CN116033428A - Method and device for processing abnormal telecommunication users - Google Patents
Method and device for processing abnormal telecommunication usersDownload PDFInfo
- Publication number
- CN116033428A CN116033428ACN202111232820.XACN202111232820ACN116033428ACN 116033428 ACN116033428 ACN 116033428ACN 202111232820 ACN202111232820 ACN 202111232820ACN 116033428 ACN116033428 ACN 116033428A
- Authority
- CN
- China
- Prior art keywords
- abnormal
- users
- user
- communication
- telecommunications
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Translated fromChinese
Description
Translated fromChinese技术领域technical field
本发明涉及大数据处理技术领域，具体涉及一种电信异常用户的处理方法及装置。另外，还涉及一种电子设备及处理器可读存储介质。The invention relates to the technical field of big data processing, in particular to a method and device for processing abnormal users of telecommunications. In addition, it also relates to an electronic device and a processor-readable storage medium.
背景技术Background technique
近年来，随着通信技术的快速发展，电信诈骗事件频发，引发极大社会危害，相关防治工作越来越受到人们的重视。电信诈骗是一项社会危害巨大且高发的犯罪行为，诈骗分子通过电话冒充他人恶意骗取财产,影响社会安定，并且随着互联网新技术的应用，电话诈骗日益呈现出多变性、对抗性等特点，快速针对电信异常用户进行预测和分析，以实现更有效的诈骗识别与态势分析成为当前研究重点。In recent years, with the rapid development of communication technology, telecommunication fraud incidents have occurred frequently, causing great social harm, and related prevention and control work has attracted more and more attention. Telecom fraud is a socially harmful and frequent crime. Fraudsters pretend to be others to maliciously defraud property through the phone, affecting social stability. With the application of new Internet technologies, telephone fraud is increasingly showing characteristics of variability and confrontation. Rapidly predicting and analyzing abnormal telecom users to achieve more effective fraud identification and situation analysis has become the focus of current research.
然而,目前现有的反诈方法通常仅建模单一的诈骗通讯行为特征，无法实现深入分析通话数据，不能形成更有效的诈骗识别与态势分析。因此，如何设计高效、精确的电信异常用户的预测处理方案成为亟待解决的难题。However, the existing anti-fraud methods usually only model a single characteristic of fraudulent communication behavior, and cannot realize in-depth analysis of call data, and cannot form more effective fraud identification and situation analysis. Therefore, how to design an efficient and accurate prediction and processing scheme for telecom abnormal users has become an urgent problem to be solved.
发明内容Contents of the invention
为此，本发明提供一种电信异常用户的处理方法及装置，以解决现有技术存在的电信异常用户的预测处理方案局限性较高，电信异常用户的识别精度和效率较差的问题。To this end, the present invention provides a method and device for processing abnormal telecommunication users to solve the problems in the prior art that the prediction and processing schemes for abnormal telecommunication users are relatively limited and the identification accuracy and efficiency of abnormal telecommunication users are poor.
第一方面，本发明提供一种电信异常用户的处理方法，包括：In the first aspect, the present invention provides a method for processing abnormal telecommunications users, including:
获取待分析用户的业务域数据和运营域数据；Obtain the business domain data and operational domain data of the user to be analyzed;
将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，得到所述待分析用户中的人证分离用户；Inputting the business domain data and the operation domain data into a recognition model for separation of witnesses and certificates for analysis, and obtaining users of separation of witnesses and certificates among the users to be analyzed;
将所述人证分离用户的通讯特征数据输入到电信异常用户识别模型，得到异常用户清单；其中，所述异常用户清单包含电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别；Inputting the communication characteristic data of the user whose identity is separated into the identification model of the abnormal user of telecommunications to obtain a list of abnormal users; wherein, the list of abnormal users includes the communication number of the abnormal user of telecommunications and the abnormal warning level corresponding to the communication number;
根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理。The communication numbers of the abnormal telecommunications users are processed according to the abnormal warning levels corresponding to the communication numbers in the abnormal user list and the preset hierarchical processing mechanism.
在一个实施例中，根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理，具体包括：In one embodiment, according to the abnormality warning level corresponding to the communication number in the abnormal user list and the preset hierarchical processing mechanism, the communication number of the abnormal telecommunications user is processed, specifically including:
若所述异常用户清单中所述通讯号码对应的异常预警级别为高危级别，则针对属于高危级别的电信异常用户的通讯号码进行通讯功能关停处理；和/或，If the abnormal warning level corresponding to the communication number in the abnormal user list is a high-risk level, the communication function shutdown processing is performed for the communication numbers of the abnormal telecommunications users belonging to the high-risk level; and/or,
若所述异常用户清单中的异常预警级别为中危级别，则将对应中危级别的电信异常用户的通讯号码按照预设的中级处理机制进行审核，若审核结果为不通过，则针对属于中危级别的电信异常用户的通讯号码进行通讯功能关停处理；和/或，If the abnormal warning level in the list of abnormal users is a medium-risk level, the communication numbers of the telecommunications abnormal users corresponding to the medium-risk level will be reviewed according to the preset intermediate-level processing mechanism. and/or,
若所述异常用户清单中所述人证分离用户的异常预警级别为低危级别，则将对应低危级别的电信异常用户的通讯号码按照预设的低级处理机制进行审核，若审核结果为不通过且预设通话时长内未完成二次实名认证，则针对属于低危级别的电信异常用户的通讯号码进行通讯功能关停处理。If the abnormal early warning level of the person-identity separation user mentioned in the abnormal user list is a low-risk level, the communication number of the telecommunications abnormal user corresponding to the low-risk level will be reviewed according to the preset low-level processing mechanism, and if the result of the review is not If it passes and the second real-name authentication is not completed within the preset call time, the communication function will be shut down for the communication numbers of low-risk telecom abnormal users.
在一个实施例中，所述的电信异常用户的处理方法，还包括：In one embodiment, the method for handling abnormal telecommunications users further includes:
对通讯功能关停处理的电信异常用户进行二次实名认证；Carry out secondary real-name authentication for abnormal telecom users whose communication functions are shut down;
若所述通讯功能关停处理的电信异常用户中存在二次实名认证结果为通过的目标电信异常用户，则将所述目标电信异常用户的通讯号码的通讯功能进行恢复。If there is a target telecommunication abnormal user whose second real-name authentication result is passed among the telecommunication abnormal users whose communication function is shut down, restore the communication function of the communication number of the target telecommunication abnormal user.
在一个实施例中，所述的电信异常用户的处理方法，还包括：In one embodiment, the method for handling abnormal telecommunications users further includes:
获取所述人证分离用户的运营域位置数据；Obtaining the location data of the operation domain of the user for separation of identity and identification;
基于所述运营域位置数据按照预设的时间间隔获取所述人证分离用户在预设高危位置的高危通信行为数据；Obtaining the high-risk communication behavior data of the user at a preset high-risk location for separation of identity and identification at a preset time interval based on the location data of the operation domain;
根据所述高危通信行为数据，确定所述人证分离用户中的电信异常用户及所述电信异常用户的异常预警级别；According to the high-risk communication behavior data, determine the abnormal telecommunications users among the users with identity separation and the abnormal warning level of the abnormal telecommunications users;
基于所述电信异常用户及所述电信异常用户的异常预警级别，更新所述异常用户清单。The abnormal user list is updated based on the abnormal telecommunications user and the abnormal warning level of the abnormal telecommunications user.
在一个实施例中，所述的电信异常用户的处理方法，还包括：构建所述电信异常用户识别模型；In one embodiment, the method for processing abnormal telecom users further includes: constructing an identification model for abnormal telecom users;
其中，所述构建所述电信异常用户识别模型，具体包括：Wherein, the construction of the telecom abnormal user identification model specifically includes:
获取异常用户的通讯特征数据，将所述异常用户的通讯特征数据确定为正样本集；Obtaining the communication feature data of the abnormal user, and determining the communication feature data of the abnormal user as a positive sample set;
获取非异常用户的通讯特征数据，将所述非异常用户的通讯特征数据确定为负样本集；Obtaining the communication characteristic data of the non-abnormal user, and determining the communication characteristic data of the non-abnormal user as a negative sample set;
利用预设的随机森林模型根据所述正样本集和所述负样本集构造相应决策树子模型，基于所述决策树子模型构建所述电信异常用户识别模型。A preset random forest model is used to construct a corresponding decision tree sub-model according to the positive sample set and the negative sample set, and the identification model of the abnormal telecommunications user is constructed based on the decision tree sub-model.
