专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

AU2022263497A1 SYSTEMS AND METHODS FOR ADAPTIVE PROPER NAME ENTITY RECOGNITION AND UNDERSTANDING 未知
公开(公告)号：AU2022263497A1
公开(公告)日：2022-12-22
申请号：AU2022263497
申请日：2022-11-02
申请人： PROMPTU SYSTEMS CORP
发明人： PRINTZ HARRY WILLIAM
IPC分类号： G10L15/02 , G01C21/36 , G06F16/332 , G06F40/295 , G10L15/18 , G10L15/19 , G10L15/22
摘要： Abstract A computer-implemented method for recognizing and understanding spoken commands that include one or more proper name entities, comprising: receiving an utterance from a user, performing primary automatic speech recognition (ASR) processing upon said utterance with a primary automatic speech recognizer to output a dataset comprising at least a sequence of nominal transcribed words and putative start and end times for each nominal transcribed word within said utterance, performing understanding processing upon said dataset with a natural language understanding (NLU) processor to generate and augment the dataset with a nominal meaning for the utterance and to determine putative presence and type of one or more spoken proper name entities within said utterance, wherein a contiguous section of audio within said utterance corresponding to each putative proper name entity, as determined from said start and end times of the words of the putative proper name entity as transcribed by the primary automatic speech recognizer, comprises an acoustic span, performing secondary automatic speech recognition (ASR) processing upon each said acoustic span with a secondary automatic speech recognizer, in each instance said secondary automatic speech recognizer specialized to process a given putative type of acoustic span to generate a nominal correct transcription and associated meaning for each said acoustic span, substituting the nominal correct transcription and associated meaning obtained from each secondary recognition as appropriate within the dataset to revise the results of the primary automatic speech recognizer and natural language understanding processor and to create a plurality of complete transcriptions and associated meanings, preparing a complete hypothesis ranking grammar comprised of said plurality of complete transcriptions and decoding the utterance against said complete hypothesis ranking grammar to determine an acoustic confidence score for each complete transcription, determining, for each acoustic span of each complete transcription, an NLU confidence score for each transcription of each acoustic span, normalizing said NLU confidence scores across the plurality of complete transcriptions to determine a normalized NLU confidence score of each complete transcription, combining said acoustic confidence score and NLU confidence score of each complete transcription to generate a final confidence score that each complete transcription and associated meaning is correct, which is used to rank the plurality of aforesaid complete transcriptions and associated meanings, and outputting a ranked list of complete transcriptions and associated meanings for the entire utterance.

2. 发明专利

AU2021201071B2 METHOD AND SYSTEM FOR AUTOMATED TEXT ANONYMISATION 未知
公开(公告)号：AU2021201071B2
公开(公告)日：2022-04-28
申请号：AU2021201071
申请日：2021-02-18
申请人： HARRISON AI PTY LTD
发明人： HACHEY BENJAMIN CLAYTON
IPC分类号： G06F21/62 , G06F40/295 , G06N20/00 , G06V30/413 , G16H50/70
摘要： A system for automated text anonymisation of clinical text, the system including an Al pipeline module to configure symbolic Al pipeline components for detecting protected health information (PHI) in the clinical text; a masking module for masking the detected PHI in the clinical text and generating a de-identified clinical text output file as well as a corresponding label file with de-identified information. The pipeline components may include at least one non-symbolic Al pipeline component or machine learning model.

