DATA PROCESSING METHOD AND DEVICE
    1.
    发明公开
    DATA PROCESSING METHOD AND DEVICE 审中-公开
    DATENVERARBEITUNGSVERFAHREN UND -VORRICHTUNG

    公开(公告)号:EP3188081A4

    公开(公告)日:2017-09-20

    申请号:EP15842270

    申请日:2015-09-02

    IPC分类号: G06K9/00 G06K9/62 G10L15/06

    摘要: Embodiments of the present invention provide a data processing method and device. The method includes: obtaining target perception data, where the target perception data is any type of the following data: image data, video data, or voice data; determining a target scenario to which the target perception data belongs; determining a target perception model corresponding to the target scenario; and computing a recognition result of the target perception data according to the target perception model. Therefore, according to the data processing method and device in the embodiments of the present invention, a scenario to which perception data belongs is determined, and a recognition result of the perception data is obtained through computation by using a perception model corresponding to the scenario, which can reduce computation complexity and therefore can improve data processing efficiency in comparison with the prior art.

    摘要翻译: 本发明的实施例提供了一种数据处理方法和设备。 该方法包括:获取目标感知数据,目标感知数据为以下任何一种数据:图像数据,视频数据或语音数据; 确定目标感知数据所属的目标场景; 确定与所述目标场景对应的目标感知模型; 根据目标感知模型计算目标感知数据的识别结果。 因此,根据本发明实施例的数据处理方法和装置,确定感知数据所属的场景,通过使用该场景对应的感知模型进行计算,得到感知数据的识别结果, 与现有技术相比,可以降低计算复杂度并且因此可以提高数据处理效率。

    THREAT DETECTION METHOD AND APPARATUS, AND NETWORK SYSTEM

    公开(公告)号:EP3306510A4

    公开(公告)日:2018-07-18

    申请号:EP17795253

    申请日:2017-01-13

    发明人: CHEN JIA

    摘要: A threat detection method and apparatus, and a network system are disclosed, which relate to the field of computer technologies, and can resolve a missed detection problem of a Web sandbox in a delay loading scenario. The method includes: when loading a URL in a browser of a Web sandbox, obtaining page code of a first display page group identified by the URL and an overall size occupied by the first display page group in a display area of the browser; injecting preset dynamic code into the page code of the first display page group; parsing and executing the page code that is of the first display page group and that includes the preset dynamic code; sending a request message when detecting that a value of a display variable is greater than or equal to a preset value, to request to obtain page code of a second display page group; receiving a response message that carries the page code of the second display page group; and detecting, in the Web sandbox, whether the page code of the second display page group carries attack code.

    SYSTEM AND METHOD FOR TRAINING PARAMETER SET IN NEURAL NETWORK

    公开(公告)号:EP3196809A4

    公开(公告)日:2017-11-22

    申请号:EP15879628

    申请日:2015-08-04

    发明人: CHEN JIA ZENG JIA

    IPC分类号: G06N3/08 G06N3/04 G06N5/04

    摘要: The present invention discloses a system and a method for training a parameter set in a neural network. The system includes: a main-control-node set, used for controlling a training process and storing a data set and a parameter set that are used for training, where the main-control-node set includes M main control nodes, every two of the M main control nodes are in a communication connection, and at least one main control node of the M main control nodes is configured to back up the parameter set; and N training-node sets, where the training-node set includes multiple training nodes, and the training node is configured to perform training according to a data set and a parameter set that are delivered by the main-control-node set, and send a training result to a corresponding main control node. The system and the method in embodiments of the present invention can avoid a case in which entire training fails when a main control node is disabled, which improves reliability of a training process, and enhances training efficiency by performing training in parallel by configuring multiple training-node sets.