在一个实施例中，所述的电信异常用户的处理方法，还包括：优化所述电信异常用户识别模型；In one embodiment, the method for processing abnormal telecommunications users further includes: optimizing the identification model for abnormal telecommunications users;
其中，所述优化所述电信异常用户识别模型，具体包括：Wherein, the optimization of the telecom abnormal user identification model specifically includes:
获取未通过二次实名认证的电信异常用户的通讯号码，基于所述未通过二次实名认证的电信异常用户的通讯号码构建异常号码库；Obtaining the communication numbers of the abnormal telecom users who have not passed the second real-name authentication, and constructing an abnormal number database based on the communication numbers of the abnormal telecom users who have not passed the second real-name authentication;
基于预设的粒子群优化模型及所述异常号码库，并利用径向基函数对所述电信异常用户识别模型进行自适应训练，基于自适应训练过程优化所述电信异常用户识别模型。Based on the preset particle swarm optimization model and the abnormal number library, the radial basis function is used to perform adaptive training on the identification model of the abnormal telecom user, and the abnormal user identification model is optimized based on the adaptive training process.
在一个实施例中，所述获取待分析用户的业务域数据和运营域数据，具体包括：获取所述待分析用户的通讯号码，基于所述通讯号码获得所述待分析用户的业务域数据和运营域数据。In one embodiment, the acquiring the service domain data and operation domain data of the user to be analyzed specifically includes: acquiring the communication number of the user to be analyzed, and obtaining the business domain data and the data of the user to be analyzed based on the communication number Operational domain data.
第二方面，本发明还提供一种电信异常用户的处理装置，包括：In the second aspect, the present invention also provides a processing device for abnormal telecommunications users, including:
用户数据获取单元，用于获取待分析用户的业务域数据和运营域数据；a user data acquisition unit, configured to acquire business domain data and operational domain data of users to be analyzed;
人证分离用户识别单元，用于将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，得到所述待分析用户中的人证分离用户；A user identification unit for separation of identity and identification, configured to input the business domain data and the data of the operation domain into an identification model for separation of identification and identification for analysis, and obtain the identification separation user among the users to be analyzed;
异常预警级别确定单元，用于将所述人证分离用户的通讯特征数据输入到电信异常用户识别模型，得到异常用户清单；其中，所述异常用户清单包含电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别；An abnormal early warning level determination unit, configured to input the communication feature data of the user whose identity is separated into the identification model of the abnormal telecom user to obtain a list of abnormal users; wherein, the list of abnormal users includes the communication number of the abnormal telecom user and the communication number of the abnormal user The abnormal warning level corresponding to the number;
分层分级处置单元，用于根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理。The layered and hierarchical processing unit is configured to process the communication numbers of the abnormal telecommunications users according to the abnormality warning levels corresponding to the communication numbers in the abnormal user list and the preset hierarchical processing mechanism.
在一个实施例中，所述分层分级处置单元，聚义用于：In one embodiment, the hierarchical processing unit is used for:
若所述异常用户清单中所述通讯号码对应的异常预警级别为高危级别，则针对属于高危级别的电信异常用户的通讯号码进行通讯功能关停处理；和/或，If the abnormal warning level corresponding to the communication number in the abnormal user list is a high-risk level, the communication function shutdown processing is performed for the communication numbers of the abnormal telecommunications users belonging to the high-risk level; and/or,
若所述异常用户清单中的异常预警级别为中危级别，则将对应中危级别的电信异常用户的通讯号码按照预设的中级处理机制进行审核，若审核结果为不通过，则针对属于中危级别的电信异常用户的通讯号码进行通讯功能关停处理；和/或，If the abnormal warning level in the list of abnormal users is a medium-risk level, the communication numbers of the telecommunications abnormal users corresponding to the medium-risk level will be reviewed according to the preset intermediate-level processing mechanism. and/or,
若所述异常用户清单中所述人证分离用户的异常预警级别为低危级别，则将对应低危级别的电信异常用户的通讯号码按照预设的低级处理机制进行审核，若审核结果为不通过且预设通话时长内未完成二次实名认证，则针对属于低危级别的电信异常用户的通讯号码进行通讯功能关停处理。If the abnormal early warning level of the person-identity separation user mentioned in the abnormal user list is a low-risk level, the communication number of the telecommunications abnormal user corresponding to the low-risk level will be reviewed according to the preset low-level processing mechanism, and if the result of the review is not If it passes and the second real-name authentication is not completed within the preset call time, the communication function will be shut down for the communication numbers of low-risk telecom abnormal users.
在一个实施例中，所述的电信异常用户的处理装置，还包括：In one embodiment, the processing device for abnormal telecommunications users further includes:
实名认证处置单元，用于对通讯功能关停处理的电信异常用户进行二次实名认证；The real-name authentication processing unit is used to perform second real-name authentication on abnormal telecom users whose communication functions are shut down;
通讯恢复处置单元，用于若所述通讯功能关停处理的电信异常用户中存在二次实名认证结果为通过的目标电信异常用户，则将所述目标电信异常用户的通讯号码的通讯功能进行恢复。The communication restoration processing unit is used to restore the communication function of the communication number of the target abnormal telecom user if there is a target telecom abnormal user whose second real name authentication result is passed among the telecom abnormal users whose communication function is shut down. .
在一个实施例中，所述的电信异常用户的处理装置，还包括：In one embodiment, the processing device for abnormal telecommunications users further includes:
运营域位置数据获取单元，用于获取所述人证分离用户的运营域位置数据；An operation domain location data acquisition unit, configured to obtain the operation domain location data of the user whose identity and identity are separated;
通信行为数据获取单元，用于基于所述运营域位置数据按照预设的时间间隔获取所述人证分离用户在预设高危位置的高危通信行为数据；A communication behavior data acquisition unit, configured to acquire the high-risk communication behavior data of the user at a preset high-risk location based on the location data of the operation domain according to a preset time interval;
根据所述高危通信行为数据，确定所述人证分离用户中的电信异常用户及所述电信异常用户的异常预警级别；According to the high-risk communication behavior data, determine the abnormal telecommunications users among the users with identity separation and the abnormal warning level of the abnormal telecommunications users;
异常用户清单更新单元，用于基于所述电信异常用户及所述电信异常用户的异常预警级别，更新所述异常用户清单。An abnormal user list updating unit, configured to update the abnormal user list based on the abnormal telecommunications user and the abnormal warning level of the abnormal telecommunications user.
在一个实施例中，所述的电信异常用户的处理装置，还包括：模型构建单元，用于构建所述电信异常用户识别模型；In one embodiment, the processing device for abnormal telecom users further includes: a model construction unit, configured to construct the identification model for abnormal telecom users;
其中，所述构建所述电信异常用户识别模型，具体包括：Wherein, the construction of the telecom abnormal user identification model specifically includes:
获取异常用户的通讯特征数据，将所述异常用户的通讯特征数据确定为正样本集；Obtaining the communication feature data of the abnormal user, and determining the communication feature data of the abnormal user as a positive sample set;
获取非异常用户的通讯特征数据，将所述非异常用户的通讯特征数据确定为负样本集；Obtaining the communication characteristic data of the non-abnormal user, and determining the communication characteristic data of the non-abnormal user as a negative sample set;
利用预设的随机森林模型根据所述正样本集和所述负样本集构造相应决策树子模型，基于所述决策树子模型构建所述电信异常用户识别模型。A preset random forest model is used to construct a corresponding decision tree sub-model according to the positive sample set and the negative sample set, and the identification model of the abnormal telecommunications user is constructed based on the decision tree sub-model.
在一个实施例中，所述的电信异常用户的处理装置，还包括：模型优化单元，用于优化所述电信异常用户识别模型；In one embodiment, the processing device for abnormal telecom users further includes: a model optimization unit, configured to optimize the identification model for abnormal telecom users;
其中，所述优化所述电信异常用户识别模型，具体包括：Wherein, the optimization of the telecom abnormal user identification model specifically includes:
获取未通过二次实名认证的电信异常用户的通讯号码，基于所述未通过二次实名认证的电信异常用户的通讯号码构建异常号码库；Obtaining the communication numbers of the abnormal telecom users who have not passed the second real-name authentication, and constructing an abnormal number database based on the communication numbers of the abnormal telecom users who have not passed the second real-name authentication;
基于预设的粒子群优化模型及所述异常号码库，并利用径向基函数对所述电信异常用户识别模型进行自适应训练，基于自适应训练过程优化所述电信异常用户识别模型。Based on the preset particle swarm optimization model and the abnormal number library, the radial basis function is used to perform adaptive training on the identification model of the abnormal telecom user, and the abnormal user identification model is optimized based on the adaptive training process.