3. 发明专利

AU2021106425A4 Method, system and apparatus for extracting entity words of diseases and their corresponding laboratory indicators from Chinese medical texts 未知
公开(公告)号：AU2021106425A4
公开(公告)日：2021-11-04
申请号：AU2021106425
申请日：2021-08-22
申请人： FENG HONGHAI , FENG RUI MS
发明人： FENG HONGHAI , WEI YAJU , HOU RUIHUI , LI JUN , ZHOU PENGCHENG , LU XUZHAO
IPC分类号： G06F40/295 , G06F40/279 , G06F40/30 , G09B17/02 , G16H70/20
摘要： Method and system and apparatus for extracting entity words of diseases and their corresponding laboratory indicators from Chinese medical texts The present invention discloses a method and system and device for extracting disease and laboratory index entities from Chinese medical texts and extracting relationships between their entities, and relates to the field of information extraction. It includes three major parts, the first consists of the reading subsystem that mainly contains the modules read by the computer. The second consists of the computation subsystem that mainly captures four major entity components, that is, laboratory indicators, abnormal values, relational words, and disease names after decomposing the sentence components. Every entity component is a set that is composed of some words or terms. The first step is inputting three of the four initial word sets to learn the left relational word set. Once the relational word set is updated, it is taken as one of the new inputs, and taking one of the previous three word sets as the one being learned. The iteration is run until every word set cannot be updated. The third consists of an output subsystem that contains two parts: a storage unit and an output unit. The storage one is the set of relevant medical entity words, and the output one displays the relationship between the relevant entities. System for extracting entity words for diseases and their corresponding lab indicators from medical texts Reading Computing output subsystem subsystems Subsystem Module Decont Inform for Informat aminat action or Reading Input ion ion at Storage Output storing Module Unit matching verific extract Unit Unit 1mitial .oul .nt mthi ei ion Untni values g unit ation unit unit Figure 1: System Block Diagram

4. 发明专利

AU2020272235A1 Methods, systems and computer program products for implementing neural network based optimization of database search functionality 未知
公开(公告)号：AU2020272235A1
公开(公告)日：2021-11-04
申请号：AU2020272235
申请日：2020-04-10
申请人： INCYZR PTY LTD
发明人： KERMODE ROGER , REED ARCHIE
IPC分类号： G06F40/295 , G06F16/245 , G06F40/284 , G06N3/02 , G06N20/00
摘要： The invention provides methods, systems and computer programs for optimizing database search functionality. In an embodiment, the invention comprises (i) receiving at least one sequence of words, (ii) identifying within the received sequence of words, one or more strings, based on string attributes, (iii) identifying a class corresponding to each identified string, (iv) generating a tokenized sequence of words by substituting the detected one or more strings with corresponding class descriptor tokens associated with an identified class for such string, (v) generating word embedding vector representations corresponding to an individual word within the generated tokenized sequence of words, (vi) generating a neural network corresponding to the tokenized sequence of words, and (vii) recording an association between the selected sequence of words and the generated neural network or the vector representation of the tokenized sequence of words.

5. 发明专利

DE102019210994A1 Vorrichtung und Verfahren zum Verarbeiten eines digitalen Datenstroms 未知
公开(公告)号：DE102019210994A1
公开(公告)日：2021-01-28
申请号：DE102019210994
申请日：2019-07-24
申请人： BOSCH GMBH ROBERT
发明人： STROETGEN JANNIK , ADEL HEIKE , LANGE LUKAS
IPC分类号： G06F40/295 , G06F16/35 , G06F21/60 , G06N20/00 , G16H10/60
摘要： Computerimplementiertes Verfahren zum Maschinenlernen und Verarbeiten eines digitalen Datenstroms sowie Vorrichtungen dafür, wobei eine Repräsentation eines Textes unabhängig von einer Domäne bereitgestellt wird (302), wobei eine Repräsentation einer Struktur der Domäne bereitgestellt wird (304), und wobei ein Modell zur automatischen Erkennung von sensitiven Textbausteinen abhängig von den Repräsentationen trainiert wird (306), und wobei Daten aus zumindest einem Teil des Datenstroms, die ein Wort repräsentieren, durch Daten ersetzt werden, die einen Platzhalter für das Wort repräsentieren, wobei abhängig von den Daten eine Ausgabe des Modells bestimmt wird, wobei in den Daten zu ersetzende Daten und Daten, welche die zu ersetzenden Daten ersetzen, abhängig von der Ausgabe des Modells bestimmt werden.

