Natural Language Patents (Class 704/9)
  • Patent number: 10147050
    Abstract: Methods to provide selective supporting evidence processing by applying a first machine learning (ML) model to a first candidate answer to generate a first confidence score that does not consider supporting evidence for the first candidate answer, determining, from a second ML model, an expected contribution of processing supporting evidence for the first candidate answer, and upon determining that the expected contribution does not exceed a specified threshold, skipping supporting evidence processing for the first candidate answer.
    Type: Grant
    Filed: September 22, 2015
    Date of Patent: December 4, 2018
    Assignee: International Business Machines Corporation
    Inventors: Michael W. Britt, Adam C. Lange-Pearson, Thomas P. Pitzen, Richard J. Stevens
  • Patent number: 10146751
    Abstract: System and methods for creating structured or semi-structured representations of information extracted from unstructured text data sources are described. In some embodiments, without requiring a predefined target data structure, the methods identify the grammatical and semantic attributes and context information in a text content, and create object-properties association data as knowledge and information extracted from the unstructured data, and represent such information in a structured or semi-structured format to facilitate search and trend analysis. In some other embodiments, the methods identify the types of information contained in the unstructured data, and for a pre-defined target information type, the methods identify the context and content of the portion of the text that represents the target information type, and extract the text, attach a tag or label to the extracted text, and store or display the data in a database table format or xml format for further pattern and trend analysis.
    Type: Grant
    Filed: December 7, 2015
    Date of Patent: December 4, 2018
    Inventor: Guangsheng Zhang
  • Patent number: 10146395
    Abstract: Techniques described herein provide quality of experience (QoE) diagnosis and analysis for wireless communications. Implementations include generating, by a user device, a user interaction event. The user device may obtain data associated with one or more QoE metrics, and cause analysis on the data associated with the one or more metrics to identify QoE issues.
    Type: Grant
    Filed: March 26, 2015
    Date of Patent: December 4, 2018
    Assignee: T-Mobile USA, Inc.
    Inventors: Jie Hui, Qi Chen, Haokun Luo, Kevin Lau, Karthik Iyer, Kranthi Sontineni
  • Patent number: 10140978
    Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting alternates in speech recognition. In some implementations, data is received that indicates multiple speech recognition hypotheses for an utterance. Based on the multiple speech recognition hypotheses, multiple alternates for a particular portion of a transcription of the utterance are identified. For each of the identified alternates, one or more features scores are determined, the features scores are input to a trained classifier, and an output is received from the classifier. A subset of the identified alternates is selected, based on the classifier outputs, to provide for display. Data indicating the selected subset of the alternates is provided for display.
    Type: Grant
    Filed: September 13, 2017
    Date of Patent: November 27, 2018
    Assignee: Google LLC
    Inventors: Alexander H. Gruenstein, Dave Harwath, Ian C. McGraw
  • Patent number: 10140292
    Abstract: The embodiments herein achieve a picture based communication system. The system allows users option to select one or more pictures, and any associated attributes. The selection of one or more pictures, and any associated attributes is taken as input. The selected words and attributes are converted to a graph representation, and subsequently the graph representation is converted to a sentence in target language. The method further involves predicting new relations, words, and attributes for further selection by user.
    Type: Grant
    Filed: January 26, 2017
    Date of Patent: November 27, 2018
    Assignee: AVAZ, INC.
    Inventor: Ajit Narayanan
  • Patent number: 10140972
    Abstract: A method of training an acoustic model for a text-to-speech system, the method comprising: receiving speech data; said speech data comprising data corresponding to different values of a first speech factor, and wherein said speech data is unlabeled, such that for a given item of speech data, the value of said first speech factor is unknown; clustering said speech data according to the value of said first speech factor into a first set of clusters; and estimating a first set of parameters to enable the acoustic model to accommodate speech for the different values of the first speech factor, wherein said clustering and said first parameter estimation are jointly performed according to a common maximum likelihood criterion.
    Type: Grant
    Filed: August 22, 2014
    Date of Patent: November 27, 2018
    Assignee: Kabushiki Kaisha Toshiba
    Inventor: Langzhou Chen
  • Patent number: 10140101
    Abstract: A method, system and computer-usable medium for linking a set of executable code snippets to perform a complicated task, comprising: decomposing a natural language statement into a plurality of decomposed natural language components; searching a repository of code snippets to identify code snippets corresponding to each of the decomposed natural language components; ordering execution of the code snippets based upon the plurality of decomposed natural language components; and, executing the code snippets in order of the natural language statement requests until a final outcome is achieved.
