-
公开(公告)号：US20230095606A1
公开(公告)日：2023-03-30
申请号：US18070682
申请日：2022-11-29
Applicant: Huawei Technologies Co., Ltd.
Inventor： Chanfei SU , Yong Wen , Kalander Marcus , Lujia Pan
IPC: G06N3/04 , G06V10/774
Abstract: A data processing method and apparatus are disclosed. The method includes: obtaining a sample dataset, where each sample in the sample dataset includes a first label; dividing the sample dataset into K sample sub-datasets, determining a group of data from the K sample sub-datasets as a test dataset, and using sample sub-datasets other than the test dataset as a train dataset; training the classifier by using the train dataset, and classifying the test dataset by using a trained classifier, to obtain a second label of each sample in the test dataset; obtaining a first indicator and a first hyper-parameter at least based on the first label and the second label; obtaining a loss function of the classifier at least based on the first hyper-parameter, where the loss function is used to update the classifier; and completing training of the classifier when the first indicator meets a preset condition.
-
公开(公告)号：US09740816B2
公开(公告)日：2017-08-22
申请号：US13953486
申请日：2013-07-29
Applicant: Huawei Technologies Co., Ltd.
CPC classification number: G06F17/509 , G06F11/3447 , G06F11/3457 , H04L41/145 , H04L43/08 , H04L63/1408
Abstract: Embodiments of the present invention provide a method and an apparatus for network traffic simulation. The method includes: obtaining traffic information and user requirement information of a to-be-simulated IP network; analyzing the traffic information to generate a traffic characteristic; according to correspondence between user requirement information and traffic models, obtaining a traffic model corresponding to the user requirement information; and generating, through simulation, traffic of the to-be-simulated IP network according to the traffic characteristic and the traffic model. The embodiments of the present invention can improve flexibility and an emulation degree of traffic simulation.
-
公开(公告)号：US11140021B2
公开(公告)日：2021-10-05
申请号：US16351033
申请日：2019-03-12
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor： Lujia Pan , Caifeng He , Jianfeng Zhang
Abstract: This application discloses a method and an apparatus for determining a fault type. The method includes: performing online real-time calculation on operating data generated by each of a plurality of users within a preset period, to obtain an operating feature value corresponding to the operating data generated by each of the plurality of users within the preset period; receiving a fault classification request, where the fault classification request requests to determine a fault type of a fault that is caused for a target user before a target moment; and determining, according to the fault classification request, the fault type of the fault that is caused for the target user before the target moment based on a fault classification model and an operating feature value that is of the target user within at least one preset period.
-
公开(公告)号：US20150112903A1
公开(公告)日：2015-04-23
申请号：US14587724
申请日：2014-12-31
Applicant: Huawei Technologies Co., Ltd.
Inventor： Wun Wah Edmond Chan , Lujia Pan
CPC classification number: G06N5/04 , G06F17/30961 , G06N99/005
Abstract: Embodiments of the present invention disclose a defect prediction method and apparatus, which relate to the data processing field, and implement accurate and quick locating of a defect in a faulty product. A specific solution is as follows: selecting a training attribute set from a pre-stored product fault record according to a target attribute, and combining the target attribute and the training attribute set into a training set, where the target attribute is a defect attribute of a historical faulty product; generating a classifier set according to the training set, where the classifier set includes at least two tree classifiers; and predicting a defect of a faulty product by using the classifier set as a prediction model. The present invention is used in a process of predicting a defect of a faulty product.
Abstract translation: 本发明的实施例公开了与数据处理领域相关的缺陷预测方法和装置，并且对故障产品中的缺陷进行准确和快速的定位。 具体方案如下：根据目标属性从预先存储的产品故障记录中选择训练属性集合，并将目标属性和训练属性集合组合成训练集合，其中目标属性是缺陷属性 一个历史错误的产品; 根据训练集生成分类器组，其中分类器集包括至少两个树分类器; 并通过使用分类器集作为预测模型来预测故障产品的缺陷。 本发明用于预测故障产品的缺陷的过程。
-
公开(公告)号：US10079770B2
公开(公告)日：2018-09-18
申请号：US14836642
申请日：2015-08-26
Applicant: Huawei Technologies Co., Ltd.
Inventor： Shuqiang Wang , Lujia Pan , Cheng He
IPC: G06F15/16 , H04L12/823 , H04L29/06 , H04L29/08 , G06F12/00
Abstract: A junk information filtering method, including: detecting an Hypertext Transfer Protocol (HTTP) request data packet set generated according to that a user visits a target uniform resource locator (URL) by using a browser of a client, where the HTTP request data packet set includes at least one junk information request data packet and at least one non-junk information request data packet; identifying the junk information request data packet in the HTTP request data packet set according to a preset classification rule; and preventing, according to a preset interception policy, a destination host corresponding to the junk information request data packet from returning a junk information response data packet to the client. The embodiments of the present invention further disclose a junk information filtering apparatus. According to the present invention, accuracy of web page junk information filtering can be improved and user traffic can be saved.
-
Patent Agency Ranking
公开(公告)号：US09813549B2
公开(公告)日：2017-11-07
申请号：US14971103
申请日：2015-12-16
Applicant: Huawei Technologies Co., Ltd.
Inventor： Yijun Liu , Lujia Pan , Guangjian Tian
CPC classification number: H04M3/2227 , G06F17/30 , G06Q10/06
Abstract: A voice service evaluation processing method and apparatus may include acquiring record data related to a to-be-evaluated voice service; using a voice service evaluation model that is acquired according to record data related to a voice service to evaluate the record data related to the to-be-evaluated voice service, so as to acquire an evaluation value of the record data related to the to-be-evaluated voice service; and performing, according to the evaluation value, evaluation processing on a voice service corresponding to the record data related to the to-be-evaluated voice service. The embodiments of the present disclosure provide a more reliable basis for improvement and optimization of a voice service. In addition, an actual status of a network may further be reflected according to evaluation performed on record data related to the voice service, and adjustment on a network indicator may further be performed according to the reflected network status.