6. 发明专利

AU2022204188A1 SYSTEMS AND METHODS FOR PROCESSING CLAIMS 未知
公开(公告)号：AU2022204188A1
公开(公告)日：2022-07-21
申请号：AU2022204188
申请日：2022-06-15
申请人： DATAINFOCOM USA INC
发明人： WANG WENSU , WANG CHUN , WEI ZHIYONG , CHENG WANLI , GAO KUIKUI , SUN YUHAO , NAZIR MUBBASHIR , THIELKE RICK
IPC分类号： G06Q40/08 , G06F16/93 , G06F40/295 , G06F40/40 , G06N3/02 , G06Q10/10 , G06Q30/00 , G06Q40/00 , G06V30/20 , G06V30/242 , G06V30/244 , G06V30/262 , G16H10/60
摘要： Methods, systems, and apparatuses, including computer programs encoded on computer storage media, are provided for processing claims using both unstructured and structured policy documents, claim data, and customer and policy data, in conjunction with automatic requests for human intervention. Policy rules, benefit calculation formulae, necessary data points, and benefit requirements are extracted from policy documents using NLP and Al techniques. Unstructured claim data is converted to a structured form using natural language processing, information extraction, and Al techniques to identify and extract relevant information, including values for the data points and benefit conditions, then the combined structured data and converted unstructured data is processed to get all values for the data points and applicable benefit conditions. The relevant claim information is then further processed against the policy rules for eligibility assessment and benefit calculation formulae to generate a benefit payment amount and entitled additional benefits. RECEIVE CLAIM 304 INFORMATION EXTRACT CLAIM 306 INFORMATION DETERMINE CLAIM TYPE 308 IDENTIFY APPLICABLE ASSESS AND ANALYZE CLAIM INFORMATION 316 AGAINST POLICIES TO DETERMINE BENEFITS CALCULATE BENEFIT 320 PAY CUSTOMER 324

7. 发明专利

IL291724D0 Technologies for dynamically creating representations for regulations 未知
公开(公告)号：IL291724D0
公开(公告)日：2022-05-01
申请号：IL29172422
申请日：2022-03-27
申请人： UL LLC , SHARPE SPENCER , RANA ANNIE IBRAHIM , LIBERMAN VALERIY , ARNOLD MICHAEL , CAULFIELD KYLE MICHAEL , COGLEY JAMES , EPSTEIN LISA , SHEEHAN TRICIA , MEHDIYEV RASHID , ACHARYA SAURAV
发明人： SHARPE SPENCER , RANA ANNIE IBRAHIM , LIBERMAN VALERIY , ARNOLD MICHAEL , CAULFIELD KYLE MICHAEL , COGLEY JAMES , EPSTEIN LISA , SHEEHAN TRICIA , MEHDIYEV RASHID , ACHARYA SAURAV
IPC分类号： G06F40/131 , G06F16/35 , G06F40/197 , G06F40/205 , G06F40/284 , G06F40/295
摘要： Systems and methods for dynamically generating object models corresponding to regulations. According to certain aspects, a server computer may access a regulation and automatically generate a summary of the regulation based on a specific set of sentences. The server computer may additionally determine a set of topics and named-entity attributes for text within a regulation object model, as well as a probability that a topic or attribute is applicable to the regulation. Further, the server computer may generate and enrich object models according to the various analyses and avail the enriched object models for review by entities and users of regulatory compliance services.