    Type: Grant
    Filed: August 26, 2015
    Date of Patent: November 27, 2018
    Assignee: International Business Machines Corporation
    Inventors: Corville O. Allen, Heather L. Duschl, Marit L. Imsdahl, Alexandra D. Markello, Dana L. Price
  • Patent number: 10140260
    Abstract: Disclosed is an improved system for performing intelligent text reduction on text strings for optimal display in a variety of user interface display elements. The system includes a smart text reduction algorithm that can shorten text strings as necessary in response to resizing the user interface or elements therein. The system also includes a design time user interface for application developers (or other users) to design graphical user interfaces in a collaborative workspace. The design time user interface incorporates the smart text reduction algorithm to enable designers to modify user interface display elements and run the text reduction algorithm on any text strings that no longer fit within the modified display elements. The described embodiments can also translate text strings of user interface display elements and can execute the smart text shortening algorithm on the translated text strings accordingly.
    Type: Grant
    Filed: July 15, 2016
    Date of Patent: November 27, 2018
    Assignee: SAP SE
    Inventors: Michail Vasiltschenko, Ralph Debusmann, Florian Hoffmann, Iuliia Konstantinova, Christian Lieske, Charles Monte
  • Patent number: 10140273
    Abstract: First content containing a plurality of list items in one or more lists can be parsed for conjunctions and implied list indicators. One or more modifications can occur at one or more conjunctions or implied list indicators. The one or more modifications can comprise one or more of expanding text, contracting text, and replacing text. The one or more modifications can generate second content conducive to natural language processing operations.
    Type: Grant
    Filed: January 19, 2016
    Date of Patent: November 27, 2018
    Assignee: International Business Machines Corporation
    Inventors: Keith P. Biegert, Brendan C. Bull, David Contreras, Robert C. Sizemore, Sterling R. Smith
  • Patent number: 10140286
    Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for optimized graph traversal are disclosed. In one aspect, a method includes the actions of receiving a given phrase that is input through a user interface by a digital component provider. The actions further include determining an entity that is being referred to by the given phrase. The actions further include identifying properties of the entity. The actions further include selecting a subset of the properties that were identified for the entity. The actions further include identifying additional phrases. The actions further include updating the user interface to present at least some of the additional phrases with programmatic controls that assign one or more of the additional phrase as distribution criteria for digital components of the digital component provider in response to activation of the programmatic controls.
    Type: Grant
    Filed: February 22, 2017
    Date of Patent: November 27, 2018
    Assignee: Google LLC
    Inventors: Christopher Jacob Durr, Hector Mauricio Ayala, Mayank Jain
  • Patent number: 10133335
    Abstract: In general, embodiments of the invention provide an approach to proactively adjust timeout settings on a display device based on user activity. Specifically, a system and method are presented to adaptively adjust the inactivity timeout settings on a display device based on the application being used, the content being displayed on the device, and an analysis of the user's history with the application. The present invention calculates an inactivity timeout modifier which is utilized to modify the default operating system timeout value.
    Type: Grant
    Filed: May 26, 2015
    Date of Patent: November 20, 2018
    Assignee: International Business Machines Corporation
    Inventors: Jeffrey D. Amsterdam, Rick A. Hamilton, II, Mauro Marzorati, Brian M. O'Connell, Keith R. Walker
  • Patent number: 10133808
    Abstract: A method, system and computer program product for generating answers to questions. In one embodiment, the method comprises receiving an input query, decomposing the input query into a plurality of different subqueries, and conducting a search in one or more data sources to identify at least one candidate answer to each of the subqueries. A ranking function is applied to each of the candidate answers to determine a ranking for each of these candidate answers; and for each of the subqueries, one of the candidate answers to the subquery is selected based on this ranking. A logical synthesis component is applied to synthesize a candidate answer for the input query from the selected the candidate answers to the subqueries. In one embodiment, the procedure applied by the logical synthesis component to synthesize the candidate answer for the input query is determined from the input query.
    Type: Grant
    Filed: May 2, 2016
    Date of Patent: November 20, 2018
    Assignee: International Business Machines Corporation
    Inventors: Eric W. Brown, Jennifer Chu-Carroll, David A. Ferrucci, Adam P. Lally, James W. Murdock, John M. Prager
  • Patent number: 10133879
    Abstract: A first character string in a logic used for the text analysis with a second character string which is different form the first character string is replaced using a mapping between the first character string and the second character string. The first character string being a character or a combination of characters in a logic. A first character string is searched for in a target document for the text analysis and the first character string, if found, is replaced with the second character string, using the mapping. The logic is sent after the replacement to a server computer. The target document after the replacement is sent to the server computer. A text analysis result for the target document after the replacement is received form the server computer.
