-
公开(公告)号：US20220076320A1
公开(公告)日：2022-03-10
申请号：US17530672
申请日：2021-11-19
Inventor： Bingjie HU , Shichen SHAO , Yongheng LI , Yufang ZHANG
Abstract: A content recommendation method, a device, and a storage medium are provided, which are related to technical fields of knowledge graph, big data, and the Internet. The specific implementation scheme includes: determining, among candidate producers, a target producer to whom a private domain content recommendation is performed according to product information of the candidate producers; establishing a recommended product set according to correlation among product information of the target producer; and performing a private domain content recommendation to the target producer based on the recommended product set.
-
公开(公告)号：US20220059074A1
公开(公告)日：2022-02-24
申请号：US17520799
申请日：2021-11-08
Inventor： Yanyan LI , Dejing DOU
Abstract: A method for evaluating satisfaction with voice interaction, a device, and a storage medium are provided, which are related to a technical field of artificial intelligence, in particular, to fields of natural language processing, knowledge graph and deep learning, and can be applied to user intention understanding. The specific implementation includes: acquiring sample interaction data of a plurality of rounds of sample voice interaction behaviors; performing feature extractions on respective sample interaction data, to obtain a sample interaction feature sequence; acquiring satisfaction marks corresponding to the respective sample interaction data, to obtain a satisfaction mark sequence; and training an initial model by using a plurality of sets of sample interaction feature sequences and of satisfaction mark sequences, to obtain the model for evaluating satisfaction.
-
公开(公告)号：US20220036880A1
公开(公告)日：2022-02-03
申请号：US17451380
申请日：2021-10-19
Inventor： Liao Zhang , Zhengxiang Jiang , Xiaoyin Fu
IPC: G10L15/06 , G06F40/30 , G06F40/253
Abstract: The present disclosure provides a method for training a linguistic model, related to fields of speech, natural language processing, deep learning technologies. A method includes: obtaining grammars corresponding to a plurality of sample texts and a slot value of a slot in each grammar by using semantic analysis; generating a grammar graph corresponding to each grammar based on the corresponding grammar and the slot value of the slot in the corresponding grammar; obtaining a weight of each grammar, a weight of each slot, and a weight of each slot value in each grammar graph based on the sample texts; determining at least one grammar frequency of each order based on the weight of each grammar, the weight of each slot, and the weight of each slot value in each grammar graph; and training the linguistic model based on the at least one grammar frequency of each order.
-
公开(公告)号：US20220036879A1
公开(公告)日：2022-02-03
申请号：US17500188
申请日：2021-10-13
Inventor： Jiaxiang GE , Zhen WU , Maoren ZHOU , Qiguang ZANG , Ming WEN , Xiaoyin FU
Abstract: A method for mining feature information, an apparatus for mining feature information and an electronic device are disclosed. The method includes: determining a usage scenario of a target device; obtaining raw audio data including real scenario data, speech synthesis data, recorded audio data and other media data; generating target audio data of the usage scenario by simulating the usage scenario based on the raw audio data; and obtaining feature information of the usage scenario by performing feature extraction on the target audio data.
-
公开(公告)号：US20220036241A1
公开(公告)日：2022-02-03
申请号：US17501003
申请日：2021-10-14
Inventor： Tianjian He , Dianhai Yu , Zhihua Wu , Daxiang Dong , Yanjun Ma
Abstract: The present disclosure discloses a method, an apparatus and a storage medium for training a deep learning framework, and relates to the artificial intelligence field such as deep learning and big data processing. The specific implementation solution is: acquiring at least one task node in a current task node cluster, that meets a preset opening condition when a target task meets a training start condition; judging whether a number of nodes of the at least one task node is greater than or equal to a preset number; synchronously training the deep learning framework of the target task by the at least one task node according to sample data if the number of nodes is greater than the preset number; and acquiring a synchronously trained target deep learning framework when the target task meets a training completion condition.