在一个实施例中，所述获取待分析用户的业务域数据和运营域数据，具体包括：获取所述待分析用户的通讯号码，基于所述通讯号码获得所述待分析用户的业务域数据和运营域数据。In one embodiment, the acquiring the service domain data and operation domain data of the user to be analyzed specifically includes: acquiring the communication number of the user to be analyzed, and obtaining the business domain data and the data of the user to be analyzed based on the communication number Operational domain data.
第三方面，本发明还提供一种电子设备，包括：存储器、处理器及存储在存储器上并在处理器上运行的计算机程序，所述处理器执行所述程序时实现如上述任意一项所述电信异常用户的处理方法的步骤。In a third aspect, the present invention also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and run on the processor, when the processor executes the program, it realizes any of the above-mentioned Describe the steps of the method for handling abnormal telecom users.
第四方面，本发明还提供一种处理器可读存储介质，所述处理器可读存储介质上存储有计算机程序，该计算机程序被处理器执行时实现如上述任意一项所述电信异常用户的处理方法的步骤。In the fourth aspect, the present invention also provides a processor-readable storage medium, where a computer program is stored on the processor-readable storage medium, and when the computer program is executed by the processor, it realizes the above-mentioned telecommunications abnormal user steps of the processing method.
本发明实施例提供的所述电信异常用户的处理方法，能够通过人证分离识别模型确定人证分离用户，再基于电信异常用户识别模型确定所述人证分离用户中电信异常用户的异常预警级别，并形成诈骗用户清单，提高了电信异常用户预测处理的效率和精确度。The method for processing the abnormal telecommunication users provided by the embodiment of the present invention can determine the user with the separation of witnesses and certificates through the recognition model of the separation of witnesses and identification, and then determine the abnormal warning level of the abnormal telecommunication users among the users of the separation of witnesses and certificates based on the identification model of the abnormal users of telecommunication , and form a list of fraudulent users, which improves the efficiency and accuracy of prediction and processing of abnormal telecom users.
附图说明Description of drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获取其他的附图。In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained based on these drawings without creative effort.
图1为本发明实施例提供的电信异常用户的处理方法的流程示意图；FIG. 1 is a schematic flowchart of a method for processing abnormal telecom users provided by an embodiment of the present invention;
图2为本发明实施例提供的为径向基函数神经网络结构的示意图；2 is a schematic diagram of a radial basis function neural network structure provided by an embodiment of the present invention;
图3为本发明实施例提供的电信异常用户的处理方法具体应用的架构示意图；FIG. 3 is a schematic diagram of a specific application of a method for processing abnormal telecom users provided by an embodiment of the present invention;
图4为本发明实施例提供的电信异常用户的处理装置的结构示意图；FIG. 4 is a schematic structural diagram of a processing device for abnormal telecommunications users provided by an embodiment of the present invention;
图5为本发明实施例提供的电子设备的实体结构示意图。FIG. 5 is a schematic diagram of a physical structure of an electronic device provided by an embodiment of the present invention.
具体实施方式Detailed ways
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获取的所有其他实施例，都属于本发明保护的范围。In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
下面基于本发明所述的电信异常用户的处理方法，对其实施例进行详细描述。如图1所示，其为本发明实施例提供的电信异常用户的处理方法的流程示意图，具体实现过程包括以下步骤：The following is a detailed description of the embodiment based on the method for processing abnormal telecommunication users of the present invention. As shown in Figure 1, it is a schematic flow chart of the processing method for abnormal telecom users provided by the embodiment of the present invention, and the specific implementation process includes the following steps:
步骤101：获取待分析用户的业务域数据和运营域数据。Step 101: Obtain the business domain data and operation domain data of the user to be analyzed.
在本发明实施例中，通过获取所述待分析用户的通讯号码，并基于所述通讯号码能够获得所述待分析用户的业务域数据和运营域数据。其中，所述业务域数据包括网络数据，比如信令、地图数据、告警、故障和网络资源等。所述运营域数据包括用户数据和业务数据，比如用户的消费习惯、终端信息、ARPU(Average revenue per payinguser)的分组、业务内容以及业务受众人群等。In the embodiment of the present invention, by obtaining the communication number of the user to be analyzed, and based on the communication number, the service domain data and the operation domain data of the user to be analyzed can be obtained. Wherein, the service domain data includes network data, such as signaling, map data, alarms, faults, and network resources. The operation domain data includes user data and business data, such as user consumption habits, terminal information, ARPU (Average revenue per paying user) grouping, business content, and business audience groups.
步骤102：将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，得到所述待分析用户中的人证分离用户。Step 102: Input the business domain data and the operation domain data into the identification model for separation of identity and identity for analysis, and obtain the identity separation users among the users to be analyzed.
具体的，将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，也就是，利用人证分离识别模型(即基于决策树算法实现的识别分类模型)，根据获取到的运营域数据和业务域数据中待分析用户的人证信息、同一身份证下号码数量、入网地离散度、常驻使用地以及相互通话记录信息进行人证分离识别，基于人证分离识别模型对所述待分析用户进行分类，识别号码非证件本人或其亲友使用的概率，并最终确定出所述待分析用户中的人证分离用户。Specifically, the business domain data and the operation domain data are input into the recognition model for separation of witnesses and certificates for analysis, that is, using the recognition model of separation of witnesses and certificates (that is, the recognition and classification model based on the decision tree algorithm), according to the obtained In the operation domain data and business domain data of the user to be analyzed, the witness information, the number of numbers under the same ID card, the dispersion of network access, the permanent place of use, and the information of mutual call records are used for the separation and recognition of witnesses and certificates, based on the recognition model of separation of witnesses and certificates Classify the users to be analyzed, identify the probability that the number is not used by the person himself or his relatives and friends, and finally determine the user whose identity is separated from the users to be analyzed.
其中，决策树算法是一种分类方法。在本发明实施例中，该决策树算法首先对通讯数据进行处理，利用归纳算法生成可读的规则和决策树，然后使用决策树对新通讯数据进行分析。具体的，决策树算法包括ID3决策树算法、C4.5决策树算法和CART(Classificationand Regression Tree)，在此不做具体限定。所述通讯数据可以是指用户的业务域数据和运营域数据等。Among them, the decision tree algorithm is a classification method. In the embodiment of the present invention, the decision tree algorithm firstly processes the communication data, uses the inductive algorithm to generate readable rules and decision trees, and then uses the decision tree to analyze the new communication data. Specifically, the decision tree algorithm includes ID3 decision tree algorithm, C4.5 decision tree algorithm and CART (Classification and Regression Tree), which are not specifically limited here. The communication data may refer to the user's business domain data and operation domain data.
步骤103：将所述人证分离用户的通讯特征数据输入到电信异常用户识别模型，得到异常用户清单；其中，所述异常用户清单包含电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别。Step 103: Input the communication characteristic data of the user whose identity is separated into the telecom abnormal user identification model to obtain a list of abnormal users; wherein, the list of abnormal users includes the communication number of the abnormal telecom user and the abnormal warning corresponding to the communication number level.