    Type: Grant
    Filed: November 3, 2015
    Date of Patent: November 20, 2018
    Assignee: International Business Machines Corporation
    Inventors: Daisuke Hayashi, Sayaka Tamai, Keisuke Nitta, Fumihiko Terui
  • Patent number: 10133920
    Abstract: One embodiment provides a method, including: receiving, at an input and display device, handwriting input; receiving, using a processor, voice input; generating, using a processor, at least one first word based on the handwriting input; generating, using a processor, at least one second word based on the voice input; and determining, using a processor, a highest probability word based on the at least one first word and the at least one second word. Other aspects are described and claimed.
    Type: Grant
    Filed: February 27, 2015
    Date of Patent: November 20, 2018
    Assignee: Lenovo (Singapore) Pte. Ltd.
    Inventors: Antoine Roland Raux, Grigori Zaitsev, Russell Speight VanBlon, Jianbang Zhang
  • Patent number: 10133728
    Abstract: The system that performs semantic parsing may automatically extract complex information from databases. Complex information may comprise nested event structures. In one example process, a processor may receive unannotated text and may access a natural-language database that includes nested events. The processor, in performing semantic parsing, may automatically generate syntactic trees that include annotations that represent the semantic information. In particular, the natural-language sentences and the database include nested event structures.
    Type: Grant
    Filed: March 20, 2015
    Date of Patent: November 20, 2018
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Hoifung Poon, Kristina Toutanova, Ankur P. Parikh
  • Patent number: 10134391
    Abstract: System and method to adjust an automatic speech recognition (ASR) engine, the method including: receiving social network information from a social network; data mining the social network information to extract one or more characteristics; inferring a trend from the extracted one or more characteristics; and adjusting the ASR engine based upon the inferred trend. Embodiments of the method may further include: receiving a speech signal from a user; and recognizing the speech signal by use of the adjusted ASR engine. Further embodiments of the method may include: producing a list of candidate matching words; and ranking the list of candidate matching words by use of the inferred trend.
    Type: Grant
    Filed: March 13, 2017
    Date of Patent: November 20, 2018
    Assignee: Avaya Inc.
    Inventors: George W. Erhart, Valentine C. Matula, David J. Skiba
  • Patent number: 10134297
    Abstract: Computer-implemented systems and methods are provided for determining a document's complexity. For example, a computer performing the complexity analysis can receive a document. The computer can determine the content words within the document and determine an association measure for each group of content words. An association profile can be created for the document using the association measures. The computer can use the association profile to determine the complexity of the document. The complexity of the document may correspond to the document's suitable reading level or, if the document is an essay, an essay score.
    Type: Grant
    Filed: February 14, 2014
    Date of Patent: November 20, 2018
    Assignee: Educational Testing Service
    Inventors: Beata Beigman Klebanov, Michael Flor
  • Patent number: 10134388
    Abstract: An automatic speech recognition (ASR) system may add new words to an ASR system by identifying words with similar usage and replicating the variations of the identified words to create new words. A new word that is used similarly to a known word may be varied to create new word forms that are similar to the word forms of a known word. The new word forms may then be incorporated into an ASR model to allow the ASR system to recognize those words when they are detected in speech. Such a system may allow flexible incorporation and recognition of varied forms of new words entering a general lexicon.
    Type: Grant
    Filed: December 23, 2015
    Date of Patent: November 20, 2018
    Assignee: Amazon Technologies, Inc.
    Inventor: Jeffrey Paul Lilly
  • Patent number: 10133739
    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for neural translation systems with rare word processing. One of the methods is a method training a neural network translation system to track the source in source sentences of unknown words in target sentences, in a source language and a target language, respectively and includes deriving alignment data from a parallel corpus, the alignment data identifying, in each pair of source and target language sentences in the parallel corpus, aligned source and target words; annotating the sentences in the parallel corpus according to the alignment data and a rare word model to generate a training dataset of paired source and target language sentences; and training a neural network translation model on the training dataset.
    Type: Grant
    Filed: October 23, 2015
    Date of Patent: November 20, 2018
    Assignee: Google LLC
    Inventors: Quoc V. Le, Minh-Thang Luong, Ilya Sutskever, Oriol Vinyals, Wojciech Zaremba
  • Patent number: 10135979
    Abstract: A method, system, and a computer program product are provided for monitoring an emotional state of a conversation by monitoring a communication between first and second persons to determine an emotional state of the communication, providing a visual representation of the emotional state of the communication for display to a person having access to the communication, and performing an intervention action to improve the emotional state of the communication in response to detecting the emotional state of the communication meets one or more specified intervention criteria.
