利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

158 条记录 (耗时 0.03 秒)

    INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM
    2.
    发明公开
    INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM 审中-公开

    公开(公告)号：US20240354672A1

    公开(公告)日：2024-10-24

    申请号：US18685926

    申请日：2021-08-26

    申请人： NIPPON TELEGRAPH AND TELEPHONE CORPORATION

    发明人： Yoshiaki SHOJI ,  Shiro OGASAWARA

    IPC分类号： G06Q10/0631 ,  G06F16/383

    CPC分类号： G06Q10/063112 ,  G06F16/383

    摘要： An information processing apparatus includes an information acquisition unit that acquires first text information handled in a terminal apparatus of a first user in response to a search request issued by the first user to search for candidate users from among a plurality of users. There is a condition generation unit that divides words included in the first text information into a first word group having an appearance frequency lower than a first threshold and a second word group having an appearance frequency equal to or higher than the first threshold according to an appearance frequency in second text information related to past work of the first user. Further, there is a search processing unit that extracts the candidate users from among users according to an appearance frequency of a word belonging to the first word group in third text information related to past work of each user.

    CLINICAL CONTEXT CENTRIC NATURAL LANGUAGE PROCESSING SOLUTIONS
    4.
    发明公开
    CLINICAL CONTEXT CENTRIC NATURAL LANGUAGE PROCESSING SOLUTIONS 审中-公开

    公开(公告)号：US20240311580A1

    公开(公告)日：2024-09-19

    申请号：US18183295

    申请日：2023-03-14

    申请人： GE Precision Healthcare LLC

    发明人： Akshit Achara ,  Shivappa Goravar ,  Sanand Sasidharan ,  Anuradha Kanamarlapudi

    IPC分类号： G06F40/40 ,  G06F16/35 ,  G06F16/383

    CPC分类号： G06F40/40 ,  G06F16/35 ,  G06F16/383

    摘要： Techniques are described for generating natural language processing (NLP) solutions in the medical domain tailored based on clinical context. In one example a method comprises receiving, by a system comprising a processor, a plurality of medical reports respectively comprising medical text and separating the medical reports into different groups of similar reports respectively corresponding to different clinical contexts based on contextual attributes associated with the medical reports. The method further comprises training, by the system, training, by the system, a separate machine learning model to perform a natural language processing task on each group of the different groups of similar reports, resulting in different machine learning models respectively adapted to perform natural language processing tasks on the different groups such that the natural language processing tasks are tailored to the different clinical contexts respectively associated with the different groups.

    Method for providing enhanced search result by fusioning more than two information retrievals
    6.
    发明公开
    Method for providing enhanced search result by fusioning more than two information retrievals 审中-公开

    公开(公告)号：US20240168988A1

    公开(公告)日：2024-05-23

    申请号：US18527499

    申请日：2023-12-04

    申请人： Hoseo University Academic Cooperation Foundation ,  Livin AI Inc.

    发明人： Sung-Bum PARK ,  Suehyun CHANG

    IPC分类号： G06F16/383 ,  G06F16/338

    CPC分类号： G06F16/383 ,  G06F16/338

    摘要： A method is presented for enhancing search results by segmenting documents into smaller passages and utilizing those passages as the search unit. This method integrates the passage rankings from two search models to produce a new document ranking and arranges the documents accordingly. The ranking of document is also rearranged based on the proportion of passages taken from the same document versus the total number of passages in that document. The final search ranking system combines document-level and passage-level search rankings to rank documents.
    This method ensures that when conducting a passage search, the search results consider the general relevance of the entire document, which leads to better performance compared to searching only by document-level or passage-level searches.

    Automatic quoting or forking of chats using deep learning in cloud based chat applications
    8.
    发明授权
    Automatic quoting or forking of chats using deep learning in cloud based chat applications 有权

    公开(公告)号：US11971917B2

    公开(公告)日：2024-04-30

    申请号：US17227781

    申请日：2021-04-12

    申请人： Cisco Technology, Inc.

    发明人： Karthik Babu Harichandra Babu ,  Akaash Dev S C ,  Amit Kumar

    IPC分类号： G06F16/383 ,  G06F16/338

    CPC分类号： G06F16/383 ,  G06F16/338

    摘要： Presented herein are techniques for a non-intruding auto quoting feature on-the-fly for chat applications based on previous conversations and effectively forking/quoting such conversations to ease the burden of the user going through all the chat history. The techniques may involve monitoring an ongoing first text communication session between a first user and at least one second user; detecting one or more salient words or phrases in the first text communication session based on the monitoring; based on the one or more salient words or phrases, identifying a second text communication session that is topically related to the first text communication session; and displaying in a user interface presented to the first user a prompt to quote relevant content from the second text communication session into the first text communication session or to fork the first user to be a participant in the second text communication session.

    CLASSIFICATION SYSTEM, METHOD, AND PROGRAM
    9.
    发明公开
    CLASSIFICATION SYSTEM, METHOD, AND PROGRAM 审中-公开

    公开(公告)号：US20240119079A1

    公开(公告)日：2024-04-11

    申请号：US18276378

    申请日：2021-02-26

    申请人： NEC Corporation

    发明人： Taro YANO ,  Kunihiro TAKEOKA ,  Masafumi OYAMADA

    IPC分类号： G06F16/35 ,  G06F16/383

    CPC分类号： G06F16/35 ,  G06F16/383

    摘要： The input means 181 accepts inputs of test data, a hierarchical structure in which a node of bottom layer represents a target class, and a classification score of a seen class as the classification score indicating a probability that the test data is classified into each class. The unseen class score calculation means 182 calculates the classification score of an unseen class based on uniformity of the classification score of each seen class. The matching score calculation means 183 calculates a matching score indicating similarity between the test data and each class label. The final classification score calculation means 184 calculates a final classification score indicating a probability that the test data is classified into the class so that the larger the classification score of each class, and the matching score, the larger the final classification score.