具体的，将人证分离用户的通讯特征进行随机组合后输入到电信异常用户识别模型，以根据通讯特征对应的属性值，将所述人证分离用户进行分类，确定所述人证分离用户的异常预警级别，并形成异常用户清单。其中，所述异常用户清单包括电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别。本发明实施例中，通过决策树算法确定人证分离用户再基于电信异常用户识别模型，确定所述人证分离用户的异常预警级别，并形成异常用户清单，提高了电信异常用户预测的精准度。其中，通讯特征包括用户的归属地、缴费金额、次均通话时长、是否合约客户、是否实名制用户、是否校园客户、高危地主叫陌生人次数、开户渠道、高危地通话次数等特征信息。单个用户的每个通讯特征对应一个固定的属性值，例如通讯特征的属性特征为A目标地区，缴费金额为100元。Specifically, the communication features of the user for identification separation are randomly combined and then input into the abnormal user identification model of telecommunications, so as to classify the user for separation of identification and identification according to the attribute value corresponding to the communication feature, and determine the identity of the user for separation of identification and identification. Abnormal warning level, and form a list of abnormal users. Wherein, the abnormal user list includes communication numbers of abnormal telecommunications users and abnormal warning levels corresponding to the communication numbers. In the embodiment of the present invention, the user whose identity is separated is determined by the decision tree algorithm, and then based on the identification model of the abnormal user in telecommunications, the abnormal warning level of the user who is separated from the identity is determined, and a list of abnormal users is formed, which improves the accuracy of the prediction of abnormal users in telecommunications . Among them, the communication characteristics include the user's location, payment amount, average call time, whether it is a contract customer, whether it is a real-name user, whether it is a campus customer, the number of calls to strangers from high-risk places, account opening channels, and the number of calls from high-risk places. Each communication feature of a single user corresponds to a fixed attribute value. For example, the attribute feature of the communication feature is A target area, and the payment amount is 100 yuan.
需要说明的是，在具体实施过程中，执行本步骤103之前，需要预先构建所述电信异常用户识别模型。所述构建所述电信异常用户识别模型，具体实现过程包括：获取异常用户的通讯特征数据，将所述异常用户的通讯特征数据确定为正样本集；获取非异常用户的通讯特征数据，将所述非异常用户的通讯特征数据确定为负样本集；利用预设的随机森林模型根据所述正样本集和所述负样本集构造相应决策树子模型，基于所述决策树子模型构建电信异常用户识别模型。It should be noted that, in the specific implementation process, before performing this
在一个构建电信异常用户识别模型的实施例中，还包括构建得到所述疑似诈骗用户识别模型，具体包括以下内容：采集异常用户的通讯特征数据作为正样本集，采集非异常用户的通讯特征数据作为负样本集；采用随机森林模型(即随机森林算法)，通过所述正样本集和负样本集生成多颗决策树；基于所述多颗决策树构建所述电信异常用户识别模型。In an embodiment of constructing a telecom abnormal user identification model, it also includes constructing the suspected fraudulent user identification model, which specifically includes the following content: collecting the communication feature data of the abnormal user as a positive sample set, and collecting the communication feature data of the non-abnormal user As a negative sample set; a random forest model (that is, a random forest algorithm) is used to generate multiple decision trees through the positive sample set and negative sample set; and the identification model of the telecom abnormal user is constructed based on the multiple decision trees.
具体地，异常用户包括实时抓取的异常用户和系统预存的异常库中的用户，采集异常用户的通讯特征作为正样本集，例如异常号码语言信息、短信信息、流量信息、位置信息、网龄信息和终端漫游指标信息；非异常用户包括已完成二次实名认证的用户和系统预存的非异常用户，采集非异常用户的通讯特征数据作为负样本集。Specifically, abnormal users include abnormal users captured in real time and users in the abnormal database pre-stored by the system, and the communication characteristics of abnormal users are collected as a positive sample set, such as language information of abnormal numbers, SMS information, traffic information, location information, and network age. Information and terminal roaming index information; non-abnormal users include users who have completed the second real-name authentication and non-abnormal users pre-stored in the system, and collect communication characteristic data of non-abnormal users as a negative sample set.
其中，随机森林算法是一种基于神经网络的机器学习模型。在本发明实施例中，随机森林算法根据正样本数据和负样本数据能够构造M棵决策树，每棵决策树用来标记一类高频异常号码类，每棵决策树随机从训练样本集N中有放回地重复随机抽取k个样本生成新的训练样本集合，同时每个样本的抽取g个特征值，新数据的分类结果按决策树投票多少形成分数而定，再根据数据分类的好坏筛选出具有最好特征值的集合。随机森林算法实质是对决策树算法的一种改进，将多个决策树合并在一起，每棵决策树的建立依赖于一个独立抽取的样品，随机森林中的每棵决策树具有相同的分布，分类误差取决于每一棵决策树的分类能力和它们之间的相关性。特征选择采用随机的方法去分裂每一个节点，然后比较不同情况下产生的误差。能够检测到的内在估计误差、分类能力和相关性决定选择哪些有价值的特征值。单棵决策树的分类能力可能很小，但在随机产生大量的决策树后，一个测试样品可以通过每一棵决策树的分类结果经统计后选择最可能的分类与最有价值的特征值。Among them, random forest algorithm is a machine learning model based on neural network. In the embodiment of the present invention, the random forest algorithm can construct M decision trees according to the positive sample data and the negative sample data, each decision tree is used to mark a class of high-frequency abnormal numbers, and each decision tree is randomly selected from the training sample set N Repeatedly randomly select k samples with replacement to generate a new training sample set, and at the same time extract g eigenvalues for each sample, the classification result of the new data depends on the number of decision tree votes to form a score, and then according to the goodness of the data classification Bad filters out the set with the best eigenvalues. The essence of the random forest algorithm is an improvement to the decision tree algorithm. Multiple decision trees are merged together. The establishment of each decision tree depends on an independently drawn sample. Each decision tree in the random forest has the same distribution. The classification error depends on the classification ability of each decision tree and the correlation between them. Feature selection uses a random method to split each node, and then compares the errors generated in different situations. The intrinsic estimation errors that can be detected, classification power and correlation determine which valuable feature values are selected. The classification ability of a single decision tree may be small, but after a large number of decision trees are randomly generated, a test sample can select the most probable classification and the most valuable feature value through the statistics of the classification results of each decision tree.
决策树构建的关键在于分割点的选取，通过采用贪心算法考虑当前分割点纯度差的大小作为要素进行从大到小优先排序。对于纯度的量化使用ID3算法，以信息增益度量属性选择，选择分裂后信息增益最大的属性进行分裂。计算集合的信息熵的公式如下：The key to constructing the decision tree is the selection of the split point, and the greedy algorithm is used to consider the size of the purity difference of the current split point as an element to prioritize from large to small. The ID3 algorithm is used for the quantification of purity, and the attribute selection is measured by information gain, and the attribute with the largest information gain after splitting is selected for splitting. The formula for calculating the information entropy of a set is as follows:
其中，info(D)为集合D的信息熵，Pi为第i类别在集合D中出现的概率。Among them, info(D) is the information entropy of the set D, and Pi is the probability that the i-th category appears in the set D.
对集合按照特征属性划分后的期望信息熵计算公式如下：The formula for calculating the expected information entropy after dividing the set according to the characteristic attributes is as follows:
其中，infoA(D)表示A对D所划分的期望信息熵，D为训练集合，A为特征属性。Among them, infoA (D) represents the expected information entropy divided by A to D, D is the training set, and A is the feature attribute.
对集合按照特征属性划分后的信息熵增益计算公式如下：The formula for calculating the information entropy gain after dividing the set according to the characteristic attributes is as follows:
gain(A)＝info(D)-infoA(D)gain(A)＝info(D)-infoA (D)
其中，gain(A)为按照A特征属性划分后所得到的信息增益，info(D)为集合D的信息熵，infoA(D)表示A对D所划分的期望信息熵。Among them, gain(A) is the information gain obtained after dividing according to the characteristic attributes of A, info(D) is the information entropy of the set D, and infoA (D) indicates the expected information entropy of A divided by D.
所有的特征值都按照信息增益来进行递归排序，从而构建整个决策树，在随机森林体系所构建的决策树中不需要进行减枝，这样对训练数据就会表现很精确，尽管对其他数据没有那么精确会出现过拟合，但对于集成学习来说可以通过多个决策树共同决策来避免单个决策树的过拟合。具体地，可选取多个训练集样本，根据训练集样本及其特征值，建立决策树，重复该步骤，建立多颗决策树。All the eigenvalues are recursively sorted according to the information gain, so as to construct the entire decision tree. In the decision tree constructed by the random forest system, there is no need to cut branches, so that the training data will be very accurate, although there is no other data. Then accurate overfitting will occur, but for integrated learning, multiple decision trees can make joint decisions to avoid overfitting of a single decision tree. Specifically, a plurality of training set samples can be selected, and a decision tree is established according to the training set samples and their feature values, and this step is repeated to establish multiple decision trees.