    Type: Grant
    Filed: November 2, 2016
    Date of Patent: November 20, 2018
    Assignee: International Business Machines Corporation
    Inventor: Thomas G. Zimmerman
  • Patent number: 10133729
    Abstract: Systems, methods, and computer-readable media for providing semantically-relevant discovery of solutions are described herein. In some examples, a computing device can receive an input, such as a query. The computing device can process each word of the input sequentially to determine a semantic representation of the input. Techniques and technologies described herein determine a response to the input, such as an answer, based on the semantic representation of the input matching a semantic representation of the response. An output including one or more relevant responses to the request can then be provided to the requestor. Example techniques described herein can apply machine learning to train a model with click-through data to provide semantically-relevant discovery of solutions. Example techniques described herein can apply recurrent neural networks (RNN) and/or long short term memory (LSTM) cells in the machine learning model.
    Type: Grant
    Filed: August 28, 2015
    Date of Patent: November 20, 2018
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Xiaodong He, Jianfeng Gao, Hamid Palangi, Xinying Song, Yelong Shen, Li Deng, Jianshu Chen
  • Patent number: 10127271
    Abstract: A system for generating a query includes a term unit (1) for extracting a term from at least one input document (51), to obtain an extracted term. A category unit (2) is arranged for associating the extracted term with a category that is semantically related with the extracted term. A query unit (3) is arranged for generating a query in dependence on the extracted term and the category. The query unit (3) includes an additional term unit (4) for generating at least one additional search term based on the category, and the query unit (3) is arranged for including the additional search term in the query. A submit unit (5) is arranged for submitting the query to at least one search engine (50), to obtain a plurality of found documents.
    Type: Grant
    Filed: September 6, 2013
    Date of Patent: November 13, 2018
    Assignee: Koninklijke Philips N.V.
    Inventors: Mauro Barbieri, Johannes Henricus Maria Korst, Pavankumar Murli Dadlani Mahtani, Zarko Aleksovski, Marc Andre Peters
  • Patent number: 10127289
    Abstract: Computer implemented systems and methods are disclosed for automatically clustering and canonically identifying related data in various data structures. Data structures may include a plurality of records, wherein each record is associated with a respective entity. In accordance with some embodiments, the systems and methods further comprise identifying clusters of records associated with a respective entity by grouping the records into pairs, analyzing the respective pairs to determine a probability that both members of the pair relate to a common entity, and identifying a cluster of overlapping pairs to generate a collection of records relating to a common entity. Clusters may further be analyzed to determine canonical names or other properties for the respective entities by analyzing record fields and identifying similarities.
    Type: Grant
    Filed: August 10, 2016
    Date of Patent: November 13, 2018
    Assignee: Palantir Technologies Inc.
    Inventors: Lawrence Manning, Rahul Mehta, Daniel Erenrich, Guillem Palou Visa, Roger Hu, Xavier Falco, Rowan Gilmore, Eli Bingham, Jason Prestinario, Yifei Huang, Daniel Fernandez, Jeremy Elser, Clayton Sader, Rahul Agarwal, Matthew Elkherj, Nicholas Latourette, Aleksandr Zamoshchin
  • Patent number: 10127442
    Abstract: Embodiments of the present disclosure relate to non-sequential document comparison. A first plurality of segments in a first document and a second plurality of segments in a second document are obtained. In response to a first segment from the first plurality of segments being associated with a second segment from the second plurality of segments, a third segment from the first plurality of segments is associated with a fourth segment from the second plurality of segments.
    Type: Grant
    Filed: June 9, 2016
    Date of Patent: November 13, 2018
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Ke Ke Cai, Hong Lei Guo, Zhi Li Guo, Feng Jin, Yong Qin, Zhong Su
  • Patent number: 10129400
    Abstract: A language processor includes a parser and an invoker. The parser parses a spoken statement to detect a plurality of words in the spoken statement and generates a parse tree based on the detected plurality of words. The invoker determines, based on the parse tree, a plurality of potential services to invoke to respond to the spoken statement, calculates a confidence score for each potential service of the plurality of potential services, wherein each confidence score indicates a likelihood, that a potential service is a service requested by the spoken statement, and determines, based on the calculated confidence scores, to invoke a potential service of the plurality of potential services in response to the spoken statement. This helps reduce a number of questions that a caller associated with the spoken statement has to answer before the language processor determines a service to invoke to respond to the caller.
