Patents by Inventor Yoshihiko Suhara

Yoshihiko Suhara has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Systems and methods for enhanced review comprehension using domain-specific knowledgebases
    Patent number: 11934783
    Abstract: Disclosed embodiments relate to natural language processing. Techniques can include receiving input text, extracting, from the input text, at least one modifier and aspect pair, receiving data from a knowledgebase, based on the at least one modifier and aspect pair and commonsense data, generate one or more premise embeddings, convert the input text into tokens, generating at least one vector for one or more of the tokens based on an analysis of the tokens, combine the at least one vector with the one or more premise embeddings to create at least one combined vector, and analyze the at least one combined vector wherein the analysis generates an output indicative of a feature of the input text.
    Type: Grant
    Filed: April 4, 2023
    Date of Patent: March 19, 2024
    Assignee: RECRUIT CO., LTD.
    Inventors: Yoshihiko Suhara, Behzad Golshan, Yuliang Li, Chen Chen, Xiaolan Wang, Jinfeng Li, Wang-Chiew Tan, çagatay Demiralp, Aaron Traylor
  • SYSTEMS AND METHODS FOR ENHANCED REVIEW COMPREHENSION USING DOMAIN-SPECIFIC KNOWLEDGEBASES
    Publication number: 20230281390
    Abstract: Disclosed embodiments relate to natural language processing. Techniques can include receiving input text, extracting, from the input text, at least one modifier and aspect pair, receiving data from a knowledgebase, based on the at least one modifier and aspect pair and commonsense data, generate one or more premise embeddings, convert the input text into tokens, generating at least one vector for one or more of the tokens based on an analysis of the tokens, combine the at least one vector with the one or more premise embeddings to create at least one combined vector, and analyze the at least one combined vector wherein the analysis generates an output indicative of a feature of the input text.
    Type: Application
    Filed: April 4, 2023
    Publication date: September 7, 2023
    Inventors: Yoshihiko SUHARA, Behzad GOLSHAN, Yuliang LI, Chen CHEN, XIAOLAN WANG, JINFENG LI, WANG-CHIEW TAN, ÇAGATAY DEMIRALP, AARON TRAYLOR
  • Systems and methods for enhanced review comprehension using domain-specific knowledgebases
    Patent number: 11620448
    Abstract: Disclosed embodiments relate to natural language processing. Techniques can include receiving input text, extracting, from the input text, at least one modifier and aspect pair, receiving data from a knowledgebase, based on the at least one modifier and aspect pair and commonsense data, generate one or more premise embeddings, convert the input text into tokens, generating at least one vector for one or more of the tokens based on an analysis of the tokens, combine the at least one vector with the one or more premise embeddings to create at least one combined vector, and analyze the at least one combined vector wherein the analysis generates an output indicative of a feature of the input text.
    Type: Grant
    Filed: August 31, 2020
    Date of Patent: April 4, 2023
    Assignee: RECRUIT CO., LTD.
    Inventors: Yoshihiko Suhara, Behzad Golshan, Yuliang Li, Chen Chen, Xiaolan Wang, Jinfeng Li, Wang-Chiew Tan, Çağatay Demiralp, Aaron Traylor
  • SYSTEMS AND METHODS FOR ENHANCED REVIEW COMPREHENSION USING DOMAIN-SPECIFIC KNOWLEDGEBASES
    Publication number: 20220067282
    Abstract: Disclosed embodiments relate to natural language processing. Techniques can include receiving input text, extracting, from the input text, at least one modifier and aspect pair, receiving data from a knowledgebase, based on the at least one modifier and aspect pair and commonsense data, generate one or more premise embeddings, convert the input text into tokens, generating at least one vector for one or more of the tokens based on an analysis of the tokens, combine the at least one vector with the one or more premise embeddings to create at least one combined vector, and analyze the at least one combined vector wherein the analysis generates an output indicative of a feature of the input text.
    Type: Application
    Filed: August 31, 2020
    Publication date: March 3, 2022
    Inventors: Yoshihiko SUHARA, Behzad GOLSHAN, Yuliang LI, Chen CHEN, XIAOLAN WANG, JINFENG LI, WANG-CHIEW TAN, ÇAGATAY DEMIRALP, AARON TRAYLOR
  • SYSTEMS AND METHODS FOR MULTILINGUAL SENTENCE EMBEDDINGS
    Publication number: 20220067279
    Abstract: Disclosed embodiments relate to natural language processing.
    Type: Application
    Filed: August 31, 2020
    Publication date: March 3, 2022
    Inventors: Wataru HIROTA, Yoshihiko SUHARA, Behzad GOLSHAN, Wang-Chiew TAN
  • Similarity Learning System and Similarity Learning Method
    Publication number: 20190164109
    Abstract: Even if the number of documents for the number of words is insufficient, appropriate similarity degree learning is performed. An analysis method by a topic model is used to perform learning of a degree of similarity between recruitment information and resume information. By analyzing recruitment information registered with a recruitment card database DB 310 and resume information registered with a resume database DB 320 using a topic model, a characteristics extracting portion 330 collects words (keywords) extracted from documents constituting the recruitment information and the resume information for each topic; and a similarity degree learning portion 360 performs similarity degree learning for each topic.
    Type: Application
    Filed: January 25, 2017
    Publication date: May 30, 2019
    Inventors: Yoshihiko Suhara, Hideki Awashima, Hidekazu Oiwa