8. 发明专利

DE112020003141T5 ABFRAGE-RELAXATION UNTER VERWENDUNG EXTERNEN DOMÄNENWISSENES ZUR BEANTWORTUNG VON ABFRAGEN 未知
公开(公告)号：DE112020003141T5
公开(公告)日：2022-03-24
申请号：DE112020003141
申请日：2020-08-03
申请人： IBM
发明人： LEI CHUAN , OZCAN FATMA , MILLER DORIAN BORIS , KREULEN JEFFREY , GEIS REBECCA
IPC分类号： G06F16/332 , G06F16/901 , G06F40/295 , G16H50/20
摘要： Ein System, ein Verfahren und ein computerlesbares Medium führen ein Verfahren zur Abfrage-Relaxation aus. Eine Abfrage, die von einem Konversationssystem ausgegeben wird, wird empfangen. Mindestens ein Suchbegriff in der Abfrage wird ermittelt. Instanzdaten werden an das Konversationssystem in Reaktion auf die Ermittlung ausgegeben, dass die Instanzdaten in einem Datenspeicher mit dem mindestens einen Suchbegriff in der Abfrage übereinstimmen. Das Computergerät gibt die empfangene Abfrage in Reaktion auf die Ermittlung, dass der mindestens eine Suchbegriff nicht mit den Instanzdaten im Datenspeicher übereinstimmt, an eine externe domänenspezifische Wissensquelle aus. Das Computergerät empfängt die relaxierten Datenübereinstimmungen von der externen domänenspezifischen Wissensquelle, die auf Grundlage einer Mehrzahl von mit der Abfrage verknüpften Kriterien mit mindestens einem Suchbegriff in der Abfrage semantisch verwandt sind. Das Computergerät erzeugt eine Rückmeldung auf die Abfrage basierend auf kontextueller Information und struktureller Information.

9. 发明专利

AU2020278972A1 Systems and methods for event summarization from data 未知
公开(公告)号：AU2020278972A1
公开(公告)日：2021-11-25
申请号：AU2020278972
申请日：2020-04-28
申请人： THOMSON REUTERS ENTPR CENTRE GMBH
发明人： HAGERMAN ELEANOR , HOWALD BLAKE , EKMEKCI BERK
IPC分类号： G06F16/738 , G06F16/953 , G06F40/10 , G06F40/20 , G06F40/295 , G06Q50/10 , G06Q50/26
摘要： In some aspects, a method includes extracting sentences from data corresponding to documents. Each extracted sentence includes at least one matched pair (a keyword from a first or second keyword set and an entity from an entity set). The method includes ordering the plurality of extracted sentences based on a distance between a respective keyword and a respective entity in each extracted sentence. The method includes identifying a first type and a second type of extracted sentences from the ordered plurality of extracted sentences. Sentences having the first type include keywords of the first keyword set. Sentences having the second type include keywords of the second keyword set. The method includes generating an extracted summary including at least one sentence having the first type and at least one sentence having the second type, intermixed based on a predetermined order rule set. The method includes outputting the extracted summary.

10. 发明专利

DE102020200455A1 Verfahren, Computerprogramm und Vorrichtung zum Verarbeiten einer Nutzereingabe 未知
公开(公告)号：DE102020200455A1
公开(公告)日：2021-07-15
申请号：DE102020200455
申请日：2020-01-15
申请人： VOLKSWAGEN AG
发明人： MEENA RAVEESH , PLESCHKA MARK , KOUSIDIS SPYROS , BUSS OKKO
IPC分类号： G06F40/35 , G06F3/16 , G06F16/36 , G06F40/247 , G06F40/295 , G10L15/22
摘要： Die vorliegende Erfindung betrifft ein Verfahren, ein Computerprogramm mit Instruktionen und eine Vorrichtung zum Verarbeiten einer Nutzereingabe. Die Erfindung betrifft weiterhin ein Fortbewegungsmittel, in dem ein erfindungsgemäßes Verfahren oder eine erfindungsgemäße Vorrichtung eingesetzt wird. In einem ersten Schritt wird eine Nutzereingabe empfangen (10). Aus der Nutzereingabe wird dann zumindest eine Entität extrahiert (11). Anschließend werden Daten für die zumindest eine Entität aus einer Wissensbasis abgefragt (12). Die aus der Wissensbasis abgefragten Daten werden ausgewertet (13). Unter Verwendung der Ergebnisse der Auswertung (13) wird schließlich eine Antwort auf die Nutzereingabe generiert (14).

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