具体地，对于训练集数据，每颗决策树会经过决策，确定用户的异常预警等级，对每个决策树对于分类结果进行评估，筛选出部分特征类型集合，构建所述电信异常用户识别模型。Specifically, for the training set data, each decision tree will go through decision-making to determine the user's abnormal warning level, evaluate the classification results of each decision tree, and filter out some feature type sets to build the telecom abnormal user identification model.
本发明实施例，通过随机森林算法，构建得到电信异常用户识别模型，基于随机树算法确定人证分离用户再基于构建得到的电信异常用户识别模型，确定所述人证分离用户的异常预警级别，并形成异常用户清单，提高了电信异常用户预测的精准度。In the embodiment of the present invention, the identification model of the abnormal user in telecommunications is constructed through the random forest algorithm, and the user with identification and identification is determined based on the random tree algorithm, and then based on the identification model for the abnormal user with identification constructed, the abnormal warning level of the user with identification and identification is determined. And a list of abnormal users is formed, which improves the accuracy of prediction of abnormal telecom users.
进一步的，在具体实施过程中，还可优化所述电信异常用户识别模型。具体的，所述优化所述电信异常用户识别模型，实现过程包括：获取未通过二次实名认证的电信异常用户的通讯号码，基于所述未通过二次实名认证的电信异常用户的通讯号码构建异常号码库；基于预设的粒子群优化模型及所述异常号码库，并利用径向基函数对所述电信异常用户识别模型进行自适应训练，基于自适应训练过程优化所述电信异常用户识别模型。Further, in the specific implementation process, the abnormal telecom user identification model can also be optimized. Specifically, the implementation process of optimizing the identification model of the abnormal telecom user includes: obtaining the communication number of the abnormal telecom user who has not passed the second real-name authentication, and constructing the communication number based on the communication number of the abnormal telecom user who has not passed the second real-name authentication Abnormal number library; based on the preset particle swarm optimization model and the abnormal number library, and using radial basis function to perform adaptive training on the abnormal telecom user identification model, and optimize the abnormal telecom user identification based on the adaptive training process Model.
在一个优化所述电信异常用户识别模型的实施例中，具体包括以下内容：获取未通过所述二次实名认证的用户号码，构建异常号码库；基于粒子群优化算法及所述异常号码库，通过径向基函数实现所述疑似异常用户识别模型的自适应训练，通过所述自适应训练过程优化所述疑似异常用户识别模型。In an embodiment of optimizing the abnormal user identification model of telecommunications, it specifically includes the following content: obtain user numbers that have not passed the second real-name authentication, and construct an abnormal number database; based on the particle swarm optimization algorithm and the abnormal number database, The adaptive training of the suspected abnormal user identification model is realized through the radial basis function, and the suspected abnormal user identification model is optimized through the adaptive training process.
具体地，为了更好的提升疑似异常用户识别模型的准确度，实现特征变量、参数数量和参数权重系数的智能优化，本实施例将系统识别出来的高危级别、中危级别和低危级别号码在经过二次实名认证后，将认证结论、异常通讯号码的基础属性、漫游属性、行为属性、位置轨迹等历史数据结合，将未通过实名认证的号码传至异常号码库，构建异常号码库。Specifically, in order to better improve the accuracy of the suspected abnormal user identification model and realize the intelligent optimization of characteristic variables, parameter numbers, and parameter weight coefficients, this embodiment will identify the high-risk level, medium-risk level, and low-risk level numbers identified by the system After the second real-name authentication, the authentication conclusion, the basic attributes of abnormal communication numbers, roaming attributes, behavior attributes, location tracks and other historical data are combined, and the numbers that have not passed the real-name authentication are transferred to the abnormal number database to build an abnormal number database.
图2为径向基函数神经网络结构的示意图。该径向基函数神经网络从左到右依次包含的是输入层、隐层和输出层三层结构。从输入层到隐层的变换是非线性变换，输出层是输出神经元的线性加权组合。Fig. 2 is a schematic diagram of a radial basis function neural network structure. The radial basis function neural network consists of three layers of input layer, hidden layer and output layer from left to right. The transformation from the input layer to the hidden layer is a nonlinear transformation, and the output layer is a linear weighted combination of output neurons.
当选择高斯函数作为径向基函数时，径向基函数神经网络的输出公式如下：When the Gaussian function is selected as the radial basis function, the output formula of the radial basis function neural network is as follows:
其中，
算法的主要任务就是估计出径向基函数神经网络中三个参数wij、μi和σi。估计参数的方法是给定一组用于训练的输入输出数据对，通过调整参数wij、μi和σi使J的值最小，得到参数wij、μi和σi。J的计算公式如下：The main task of the algorithm is to estimate the three parameters wij , μi and σi in the radial basis function neural network. The method of estimating the parameters is given a set of input-output data pairs for training, and the parameters wij , μi and σi are obtained by adjusting the parameters wij , μi and σi to minimize the value of J. The calculation formula of J is as follows:
其中，y(k)为输入输出数据对中的输出数据，
当成功地训练出径向基函数神经网络后，就得到了径向基函数神经网络模型中未知的参数wij、μi和σi，利用此公式可由输入向量x(t)得出预测的输出结果
利用此算法，使用已识别出的疑似异常客户的基础属性、漫游属性、行为属性、位置轨迹等历史数据，能够精确地预测客户的通讯行为，以此作为疑似异常用户的识别模型的计算数据，提高识别精确度。Using this algorithm, using historical data such as basic attributes, roaming attributes, behavior attributes, and location trajectories of identified suspected abnormal customers, it is possible to accurately predict the communication behavior of customers, and use this as the calculation data for the identification model of suspected abnormal users. Improve recognition accuracy.
本发明实施例，基于粒子群优化算法及异常号码库，通过径向基函数实现所述疑似异常用户识别模型的自适应训练优化所述疑似异常用户识别模型，优化后的模型识别精准度高，提高了电信异常用户预测的精准度。In the embodiment of the present invention, based on the particle swarm optimization algorithm and the abnormal number library, the self-adaptive training of the suspected abnormal user identification model is realized through the radial basis function to optimize the suspected abnormal user identification model, and the optimized model has high identification accuracy. Improve the accuracy of forecasting abnormal telecom users.
步骤104：根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理。Step 104: Process the communication numbers of the abnormal telecommunications users according to the abnormal warning levels corresponding to the communication numbers in the abnormal user list and the preset hierarchical processing mechanism.
在本发明实施例中，若所述异常用户清单中所述通讯号码对应的异常预警级别为高危级别，则针对属于高危级别的电信异常用户的通讯号码进行通讯功能关停处理。具体地，对于已确定异常等级的异常用户，根据异常用户对应的异常预警级别进行相应的处理，若所述异常用户清单中的用户的异常预警级别为高危级别，则对所述人证分离用户的通讯号码进行通讯功能关停处理。关停后的通讯号码在恢复通讯功能之前无法进行通讯。In the embodiment of the present invention, if the abnormal warning level corresponding to the communication number in the abnormal user list is a high-risk level, the communication function shutdown processing is performed for the communication numbers of the abnormal telecommunications users belonging to the high-risk level. Specifically, for abnormal users whose abnormal levels have been determined, corresponding processing is performed according to the abnormal warning levels corresponding to the abnormal users. The communication number of the communication function is shut down. The communication number after shutting down cannot communicate until the communication function is restored.
若所述异常用户清单中的异常预警级别为中危级别，则将对应中危级别的电信异常用户的通讯号码按照预设的中级处理机制进行审核，若审核结果为不通过，则针对属于中危级别的电信异常用户的通讯号码进行通讯功能关停处理。比如，若所述异常用户清单中异常用户的异常预警级别为中危级别，则将所述中危级别异常用户的通讯号码发送至预审核模块中按照预设的中级处理机制进行审核，若接收到审核结果为通过，则不对该异常用户的通讯号码进行关停处理，若接收到审核结果为不通过，则对所述人证分离用户的通讯号码进行通讯功能关停处理。If the abnormal warning level in the list of abnormal users is a medium-risk level, the communication numbers of the telecommunications abnormal users corresponding to the medium-risk level will be reviewed according to the preset intermediate-level processing mechanism. Shut down the communication function of the communication number of the abnormal telecommunications user at the dangerous level. For example, if the abnormal warning level of the abnormal user in the abnormal user list is the medium-risk level, the communication number of the abnormal user with the medium-risk level is sent to the pre-audit module for review according to the preset intermediate-level processing mechanism. If the verification result is passed, the communication number of the abnormal user will not be shut down, and if the verification result is not passed, the communication function of the communication number of the user whose identity and identity are separated will be shut down.