    Type: Grant
    Filed: December 2, 2016
    Date of Patent: November 13, 2018
    Assignee: Bank of America Corporation
    Inventors: Viju S. Kothuvatiparambil, Priyank R. Shah, Emad Noorizadeh
  • Patent number: 10127303
    Abstract: A processor-implemented method, system, and/or computer program product measures and displays facets in context-based conformed dimensional data gravity wells. Conformed dimensional objects and synthetic context-based objects are parsed into n-tuples. A virtual mass of each parsed object is calculated, in order to define a shape of multiple context-based conformed dimensional data gravity wells that are created when data objects that are pulled into each of the context-based conformed dimensional data gravity well frameworks on a context-based conformed dimensional gravity wells membrane. Data from the multiple context-based conformed dimensional data gravity wells then populates nodes in a data model. A displayed appearance of the particular context-based conformed dimensional data gravity well is dynamically adjusted according to how many parsed synthetic context-based objects are pulled into the particular context-based conformed dimensional data gravity well.
    Type: Grant
    Filed: June 23, 2016
    Date of Patent: November 13, 2018
    Assignee: International Business Machines Corporation
    Inventors: Samuel S. Adams, Robert R. Friedlander, James R. Kraemer, Jeb R. Linton
  • Patent number: 10120843
    Abstract: One or more processors identify one or more character errors in a document. The one or more processors replace a character having the identified one or more character errors with a replacement character. The replacement of the character error with the replacement character allows deep parsing of the document to complete. The one or more processors apply to the document one or both of a deep parsing and natural language processing after the replacing.
    Type: Grant
    Filed: August 26, 2014
    Date of Patent: November 6, 2018
    Assignee: International Business Machines Corporation
    Inventors: James E. Bostick, John M. Ganci, Jr., Kimberly G. Starks, Craig M. Trim
  • Patent number: 10120858
    Abstract: Methods for analyzing a query is provided. Methods may include receiving a query from a user. Methods may include dissecting the query into a plurality of word segments. Methods may include determining, based on a predetermined ontology, which of the plurality of words satisfies a predetermined ontological threshold and which words do not satisfy the predetermined ontological threshold. Methods may include discarding the words that do not satisfy the predetermined ontological threshold. Methods may include determining a part of speech associated with each word based on the content of the query. Methods may include determining a concept associated with each word based on the content of the query. Methods may include displaying to the user word that satisfies the predetermined threshold alongside the associated part of speech and associated concept. Methods may include enabling the user to change each word, each concept and each part of speech.
    Type: Grant
    Filed: November 20, 2015
    Date of Patent: November 6, 2018
    Assignee: Bank of America Corporation
    Inventors: Ramakrishna R. Yannam, Viju Kothuvatiparambil, David H. Witting, Bharathiraja Krishnamoorthy, Mahalakshmi Chandra Sekaran, Sushil Golani
  • Patent number: 10121467
    Abstract: A language model for automatic speech processing, such as a finite state transducer (FST) may be configured to incorporate information about how a particular word sequence (N-gram) may be used in a similar manner from another N-gram. A score of a component of the FST (such as an arc or state) relating to the first N-gram may be based on information of the second N-gram. Further, the FST may be configured to have an arc between a state of the first N-gram and a state of the second N-gram to allow for cross N-gram back off, rather than backoff from a larger N-gram to a smaller N-gram during traversal of the FST during speech processing.
    Type: Grant
    Filed: June 30, 2016
    Date of Patent: November 6, 2018
    Assignee: Amazon Technologies, Inc.
    Inventors: Ankur Gandhe, Denis Sergeyevich Filimonov, Ariya Rastrow, Björn Hoffmeister
  • Patent number: 10120863
    Abstract: A regular rule learning system, including an analyzing circuit configured to analyze a corpus of sentences to find semantic relationships between sentence constituents that are responsible for specific senses of words in that sentence by describing the semantic relationships and grammatical relations that are actuated in the sentence.
    Type: Grant
    Filed: October 24, 2017
    Date of Patent: November 6, 2018
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Octavian Popescu, Vadim Sheinin
  • Patent number: 10120864
    Abstract: A method for categorizing an issue includes, for each of a plurality of categories of issue, providing at least one discourse pattern for identify text sequences that meet the discourse pattern. At least one of the discourse patterns specifies that an instance of a domain term in a domain term vocabulary be present in the text sequence for the pattern to be met. An issue is received which includes a text sequence. The text sequence is categorized based on which, if any, of the discourse patterns are met by the text sequence of the received issue. Information based on the categorization of the text sequence is output.
