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. 发明公开

EP4141695A1 RESPONSE METHOD IN MAN-MACHINE DIALOGUE, DIALOGUE SYSTEM, AND STORAGE MEDIUM 审中-实审
公开(公告)号：EP4141695A1
公开(公告)日：2023-03-01
申请号：EP21808970.4
申请日：2021-04-01
申请人： Huawei Technologies Co., Ltd.
发明人： ZENG, Yulong , WANG, Yasheng , WANG, Yadao
IPC分类号： G06F16/33
摘要： A response method in a human-computer dialogue, a dialogue system, and a storage medium are provided and belong to the field of artificial intelligence. In a process of a dialogue between a user and a machine, a user intent of a current dialogue is determined based on an expected user intent associated with a sentence replied by the machine to the user in a previous dialogue, so that a response is made. Because processing logic for an expected user intent is introduced, accuracy of a generated response sentence is improved.

2. 发明公开

EP3582118A1 METHOD AND APPARATUS FOR TRAINING CLASSIFICATION MODEL 审中-公开
公开(公告)号：EP3582118A1
公开(公告)日：2019-12-18
申请号：EP18839409.2
申请日：2018-06-29
申请人： Huawei Technologies Co., Ltd.
发明人： WANG, Yasheng , ZHANG, Yang , BI, Shuzhan , YAN, Youliang
IPC分类号： G06F17/00
摘要： This application provides a classification model training method and apparatus. The method includes: obtaining a positive training set and a first negative training set, where the positive training set includes samples of a positive sample set in a corpus, the first negative training set includes samples of an unlabeled sample set in the corpus, and the unlabeled sample set indicates a sample set that is in the corpus and that does not belong to a dictionary; performing training, by using the positive training set and the first negative training set, to obtain a first classification model; determining, by using the first classification model, a pseudo negative sample in the first negative training set, where the pseudo negative sample indicates a sample that is in the first negative training set and that is considered as a positive sample; removing the pseudo negative sample from the first negative training set, and updating the first negative training set to a second negative training set; and performing training, by using the positive training set and the second negative training set, to obtain a second classification model, where the second classification model is a target classification model. Therefore, according to the method provided in this application, accuracy of the classification model can be effectively improved. When the dictionary is extended by using the classification model, accuracy of the dictionary can also be improved.

3. 发明公开

EP3964998A1 TEXT PROCESSING METHOD AND MODEL TRAINING METHOD AND APPARATUS 审中-实审
公开(公告)号：EP3964998A1
公开(公告)日：2022-03-09
申请号：EP20804954.4
申请日：2020-01-17
申请人： Huawei Technologies Co., Ltd. , Tsinghua University
发明人： WANG, Yasheng , JIANG, Xin , CHEN, Xiao , LIU, Qun , ZHANG, Zhengyan , QI, Fanchao , LIU, Zhiyuan
IPC分类号： G06F40/00
摘要： This application relates to the field of artificial intelligence, and provides a text processing method, a model training method, and an apparatus. The method includes: obtaining target knowledge data, where the target knowledge data includes a first named entity, a second named entity, and an association between the first named entity and the second named entity; processing the target knowledge data to obtain a target knowledge vector; processing to-be-processed text to obtain a target text vector, where the to-be-processed text includes the first named entity; fusing the target text vector and the target knowledge vector based on a target fusion model, to obtain a fused target text vector and a fused target knowledge vector; and processing the fused target text vector and/or the fused target knowledge vector based on a target processing model, to obtain a processing result corresponding to a target task. The foregoing technical solution can improve accuracy of a result of processing a target task by the target processing model.

4. 发明公开

EP4102397A1 TEXT INFORMATION PROCESSING METHOD AND APPARATUS, COMPUTER DEVICE, AND READABLE STORAGE MEDIUM 审中-实审
公开(公告)号：EP4102397A1
公开(公告)日：2022-12-14
申请号：EP20917614.8
申请日：2020-09-14
申请人： Huawei Technologies Co., Ltd.
发明人： DENG, Liqun , WEI, Jiansheng , ZHANG, Yang , WANG, Yasheng , SUN, Wenhua
IPC分类号： G06F40/289 , G06F40/30 , G06F40/284 , G06F16/35
摘要： A text information processing method and apparatus, a computer device, and a readable storage medium are applicable to emotional speech synthesis in the artificial intelligence field. When an emotion category is determined for a sentence in text, the processing method considers both a predicted emotion category of each sentence and an overall emotion category of the text in which the sentence is located, and the method generates speech information for the sentence in the text, to help generate, for the text, the speech information that better conforms to human emotion expression habits, and to improve a humanoid degree of a smart speech device.

5. 发明公开

EP4404080A1 METHOD AND APPARATUS FOR DETERMINING REPLY STATEMENT 审中-公开
公开(公告)号：EP4404080A1
公开(公告)日：2024-07-24
申请号：EP22880379.7
申请日：2022-10-13
申请人： Huawei Technologies Co., Ltd.
发明人： HE, Bin , WANG, Yasheng , LI, Yitong , MI, Fei
IPC分类号： G06F16/33
CPC分类号： G06F16/33 , G06F16/332 , G06N3/02 , G06F18/214
摘要： This application discloses a response determining method. The method includes: obtaining a to-be-responded first user statement; determining first state information of the first user statement based on the first user statement by using a state determining network, where the first state information includes a first dialog type of the first user statement; and inputting the first user statement and the first dialog type into a response generation network, to obtain a response corresponding to the first user statement. In this application, a dialog type of a user dialog is identified by using the state determining network, and responses corresponding to different dialog types are generated by reusing the dialog generation network, which is equivalent to processing user statements of different dialog types by using a same model, thereby reducing model complexity and a model size of a dialog system.

6. 发明公开

EP4390753A1 TEXT DATA PROCESSING METHOD, NEURAL NETWORK TRAINING METHOD, AND RELATED DEVICES 审中-公开
公开(公告)号：EP4390753A1
公开(公告)日：2024-06-26
申请号：EP22869111.9
申请日：2022-09-08
申请人： Huawei Technologies Co., Ltd.
发明人： MENG, Xiaojun , WANG, Yasheng , JIANG, Xin , LIU, Qun
IPC分类号： G06F40/30 , G06F40/284 , G06N3/04 , G06N3/08
摘要： A text data processing method, a neural-network training method, and related devices are provided. The methods may be applied to the text data processing field in the artificial intelligence field. The method includes: obtaining a to-be-processed text, where the to-be-processed text includes a plurality of characters; and processing the to-be-processed text by using a target model to obtain a prediction result, where the prediction result indicates to split the to-be-processed text into a plurality of target character sets, the prediction result further includes a plurality of first labels, one first label indicates semantics of one target character set, and the plurality of first labels are used to determine an intention of the to-be-processed text. A reduplicated word or a modal auxiliary word can be split into another target character set, so that the intention of the to-be-processed text can still be understood even if the reduplicated word or the modal auxiliary word, or the like exists in the entire to-be-processed text. In this way, a natural language understanding method with a stronger generalization capability is provided.

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