若所述异常用户清单中所述人证分离用户的异常预警级别为低危级别，则将对应低危级别的电信异常用户的通讯号码按照预设的低级处理机制进行审核，若审核结果为不通过且预设通话时长内未完成二次实名认证，则针对属于低危级别的电信异常用户的通讯号码进行通讯功能关停处理。具体地，若所述异常用户清单中异常用户的异常预警级别为低危级别，则将所述中低级别用户的通讯号码发送至预审核模块按照预设的低级处理机制进行审核，若接收到审核结果为通过，则不对该异常用户的通讯号码进行关停处理，若接收到人工审核结果为不通过，则在异常用户在预设通话时长内未完成二次实名认证的情况下，对所述人证分离用户的通讯号码进行通讯功能关停处理。优选地，预设通话时长为3小时，用户可在3小时内完成二次实名认证，避免号码的通讯功能关停处理。其中，所述二次实名认证为对通讯功能关停处理的电信异常用户进行二次实名认证；若通讯功能关停处理的电信异常用户中存在二次实名认证结果为通过的目标电信异常用户，则将目标电信异常用户的通讯号码的通讯功能进行恢复。If the abnormal early warning level of the person-identity separation user mentioned in the abnormal user list is a low-risk level, the communication number of the telecommunications abnormal user corresponding to the low-risk level will be reviewed according to the preset low-level processing mechanism, and if the result of the review is not If it passes and the second real-name authentication is not completed within the preset call time, the communication function will be shut down for the communication numbers of low-risk telecom abnormal users. Specifically, if the abnormal warning level of the abnormal user in the abnormal user list is a low-risk level, the communication number of the low-medium level user is sent to the pre-audit module for review according to the preset low-level processing mechanism. If the audit result is passed, the communication number of the abnormal user will not be closed. Shut down the communication function of the communication number of the above-mentioned witness separation user. Preferably, the preset call duration is 3 hours, and the user can complete the second real-name authentication within 3 hours, so as to avoid shutting down the communication function of the number. Wherein, the second real-name authentication is to carry out second real-name authentication on abnormal telecom users whose communication functions are shut down; Then restore the communication function of the communication number of the target telecom abnormal user.
在本发明实施例中，通过随机树算法确定人证分离用户再基于电信异常用户识别模型，确定所述人证分离用户的异常预警级别，并形成异常用户清单，提高了电信异常用户(比如诈骗用户)预测的精准度。对于高危、中危和低危的人证分离用户分别采取不同的处理方式，有效的控制了异常通讯行为，对低危级别的人证分离用户设置预设通话时长，减小对用户的影响，提升用户体验感。In the embodiment of the present invention, the user whose identity is separated is determined by the random tree algorithm, and then based on the identification model of the abnormal user in telecommunications, the abnormal early warning level of the user who is separated from the identity is determined, and a list of abnormal users is formed, which improves the number of abnormal users of telecommunications (such as fraudulent users). user) prediction accuracy. Different processing methods are adopted for high-risk, medium-risk and low-risk users of identity separation, which effectively control abnormal communication behaviors. For low-risk users of identity separation, preset call durations are set to reduce the impact on users. Improve user experience.
在具体实施过程中，还包括若判断已进行关停处理的所述异常用户清单中的人证分离用户完成所述二次实名认证，则恢复所述人证分离用户的通讯号码的通讯功能。其中，已进行关停处理的人证分离用户包括，已进行通讯功能关停处理的高危用户、中危用户和低危用户。已进行关停处理的人证分离用户可在用户端进行二次实名认证。具体地，获取用户在用户端输入的通讯号码(比如手机号)，通过短信验证码登录，登录成功后，通过密码服务或近期通话记录进行补登记。登记成功后，获取用户上传身份证图片，系统通过光学字符识别OCR(Optical Character Recognition，光学字符识别)技术，将用户上传的身份证照片与预存的身份信息库进行对比，确保用户身份信息的真实性。核实用户的身份信息后，系统开启摄像头及录音功能，生成提示信息提示用户朗读预设的数字信息，后台通过声音识别用户朗读数字是否与系统数字一致，是否为本人朗读，同时辅以人像比对技术、静默活体技术判断拍摄视频是否本人且是否为真人活体拍摄。若确定朗读正确且是本人活体拍摄，则二次实名认证通过，将用户的身份信息传至后台系统，后台系统根据用户上传的身份信息，调用后台接口，为用户完成信息补录，同时，恢复该用户的号码的通讯功能。通过提供二次实名认证的方法，对误判的正常客户需要提供快速恢复功能，减小用户对的不友好感知，减小对正常客户的影响。In the specific implementation process, if it is judged that the identity separation user in the abnormal user list that has been shut down has completed the second real name authentication, restoring the communication function of the communication number of the identity separation user. Among them, the users whose identity separation has been shut down include high-risk users, medium-risk users, and low-risk users whose communication functions have been shut down. Users who have been shut down and whose identity is separated can perform a second real-name authentication on the user terminal. Specifically, obtain the communication number (such as a mobile phone number) entered by the user at the user end, and log in through the SMS verification code. After the login is successful, perform supplementary registration through the password service or recent call records. After the registration is successful, the ID card picture uploaded by the user is obtained, and the system uses OCR (Optical Character Recognition, Optical Character Recognition) technology to compare the ID card photo uploaded by the user with the pre-stored identity information database to ensure the authenticity of the user's identity information sex. After verifying the user's identity information, the system turns on the camera and recording function, generates a prompt message to remind the user to read the preset digital information, and the background uses voice to identify whether the user's reading number is consistent with the system number, whether it is the person's reading, and supplemented by portrait comparison Technology, silent living technology to judge whether the video is taken by the person and whether it is a real person. If it is determined that the reading is correct and the photo was taken by the person himself, the second real-name authentication will pass, and the user’s identity information will be transmitted to the background system. Communication features for this user's number. By providing a second real-name authentication method, it is necessary to provide a quick recovery function for normal customers who have been misjudged, so as to reduce the unfriendly perception of users and reduce the impact on normal customers.
除此之外，在具体实施过程中，还包括：获取所述人证分离用户的运营域位置数据；基于所述运营域位置数据按照预设的时间间隔获取所述人证分离用户在预设高危位置的高危通信行为数据；根据所述高危通信行为数据，确定所述人证分离用户中的电信异常用户及所述电信异常用户的异常预警级别；基于所述电信异常用户及所述电信异常用户的异常预警级别，更新所述异常用户清单。In addition, in the specific implementation process, it also includes: obtaining the operation domain location data of the user for separation of witnesses and certificates; High-risk communication behavior data in high-risk locations; according to the high-risk communication behavior data, determine the abnormal telecommunications users among the users with identity separation and the abnormal warning level of the abnormal telecommunications users; based on the abnormal telecommunications users and the abnormal telecommunications The abnormal warning level of the user is used to update the list of abnormal users.
具体地，可跨系统采集A口信令、地图数据等运营域位置数据，通过运营域位置数据实时跟踪漫游出访到预设的高危位置的高危通信行为数据，预设的高危位置可以根据实际情况进行设置，在此不做具体限定。根据所述高危通信行为数据及其相关用户的已知属性信息，例如是否是星级用户、在网时长和呼叫号码的离散程度等，通过流式计算方式每隔预设的时间间隔确定一次疑似异常用户及所述疑似异常用户的异常预警级别。流式计算的方式相较于离线计算具有更高的实时性，流式计算区别于实时计算又存在一定的时延。优选地，本实施例中每隔1小时更新一次异常用户的异常预警级别。根据基于流式计算确定的异常用户及其对应的异常预警级别更新异常用户清单。基于流式计算对用户的高危通信行为进行计算，对所述异常用户清单进行更新，提高了电信异常用户预测的精准度和实时性。Specifically, the operation domain location data such as port A signaling and map data can be collected across systems, and the high-risk communication behavior data of roaming and visiting preset high-risk locations can be tracked in real time through the operation domain location data. The preset high-risk locations can be based on actual conditions. It is not specifically limited here. According to the high-risk communication behavior data and the known attribute information of related users, such as whether they are star users, the length of online time, and the degree of dispersion of call numbers, etc., the suspected Abnormal users and the abnormal warning level of the suspected abnormal users. Compared with offline computing, stream computing has higher real-time performance, but stream computing differs from real-time computing in that there is a certain delay. Preferably, in this embodiment, the abnormal warning level of abnormal users is updated every hour. The list of abnormal users is updated according to the abnormal users determined based on streaming computing and their corresponding abnormal warning levels. The high-risk communication behavior of users is calculated based on stream computing, and the list of abnormal users is updated, which improves the accuracy and real-time performance of prediction of abnormal users in telecommunications.