    Type: Grant
    Filed: March 29, 2016
    Date of Patent: November 6, 2018
    Assignee: Conduent Business Services LLC
    Inventors: Ágnes Sandor, Nikolaos Lagos, Caroline Brun, Ngoc Phuoc An Vo
  • Patent number: 10120865
    Abstract: Techniques for use in connection with a system for automatically generating text. Techniques include accessing information specifying at least one referential expression for at least a first referent and at least one anaphoric expression for at least the first referent; accessing a template that includes human-language text and a first tag that serves as a placeholder for a first text portion including a reference to at least the first referent; automatically identifying, using at least one system rule and at least one processor, text to use for the first text portion at least in part by determining whether to use as the text for the first text portion the at least one referential expression or the at least one anaphoric expression; and automatically generating output text including the human-language text and the identified text for the first text portion.
    Type: Grant
    Filed: July 5, 2016
    Date of Patent: November 6, 2018
    Assignee: YSEOP SA
    Inventors: Alain Kaeser, Emmanuel Vignon, Ludan Stoeckle
  • Patent number: 10116600
    Abstract: Many computing scenarios enable a user to send a message to a user expressing a statement, such as a “happy birthday” message or a “let's meet for lunch” message, and may seek assistance in composing the message. Some services may suggest stock messages, which the user may browse and select, but the generic, impersonal style of the messages may be apparent to the recipient. Presented herein are techniques for composing a message from a user to a recipient expressing a selected statement in the personalized style of the user. Such techniques involve evaluating an expression set (e.g., a mailbox or text message set) of the user to identify an expression style of the user when expressing such statements to such recipients, and composing a message that expressing the statement in the expression style of the user.
    Type: Grant
    Filed: August 14, 2014
    Date of Patent: October 30, 2018
    Assignee: EXCALIBUR IP, LLC
    Inventor: Stewart Smith
  • Patent number: 10114883
    Abstract: An electronic device is disclosed. The electronic device stores metadata associated with a content item, the metadata including first metadata from a first source and second metadata from a second source. Both the first and second metadata are associated with a particular attribute of the content item. Based on the first and second metadata, preferred metadata is generated for a particular attribute of the content item.
    Type: Grant
    Filed: March 3, 2015
    Date of Patent: October 30, 2018
    Assignee: Apple Inc.
    Inventors: Guy Piekarz, Ilan Ben-Zeev, Tomer Y. Doron
  • Patent number: 10115170
    Abstract: In one embodiment, a method for searching for drawings found in a set of patent-related documents includes receiving a search query, and searching a parts index for part names matching one or more terms in the search query. The method further includes displaying one or more drawings, wherein the one or more drawings include at least one part name within the search query and the part name is stored within the parts index. In another embodiment, a method for searching for drawings includes receiving a request to search for drawings relating to a select drawing of a patent-related document, extracting parts information from the select drawing, and comparing the extracted parts information with parts information stored in a parts index. The method further includes determining that one or more drawings are similar to the select drawing, and displaying the one or more drawings.
    Type: Grant
    Filed: July 3, 2014
    Date of Patent: October 30, 2018
    Assignee: Lex Machina, Inc.
    Inventors: Brian K. Elias, Matthew C. Morrise, Ray Daley
  • Patent number: 10108620
    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for associating still images and videos. One method includes receiving a plurality of images and a plurality of videos and determining whether the images are related to the videos. The determining includes, for an image and a video, extracting features from the image and extracting features frames of the video, and comparing the features to determine whether the image is related to the video. The method further includes maintaining a data store storing data associating each image with each video determined to be related to the image.
    Type: Grant
    Filed: April 29, 2011
    Date of Patent: October 23, 2018
    Assignee: Google LLC
    Inventors: Ming Zhao, Yang Song, Hartwig Adam, Ullas Gargi, Yushi Jing, Henry A. Rowley
  • Patent number: 10108602
    Abstract: An approach is provided to discover new portmanteau, such as when ingesting documents into a question answering (QA) system. The approach works by analyzing a words included in electronic documents and identifies words as being possible portmanteaus. To analyze a portmanteau found in a document, the approach identifies morphemes that are included in the identified portmanteau and candidate words that correspond to each of the identified morphemes. A meaning for the new portmanteau is then derived from the meanings of the candidate word meanings.