-
公开(公告)号：US20150365334A1
公开(公告)日：2015-12-17
申请号：US14836642
申请日：2015-08-26
Applicant: Huawei Technologies Co., Ltd.
Inventor： Shuqiang Wang , Lujia Pan , Cheng He
IPC: H04L12/823 , H04L29/06 , H04L29/08
Abstract: A junk information filtering method, including: detecting an Hypertext Transfer Protocol (HTTP) request data packet set generated according to that a user visits a target uniform resource locator (URL) by using a browser of a client, where the HTTP request data packet set includes at least one junk information request data packet and at least one non-junk information request data packet; identifying the junk information request data packet in the HTTP request data packet set according to a preset classification rule; and preventing, according to a preset interception policy, a destination host corresponding to the junk information request data packet from returning a junk information response data packet to the client. The embodiments of the present invention further disclose a junk information filtering apparatus. According to the present invention, accuracy of web page junk information filtering can be improved and user traffic can be saved.
Abstract translation: 一种垃圾信息过滤方法，包括：根据用户通过使用客户端的浏览器访问目标统一资源定位符（URL）生成的超文本传输协议（HTTP）请求数据包集，其中HTTP请求数据包集合 包括至少一个垃圾信息请求数据包和至少一个非垃圾信息请求数据包; 根据预先设定的分类规则识别HTTP请求数据包集中的垃圾信息请求数据包; 以及根据预设的截取策略，将与所述垃圾信息请求数据包相对应的目的主机从所述客户端返回垃圾信息响应数据包。 本发明的实施例还公开了一种垃圾信息过滤装置。 根据本发明，可以提高网页垃圾信息过滤的准确性，并且可以节省用户流量。
-
公开(公告)号：US20140058985A1
公开(公告)日：2014-02-27
申请号：US14063715
申请日：2013-10-25
Applicant: Huawei Technologies Co., LTD
Inventor： Andrey Osin , Xiuqiang He , Lujia Pan , Cheng He
CPC classification number: G06N3/02 , G06N3/08 , H04L41/142 , H04L41/145 , H04L41/147 , H04L41/16 , H04L43/026 , H04L43/028
Abstract: Embodiments of the present invention provide a method and device for synthesis of network traffic. The method includes extracting a first real traffic composition parameter sequence and a second real traffic composition parameter sequence from real traffic. A first synthetic traffic composition parameter sequence is generated. Relational coefficients between first traffic composition parameters and second traffic composition parameters are obtained. A second synthetic traffic composition parameter sequence is generated and synthetic traffic is generated according to the first synthetic traffic composition parameter sequence and the second synthetic traffic composition parameter sequence.
Abstract translation: 本发明的实施例提供了一种用于合成网络业务的方法和装置。 该方法包括从真实业务中提取第一真实业务组合参数序列和第二实际业务组合参数序列。 生成第一个合成交通组合参数序列。 获得第一交通组合参数与第二交通组合参数之间的关系系数。 生成第二合成业务组合参数序列，并根据第一合成业务组合参数序列和第二合成业务组合参数序列生成合成业务。
-
公开(公告)号：US10068176B2
公开(公告)日：2018-09-04
申请号：US14587724
申请日：2014-12-31
Applicant: Huawei Technologies Co., Ltd.
Inventor： Wun Wah Edmond Chan , Lujia Pan
Abstract: Embodiments of the present invention disclose a defect prediction method and apparatus, which relate to the data processing field, and implement accurate and quick locating of a defect in a faulty product. A specific solution is as follows: selecting a training attribute set from a pre-stored product fault record according to a target attribute, and combining the target attribute and the training attribute set into a training set, where the target attribute is a defect attribute of a historical faulty product; generating a classifier set according to the training set, where the classifier set includes at least two tree classifiers; and predicting a defect of a faulty product by using the classifier set as a prediction model. The present invention is used in a process of predicting a defect of a faulty product.
-
10.发明授权公开(公告)号：US09202088B2
公开(公告)日：2015-12-01
申请号：US13687823
申请日：2012-11-28
Applicant: Huawei Technologies Co., Ltd.
CPC classification number: G06K7/00 , G01S5/0284 , H04W64/00
Abstract: A method and an apparatus for selecting a reference tag used for positioning are provided. The method includes: obtaining first signal strength emitted by a reference tag and second signal strength emitted by a tag to be positioned, where the first signal strength and the second signal strength are detected by each radio frequency reader; comparing the first signal strength with the second signal strength to determine a boundary of the radio frequency readers, and dividing a network into a first area and a second area with the boundary; determining an estimated direction of the tag to be positioned relative to the reference tag according to a relative position of the reference tag and the boundary; and according to the estimated direction and according to the specified number of reference tags, selecting a valid reference tag group as the reference tag used for positioning.
Abstract translation: 提供了一种用于选择用于定位的参考标签的方法和装置。 该方法包括：获取由参考标签发出的第一信号强度和由待定位的标签发射的第二信号强度，其中由每个射频读取器检测第一信号强度和第二信号强度; 将第一信号强度与第二信号强度进行比较，以确定射频读取器的边界，并将网络划分成具有边界的第一区域和第二区域; 根据参考标签和边界的相对位置确定相对于参考标签定位的标签的估计方向; 并根据估计的方向和根据指定数量的参考标签，选择有效的参考标签组作为用于定位的参考标签。
-
-
-
-
-
-
-
-
-
Search Results
17
records
(took 0.022 seconds)