图3为本发明实施例提供的电信异常用户的处理方法具体应用的架构示意图。在一个实际实施例中，包括人证分离识别器、疑似诈骗识别模块、分层分级模块、诈骗号码库、疑似涉诈号码识别优化器、实名认证处理器和涉诈号码通讯恢复处置器。该方法包括如下步骤：FIG. 3 is a schematic diagram of a specific application of the method for processing abnormal telecommunications users provided by an embodiment of the present invention. In an actual embodiment, it includes a witness separation recognizer, a suspected fraud identification module, a hierarchical classification module, a fraud number database, a suspected fraud number identification optimizer, a real name authentication processor and a fraud number communication recovery processor. The method comprises the steps of:
步骤300：获取客户信息输入接口输入的待分析用户的通讯号码。Step 300: Obtain the communication number of the user to be analyzed inputted by the customer information input interface.
步骤301：将相应的客户特征进行分析抽取，然后将客户信息分别输入人证分离识别器，以及疑似诈骗识别模块。Step 301: Analyzing and extracting the corresponding customer features, and then inputting the customer information into the identity separation recognizer and the suspected fraud recognition module.
步骤302：通过人证分离识别器进行实名实人分离号码识别，将识别出的高危疑似分离通讯号码作为一个重要因子传至疑似诈骗识别模块，将中低危疑似分离号码直接跳转至实名认证处置器。Step 302: Perform real-name and real-person separation number identification through the identity separation recognizer, pass the identified high-risk suspected separation communication number as an important factor to the suspected fraud identification module, and directly jump to the real-name authentication for low- and medium-risk suspected separation numbers disposer.
步骤303：疑似诈骗识别模块中含疑似涉诈号码识别器以及准实时涉诈号码识别器，在疑似诈骗识别模块从输入客户中识别出疑似涉诈号码，分层分级输出至分层分级处置模块。Step 303: The suspected fraud identification module includes a suspected fraud-related number recognizer and a quasi-real-time fraud-related number recognizer. The suspected fraud-related number is identified from the input customers in the suspected fraud identification module, and the hierarchical output is sent to the hierarchical and hierarchical disposal module. .
步骤304：高危涉诈号码关停处置器对识别出的高危涉诈号码进行主叫、发短信和上网功能自动关停；中危涉诈号码关停处置器先是对业务人员提供中危号码审核功能，若审核为立即关停，处置器将对号码的进行主叫、发短信和上网功能自动关停；低危涉诈号码关停处置器，对业务人员提供中危号码审核功能，审核为涉诈号码的，系统提供3小时通话时间，3小时通话时间内，未进行二次认证或认证不通过的处置器将对号码的进行主叫、发短信和上网功能自动关停。Step 304: The high-risk fraudulent number shutdown handler automatically shuts down the functions of calling, sending text messages and surfing the Internet for the identified high-risk fraudulent numbers; the medium-risk fraudulent number shutdown handler first provides business personnel with medium-risk numbers function, if the audit is immediate shut down, the processor will automatically shut down the number’s functions of calling, sending text messages and surfing the Internet; shut down the processor for low-risk fraudulent numbers, and provide business personnel with the function of auditing medium-risk numbers. For fraudulent numbers, the system provides 3 hours of talk time. During the 3 hours of talk time, processors that have not passed the secondary authentication or have not passed the authentication will automatically shut down the functions of calling, sending text messages and surfing the Internet.
步骤305：分层分级处置模块对涉诈号码进行处置后，将号码提交至实名认证处置器，由用户进行实名认证信息的提交，再由系统判断号码是否通过实名认证。Step 305: After processing the fraudulent number by the hierarchical and grading processing module, the number is submitted to the real-name authentication handler, and the user submits the real-name authentication information, and then the system judges whether the number has passed the real-name authentication.
步骤306：将通过实名认证的号码传至涉诈号码通讯恢复处置器，对被暂停主叫、发短信和上网功能的号码恢复主叫、发短信和上网功能；未通过实名认证的号码传至诈骗号码库，诈骗号码库中的号码传至疑似涉诈号码识别优化器进行号码属性、行为分析，优化疑似涉诈号码识别模型。Step 306: Pass the number that has passed the real-name authentication to the communication recovery processor for the fraudulent number, and restore the functions of calling, sending text messages, and surfing the Internet to the number whose functions of calling, sending text messages, and surfing the Internet are suspended; Fraud number database, the numbers in the fraud number database are sent to the suspected fraudulent number identification optimizer for number attribute and behavior analysis, and the suspected fraudulent number identification model is optimized.
步骤307：将实名认证处置器中的结果向用户进行展示。Step 307: Display the result in the real-name authentication handler to the user.
本发明实施例，使用基于实名认证预测疑似电信诈骗用户方法及装置，实现了“信令采集-信令传输-涉诈号码检出-涉诈号码关停-便携二次实名认证-快速自动开通”的闭环流程，全程系统自动处理。构建了完备的异常监控分析、预警处置流程，将防范关口前移，及时应对苗头性、趋势性问题，异常号码识别精度高、关停处理效率高，有效遏制涉诈举报通报上涨趋势，同时降低误判号码的客户投诉。In the embodiment of the present invention, using the method and device for predicting suspected telecommunications fraud users based on real-name authentication, "signaling collection-signaling transmission-fraud-related number detection-fraud-related number shutdown-portable secondary real-name authentication-fast automatic activation "Closed-loop process, the whole process is automatically processed by the system. Constructed a complete abnormal monitoring analysis, early warning and handling process, will prevent the gate from moving forward, and respond to emerging and trending problems in a timely manner. The identification accuracy of abnormal numbers is high, and the shutdown processing efficiency is high, effectively curbing the upward trend of fraud reports and notifications, while reducing Customer complaints about misjudged numbers.
综上所述，本发明实施例提供的所述电信异常用户的处理方法，能够通过人证分离识别模型确定人证分离用户，再基于电信异常用户识别模型确定所述人证分离用户中电信异常用户的异常预警级别，并形成诈骗用户清单，提高了电信异常用户预测处理的效率和精确度。To sum up, the method for processing the abnormal telecommunication users provided by the embodiment of the present invention can determine the users whose witnesses and certificates are separated through the recognition model of the separation of witnesses and certificates, and then determine the abnormal telecommunications among the users of the separation of witnesses and certificates based on the identification model of the abnormal users of telecommunications. The user's abnormal warning level and form a list of fraudulent users, which improves the efficiency and accuracy of telecom abnormal user prediction and processing.
与上述提供的一种电信异常用户的处理方法相对应，本发明还提供一种电信异常用户的处理装置。由于该装置的实施例相似于上述方法实施例，所以描述得比较简单，相关之处请参见上述方法实施例部分的说明即可，下面描述的电信异常用户的处理装置的实施例仅是示意性的。Corresponding to the method for processing abnormal telecommunication users provided above, the present invention also provides a device for processing abnormal telecommunication users. Since the embodiment of the device is similar to the above-mentioned method embodiment, the description is relatively simple. Please refer to the description of the above-mentioned method embodiment for relevant information. The embodiment of the processing device for the abnormal telecom user described below is only illustrative of.
请参考图4所示，其为本发明实施例提供的一种电信异常用户的处理装置的结构示意图。Please refer to FIG. 4 , which is a schematic structural diagram of an apparatus for processing abnormal telecommunication users provided by an embodiment of the present invention.