    Type: Grant
    Filed: August 27, 2017
    Date of Patent: October 23, 2018
    Assignee: International Business Machines Corporation
    Inventors: Corville O. Allen, Albert A. Chung, Andrew R. Freed, Sorabh Murgai
  • Patent number: 10108661
    Abstract: An approach is provided in which a knowledge manager analyzes a corpus of documents based upon relations corresponding to entities in a structured resource and constructs a natural language context associated with a set of the relations. The knowledge manager maps the natural language context to a database query and, in turn, invokes the database query when the knowledge manager matches a question to the natural language context.
    Type: Grant
    Filed: September 20, 2015
    Date of Patent: October 23, 2018
    Assignee: International Business Machines Corporation
    Inventors: Timothy A. Bishop, Stephen A. Boxwell, Benjamin L. Brumfield, Nirav P. Desai, Stanley J. Vernier
  • Patent number: 10108605
    Abstract: Disclosed are systems and methods for natural language processing. The method includes receiving, by a first processor, a natural language input including a raw text document and a desired output request, defining an ordered sequence of natural language tasks based at least in part on the natural language input, and generating a data structure including the raw text document. The first processor can then transmit, to a second processor, the data structure and a request to execute a first natural language task in the ordered sequence, and the second processor can execute the first natural language task based at least in part on the data structure to produce a first task output. The second processor can then insert the first task output into the data structure and transmit the data structure having the first task output to the first processor. The first processor can then generate an output based at least in part on the data structure and the desired output request.
    Type: Grant
    Filed: June 9, 2016
    Date of Patent: October 23, 2018
    Assignee: PHARM3R LLC
    Inventor: Greg Leighton
  • Patent number: 10108906
    Abstract: Systems and computer program products to provide selective supporting evidence processing by applying a first machine learning (ML) model to a first candidate answer to generate a first confidence score that does not consider supporting evidence for the first candidate answer, determining, from a second ML model, an expected contribution of processing supporting evidence for the first candidate answer, and upon determining that the expected contribution does not exceed a specified threshold, skipping supporting evidence processing for the first candidate answer.
    Type: Grant
    Filed: December 19, 2014
    Date of Patent: October 23, 2018
    Assignee: International Business Machines Corporation
    Inventors: Michael W. Britt, Adam C. Lange-Pearson, Thomas P. Pitzen, Richard J. Stevens
  • Patent number: 10108869
    Abstract: Provided is a device including: a display unit configured to display handwritten content based on an analog handwritten input of a user; a user input unit that receives a user input of selecting a portion of the handwritten content displayed on the display unit; and a control unit reproduces a segment of multimedia content, which corresponds to the portion of the handwritten content, from the multimedia content synchronized with the handwritten content.
    Type: Grant
    Filed: May 12, 2017
    Date of Patent: October 23, 2018
    Assignee: SAMSUNG ELECTRONICS CO., LTD.
    Inventors: Pil-seung Yang, Da-hye Park, Seol-hye Won, In-kuk Yun, Yong-gook Park
  • Patent number: 10110533
    Abstract: Identifying entities in email signature blocks is described. A system scores each token, in a sequence of tokens from an email signature block, based on entity types, wherein each token is a word, a punctuation symbol, or an end-of-line character. The system identifies each entity sequence which includes a number of entities that matches the number of tokens in the sequence of tokens. The system identifies an entity sequence with a highest score based on applying scores for each token in the sequence of tokens to each identified entity sequence. The system outputs the sequence of tokens as an identified set of entities based on the entity sequence with the highest score.
    Type: Grant
    Filed: October 28, 2014
    Date of Patent: October 23, 2018
    Assignee: salesforce.com, inc.
    Inventor: Arun Jagota
  • Patent number: 10102275
    Abstract: A method, system or computer usable program product for utilizing a dialectical model for providing an answer to a user query including applying NLP to the query to generate a first set of NLP structures; generating a contrary set of NLP structures contrary to the user query; evaluating the first set of NLP structures to identify a first set of evidence; evaluating the contrary set of NLP structures to identify a second set of evidence; evaluating a first set of hypotheses from the first set of NLP structures and a contrary set of hypotheses from the contrary set of NLP structures to determine a set of answers to the user query; summarizing the set of answers including indicating derivation thereof; and providing the summarized set of answers to the user.
