probability matrix to yield a reference-based ranking of the electronic documents within each cluster.

99. 发明专利

AU2019263758B2 Systems and methods for generating a contextually and conversationally correct response to a query 未知
公开(公告)号：AU2019263758B2
公开(公告)日：2022-08-11
申请号：AU2019263758
申请日：2019-05-03
申请人： THOMSON REUTERS ENTPR CENTRE GMBH
发明人： SURPRENANT MATTHEW A , LINDBERG ERIK , MCELVAIN GAYLE
IPC分类号： G06F16/14 , G06F16/20 , G06F16/332 , G06F16/338 , G06F40/166 , G06F40/20 , G06Q50/18 , G06Q50/20
摘要： The present disclosure relates to systems and methods for generating contextually, grammatically, and conversationally correct answers to input questions. Embodiments provide for linguistic and syntactic structure analysis of a submitted question in order to determine whether the submitted question may be answered by at least one headnote. The question is then further analyzed to determine more details about the intent and context of the question. A federated search process, based on the linguistic and syntactic structure analysis, and the additional analysis of the question is used to identify candidate question-answer pairs from a corpus of previously created headnotes. Machine learning models are used to analyze the candidate question-answer pairs, additional rules are applied to rank the candidate answers, and dynamic thresholds are applied to identify the best potential answers to provide to a user as a response to the submitted question.

100. 发明专利

AU2020321751B2 Neural network system for text classification 未知
公开(公告)号：AU2020321751B2
公开(公告)日：2022-07-28
申请号：AU2020321751
申请日：2020-06-17
申请人： INTUIT INC
发明人： PODGORNY IGOR A , CARVALHO VITOR R , GUPTA SPARSH
IPC分类号： G06F16/332 , G06F16/33 , G06F40/216 , G06F40/284 , G06F40/30
摘要： A computer-implemented method is provided to perform text classification with a neural network system. The method includes providing a computing device to receive input datasets including user input question text and feed the datasets to the neural network system. The neural network system includes one or more neural networks configured to extract and concatenate character-based features, word-based features from the question datasets and clickstream embeddings of clickstream data to form a representation vector indicative of the question text and user behavior. A representation vector is fed into fully connected layers of a feed-forward network. The feed-forward network is configured to predict a first class and a second class associated with respective user input questions based on the representation vector.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式