-
公开(公告)号：US11164220B2
公开(公告)日：2021-11-02
申请号：US16025677
申请日：2018-07-02
发明人： Chunhui Wang , Xihong Chen , Jiahuan Zhai , Pengfei Liu , Wanhe Zhao
IPC分类号： G06Q30/00 , G06Q30/02 , G06F16/957 , G06F16/00
摘要： Embodiments of the present disclosure provide an information processing method, a server, and a computer storage medium. The method includes: collecting first information; selecting from the first information at least one information material, and generating, according to the first information material, a media information form template supporting at least two types of scenario presentation requirements; receiving a first request for the information material initiated by a data traffic monitor; sending the information material and the media information form template to the data traffic monitor; and generating a presentation result according to the information material and the media information form template.
-
公开(公告)号：US10929879B2
公开(公告)日：2021-02-23
申请号：US15971614
申请日：2018-05-04
发明人： Quan Cheng , Yiqun Li , Chunhui Wang
摘要： This application discloses a method and an apparatus for advertisement fraud reduction. A training sample set including multiple training samples is obtained. At least one of the multiple training samples, associated with a fraudulent training user, includes a training click log associated with clicking one or more advertisements by the fraudulent training user. Feature information from the training sample set is extracted. The fraudulent training user and the feature information are associated with a fraudulent user type. A positive sample associated with the feature information is formed based on the at least one of the multiple training samples. A fraudulent user identification model associated with the fraudulent user type is trained based on at least the positive sample. Further, a sample to be identified, associated with a user to be identified, is received. Whether the user is a fraudulent user is determining using the fraudulent user identification model.
-
公开(公告)号：US11373205B2
公开(公告)日：2022-06-28
申请号：US15989997
申请日：2018-05-25
发明人： Dongyu Li , Zuojie Peng , Jie Liu , Chunhui Wang , Yu Sun , Yiqun Li
摘要： Embodiments of the present disclosure disclose an information processing method, a server, and a non-volatile storage medium. The method includes obtaining first log information in a first time period and then obtaining, based on the first log information, terminal information of a terminal that performs a hit behavior on a media information display place. The method includes determining, based on the terminal information, regional information corresponding to the terminal, where the regional information is used to indicate a region in which the terminal is located. The method also includes determining whether the number of regions in which the terminal is located is greater than a first threshold in a preset time range according to the regional information. The method further includes, when the number is greater than the first threshold, obtaining first terminal information of the terminal and determining that the terminal is an abnormal terminal.
-
公开(公告)号：US20180253755A1
公开(公告)日：2018-09-06
申请号：US15971614
申请日：2018-05-04
发明人： Quan Cheng , Yiqun Li , Chunhui Wang
CPC分类号： G06Q30/0248 , G06F15/76 , G06K9/6259 , G06N20/00 , G06Q30/02 , G06Q30/0277
摘要： This application discloses a method and an apparatus for advertisement fraud reduction. A training sample set including multiple training samples is obtained. At least one of the multiple training samples, associated with a fraudulent training user, includes a training click log associated with clicking one or more advertisements by the fraudulent training user. Feature information from the training sample set is extracted. The fraudulent training user and the feature information are associated with a fraudulent user type. A positive sample associated with the feature information is formed based on the at least one of the multiple training samples. A fraudulent user identification model associated with the fraudulent user type is trained based on at least the positive sample. Further, a sample to be identified, associated with a user to be identified, is received. Whether the user is a fraudulent user is determining using the fraudulent user identification model.
-
-
-
搜索结果
4 条记录 (耗时 0.016 秒)