    Type: Grant
    Filed: May 27, 2015
    Date of Patent: October 16, 2018
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Donna K. Byron, Scott H. Isensee, Ashok Kumar, Mark Marrara, Mary D. Swift, Cale R. Vardy
  • Patent number: 10102192
    Abstract: Systems, methods, and computer-readable media are disclosed for enhancing an email application to automatically analyze an email thread and generate a compact content summary. The content summary is based on relative content contributions provided by the constituent email messages in the email thread. The content summary may be presented in a special window without disturbing or modifying the email thread or its constituent email messages. The distinctive content summary disclosed herein comprises certain sentences that are automatically gleaned from the email thread, analyzed relative to other sentences, and presented in a chronological sequence so that the user can quickly determine what the email thread is about and/or the current status of the conversation. The content summary is based on email weights, word weights, and intersecting sentence pairs.
    Type: Grant
    Filed: October 28, 2016
    Date of Patent: October 16, 2018
    Assignee: Commvault Systems, Inc.
    Inventors: Mrityunjay Upadhyay, Divakar Radhakrishnan, Madhusudhanan Natarajan
  • Patent number: 10102289
    Abstract: An approach is provided to analyze posts included threads of an online forum. The analyzing identifies a main topic related to a parent post of the thread. Child posts of the thread are selected with the parent post being a parent to each of the child posts. Child topics are identified for each of the child posts. A relevance of each of the child posts is determined by comparing the identified main topic to each of the identified child topics. Child posts are selected based on the relevance of the child posts. Parent post data is ingested into a corpus utilized by a question answering (QA) system. Data from the selected child posts is also ingested into the corpus.
    Type: Grant
    Filed: December 2, 2014
    Date of Patent: October 16, 2018
    Assignee: International Business Machines Corporation
    Inventors: Corville O. Allen, Andrew R. Freed
  • Patent number: 10095801
    Abstract: Interaction is provided between a first content set and a second content set, both of which are loaded into a data structure. When an event associated with loading of the second content set is detected, the second content set is parsed to identify at least one sub-set of the second content set. The identified sub-set is checked against a first data set associated with the first content set to determine whether the identified sub-set matches the first data set. If a match is found, an action associated with the least one identified sub-set is executed and the data structure is modified.
    Type: Grant
    Filed: August 17, 2015
    Date of Patent: October 9, 2018
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventor: Julian L. Friedman
  • Patent number: 10095692
    Abstract: The present invention relates to a system and method for bootstrapping templates for use in natural language sentence generation. More specifically, the present invention relates to identifying a set of candidate sentences from a large corpus based on a set of original templates by using a similarity measure. The set of candidate sentences are then processed or cleaned to generate a set of templates for use in natural language sentence generation.
    Type: Grant
    Filed: May 29, 2015
    Date of Patent: October 9, 2018
    Assignee: Thornson Reuters Global Resources Unlimited Company
    Inventors: Dezhao Song, Blake Howald, Frank Schilder
  • Patent number: 10095736
    Abstract: An approach is provided in which a knowledge manager analyzes a corpus of documents based upon relations corresponding to entities in a structured resource and constructs a natural language context associated with a set of the relations. The knowledge manager maps the natural language context to a database query and, in turn, invokes the database query when the knowledge manager matches a question to the natural language context.
    Type: Grant
    Filed: November 3, 2014
    Date of Patent: October 9, 2018
    Assignee: International Business Machines Corporation
    Inventors: Timothy A. Bishop, Stephen A. Boxwell, Benjamin L. Brumfield, Nirav P. Desai, Stanley J. Vernier
  • Patent number: 10095691
    Abstract: First specification data defining a grammar corresponding to natural language, and second specification data defining a mapping between the grammar and one or more outputs to be generated, are received. An electronic object is generated using the first specification data and the second specification data, the electronic object being configured to, when executed by one or more processors, cause the one or more processors to map received natural language input to the one or more outputs. The one or more outputs may be used to control an electronically controllable machine.
    Type: Grant
    Filed: March 22, 2016
    Date of Patent: October 9, 2018
    Assignee: Wolfram Research, Inc.
    Inventor: Stephen Wolfram
  • Patent number: 10088972
    Abstract: Techniques and architectures for implementing a team of virtual assistants are described herein. The team may include multiple virtual assistants that are configured with different characteristics, such as different functionality, base language models, levels of training, visual appearances, personalities, and so on. The characteristics of the virtual assistants may be configured by trainers, end-users, and/or a virtual assistant service. The virtual assistants may be presented to end-users in conversation user interfaces to perform different tasks for the users in a conversational manner. The different virtual assistants may adapt to different contexts. The virtual assistants may additionally, or alternatively, interact with each other to carry out tasks for the users, which may be illustrated in conversation user interfaces.
    Type: Grant
    Filed: June 2, 2014
    Date of Patent: October 2, 2018
    Assignee: VERINT AMERICAS INC.
    Inventors: Fred A Brown, Tanya M Miller