-
Patent Agency Ranking
公开(公告)号：US20220036230A1
公开(公告)日：2022-02-03
申请号：US17501755
申请日：2021-10-14
Inventor： Xin WANG , Xuanqiang Zhao , Benchi Zhao , Zihe Wang , Zhixin Song , Renyu Liu
IPC: G06N10/00
Abstract: A quantum entangled state processing method, a device, and a storage medium are provided, which are related to a field of quantum calculation. The specific implementation scheme includes: determining n initial quantum states to be processed; determining at least two nodes associated with the initial quantum state; acquiring at least one first parameterized quantum circuit required by the first node and at least one second parameterized quantum circuit required by the second node matched with a preset processing scenario; controlling, based on an initial quantum operation strategy, the first node to perform a local quantum operation to obtain a first measurement result, controlling the second node to perform a local quantum operation to obtain a second measurement result; obtaining an output quantum state meeting a preset requirement of the preset processing scenario at least based on the first measurement result and the second measurement result.
-
公开(公告)号：US11235763B2
公开(公告)日：2022-02-01
申请号：US16704043
申请日：2019-12-05
Inventor： Fan Yang , Fan Zhu , Weicheng Zhu , Leibing Lv , Lin Ma
IPC: B60W30/09 , B60W30/095 , G08G1/16
Abstract: The present disclosure provides a method, an apparatus, a device and a readable storage medium for preventing a vehicle collision. The method includes determining a current conservative anti-collision region of a vehicle acquiring the omnidirectional environmental information in the current region in the current conservative anti-collision region; predicting a behavior of an obstacle according to the omnidirectional environmental information in the current region, and controlling the vehicle to conduct an anti-collision behavior according a prediction result, a reasonable prediction may be made to the behavior of the obstacle based on the conservative anti-collision region and the omnidirectional environmental information, thus controlling the vehicle to conduct an accurate anti-collision behavior, thereby preventing collisions effectively, and avoiding an occurrence of traffic accidents.
-
公开(公告)号：US20220027705A1
公开(公告)日：2022-01-27
申请号：US17494497
申请日：2021-10-05
Abstract: A building positioning method, an electronic device, a storage medium and a terminal device are disclosed, which relate to the technical fields of artificial intelligence, computer vision domains and intelligent traffic. The method includes: acquiring a building fingerprint library, the building fingerprint library including a plurality of groups of triplet data and a plurality of corresponding building information, wherein a single group of triplet data includes surveying and mapping data, GPS data and Wi-Fi data; receiving a positioning request, the positioning request including first triplet data, the first triplet data including first surveying and mapping data, first GPS data, and first Wi-Fi data; calculating a similarity between the first triplet data and the plurality of groups of triplet data in the building fingerprint library, respectively; and determining building information corresponding to the positioning request according to the calculated similarity.
-
公开(公告)号：US20220027639A1
公开(公告)日：2022-01-27
申请号：US17494682
申请日：2021-10-05
Inventor： Yingying LI , Xiao TAN , Hao SUN
Abstract: A lane line detection method, an electronic device, and a storage medium, related to the field of artificial intelligence, and particularly related to computer vision and deep learning technologies, which can be applied to intelligent traffic scenes, are provided. The method includes: dividing an image into a foreground region and a background region; determining a solid line and a dotted line included in the foreground region; determining, according to the solid line and the dotted line comprised in the foreground region, whether a dotted-and-solid line is included in the foreground region; and determining a lane line detection result according to the solid line, the dotted line, and whether a dotted-and-solid line is comprised in the foreground region. According to the technical solution, the accuracy of lane line detection can be improved.
-
公开(公告)号：US20220027606A1
公开(公告)日：2022-01-27
申请号：US17494724
申请日：2021-10-05
Inventor： Tao HU , Xiangbo SU
Abstract: A human behavior recognition method, a device, and a storage medium are provided, which are related to the field of artificial intelligence, specifically to computer vision and deep learning technologies, and applicable to smart city scenarios. The method includes: obtaining attribute information of a target object and N pieces of candidate behavior-related information of a target human from a target image, wherein N is an integer greater than or equal to 1; determining target behavior-related information based on comparison results between the N pieces of candidate behavior-related information and the attribute information of the target object; and determining a behavior recognition result of the target human based on the target behavior-related information.
-
-
-
-
-
-
-
-
-
Search Results
1522
records
(took 0.088 seconds)