本发明所述的电信异常用户的处理装置包括如下部分：The processing device for abnormal users of telecommunication according to the present invention includes the following parts:
用户数据获取单元401，用于获取待分析用户的业务域数据和运营域数据；A user
人证分离用户识别单元402，用于将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，得到所述待分析用户中的人证分离用户；The identity separation
异常预警级别确定单元403，用于将所述人证分离用户的通讯特征数据输入到电信异常用户识别模型，得到异常用户清单；其中，所述异常用户清单包含电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别；Abnormal early warning
分层分级处置单元404，用于根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理。The hierarchical and
在一个实施例中，所述分层分级处置单元，聚义用于：In one embodiment, the hierarchical processing unit is used for:
若所述异常用户清单中所述通讯号码对应的异常预警级别为高危级别，则针对属于高危级别的电信异常用户的通讯号码进行通讯功能关停处理；和/或，If the abnormal warning level corresponding to the communication number in the abnormal user list is a high-risk level, the communication function shutdown processing is performed for the communication numbers of the abnormal telecommunications users belonging to the high-risk level; and/or,
若所述异常用户清单中的异常预警级别为中危级别，则将对应中危级别的电信异常用户的通讯号码按照预设的中级处理机制进行审核，若审核结果为不通过，则针对属于中危级别的电信异常用户的通讯号码进行通讯功能关停处理；和/或，If the abnormal warning level in the list of abnormal users is a medium-risk level, the communication numbers of the telecommunications abnormal users corresponding to the medium-risk level will be reviewed according to the preset intermediate-level processing mechanism. and/or,
若所述异常用户清单中所述人证分离用户的异常预警级别为低危级别，则将对应低危级别的电信异常用户的通讯号码按照预设的低级处理机制进行审核，若审核结果为不通过且预设通话时长内未完成二次实名认证，则针对属于低危级别的电信异常用户的通讯号码进行通讯功能关停处理。If the abnormal early warning level of the person-identity separation user mentioned in the abnormal user list is a low-risk level, the communication number of the telecommunications abnormal user corresponding to the low-risk level will be reviewed according to the preset low-level processing mechanism, and if the result of the review is not If it passes and the second real-name authentication is not completed within the preset call time, the communication function will be shut down for the communication numbers of low-risk telecom abnormal users.
在一个实施例中，所述的电信异常用户的处理装置，还包括：In one embodiment, the processing device for abnormal telecommunications users further includes:
实名认证处置单元，用于对通讯功能关停处理的电信异常用户进行二次实名认证；The real-name authentication processing unit is used to perform second real-name authentication on abnormal telecom users whose communication functions are shut down;
通讯恢复处置单元，用于若所述通讯功能关停处理的电信异常用户中存在二次实名认证结果为通过的目标电信异常用户，则将所述目标电信异常用户的通讯号码的通讯功能进行恢复。The communication restoration processing unit is used to restore the communication function of the communication number of the target abnormal telecom user if there is a target telecom abnormal user whose second real name authentication result is passed among the telecom abnormal users whose communication function is shut down. .
在一个实施例中，所述的电信异常用户的处理装置，还包括：In one embodiment, the processing device for abnormal telecommunications users further includes:
运营域位置数据获取单元，用于获取所述人证分离用户的运营域位置数据；An operation domain location data acquisition unit, configured to obtain the operation domain location data of the user whose identity and identity are separated;
通信行为数据获取单元，用于基于所述运营域位置数据按照预设的时间间隔获取所述人证分离用户在预设高危位置的高危通信行为数据；A communication behavior data acquisition unit, configured to acquire the high-risk communication behavior data of the user at a preset high-risk location based on the location data of the operation domain according to a preset time interval;
根据所述高危通信行为数据，确定所述人证分离用户中的电信异常用户及所述电信异常用户的异常预警级别；According to the high-risk communication behavior data, determine the abnormal telecommunications users among the users with identity separation and the abnormal warning level of the abnormal telecommunications users;
异常用户清单更新单元，用于基于所述电信异常用户及所述电信异常用户的异常预警级别，更新所述异常用户清单。An abnormal user list updating unit, configured to update the abnormal user list based on the abnormal telecommunications user and the abnormal warning level of the abnormal telecommunications user.
在一个实施例中，所述的电信异常用户的处理装置，还包括：模型构建单元，用于构建所述电信异常用户识别模型；In one embodiment, the processing device for abnormal telecom users further includes: a model construction unit, configured to construct the identification model for abnormal telecom users;
其中，所述构建所述电信异常用户识别模型，具体包括：Wherein, the construction of the telecom abnormal user identification model specifically includes:
获取异常用户的通讯特征数据，将所述异常用户的通讯特征数据确定为正样本集；Obtaining the communication feature data of the abnormal user, and determining the communication feature data of the abnormal user as a positive sample set;
获取非异常用户的通讯特征数据，将所述非异常用户的通讯特征数据确定为负样本集；Obtaining the communication characteristic data of the non-abnormal user, and determining the communication characteristic data of the non-abnormal user as a negative sample set;
利用预设的随机森林模型根据所述正样本集和所述负样本集构造相应决策树子模型，基于所述决策树子模型构建所述电信异常用户识别模型。A preset random forest model is used to construct a corresponding decision tree sub-model according to the positive sample set and the negative sample set, and the identification model of the abnormal telecommunications user is constructed based on the decision tree sub-model.
在一个实施例中，所述的电信异常用户的处理装置，还包括：模型优化单元，用于优化所述电信异常用户识别模型；In one embodiment, the processing device for abnormal telecom users further includes: a model optimization unit, configured to optimize the identification model for abnormal telecom users;
其中，所述优化所述电信异常用户识别模型，具体包括：Wherein, the optimization of the telecom abnormal user identification model specifically includes:
获取未通过二次实名认证的电信异常用户的通讯号码，基于所述未通过二次实名认证的电信异常用户的通讯号码构建异常号码库；Obtaining the communication numbers of the abnormal telecom users who have not passed the second real-name authentication, and constructing an abnormal number database based on the communication numbers of the abnormal telecom users who have not passed the second real-name authentication;
基于预设的粒子群优化模型及所述异常号码库，并利用径向基函数对所述电信异常用户识别模型进行自适应训练，基于自适应训练过程优化所述电信异常用户识别模型。Based on the preset particle swarm optimization model and the abnormal number library, the radial basis function is used to perform adaptive training on the identification model of the abnormal telecom user, and the abnormal user identification model is optimized based on the adaptive training process.
在一个实施例中，所述获取待分析用户的业务域数据和运营域数据，具体包括：获取所述待分析用户的通讯号码，基于所述通讯号码获得所述待分析用户的业务域数据和运营域数据。In one embodiment, the acquiring the service domain data and operation domain data of the user to be analyzed specifically includes: acquiring the communication number of the user to be analyzed, and obtaining the business domain data and the data of the user to be analyzed based on the communication number Operational domain data.
本发明实施例提供的所述电信异常用户的处理装置，能够通过人证分离识别模型确定人证分离用户，再基于电信异常用户识别模型确定所述人证分离用户中电信异常用户的异常预警级别，并形成诈骗用户清单，提高了电信异常用户预测处理的效率和精确度。The device for processing abnormal telecommunications users provided by the embodiments of the present invention can determine the identity separation users through the identity separation identification model, and then determine the abnormal early warning level of the telecommunications abnormal users among the identity separation users based on the identification model of the telecommunications abnormal users , and form a list of fraudulent users, which improves the efficiency and accuracy of prediction and processing of abnormal telecom users.
与上述提供的电信异常用户的处理方法相对应，本发明还提供一种电子设备。由于该电子设备的实施例相似于上述方法实施例，所以描述得比较简单，相关之处请参见上述方法实施例部分的说明即可，下面描述的电子设备仅是示意性的。如图5所示，其为本发明实施例公开的一种电子设备的实体结构示意图。该电子设备可以包括：处理器(processor)501、存储器(memory)502和通信总线503，其中，处理器501，存储器502通过通信总线503完成相互间的通信，通过通信接口504与外部进行通信。处理器501可以调用存储器502中的逻辑指令，以执行电信异常用户的处理方法。该方法包括：获取待分析用户的业务域数据和运营域数据；将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，得到所述待分析用户中的人证分离用户；将所述人证分离用户的通讯特征数据输入到电信异常用户识别模型，得到异常用户清单；其中，所述异常用户清单包含电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别；根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理。Corresponding to the method for processing abnormal telecommunication users provided above, the present invention also provides an electronic device. Since the embodiment of the electronic device is similar to the above-mentioned method embodiment, the description is relatively simple. For related details, please refer to the description of the above-mentioned method embodiment. The electronic device described below is only illustrative. As shown in FIG. 5 , it is a schematic diagram of a physical structure of an electronic device disclosed in an embodiment of the present invention. The electronic device may include: a processor (processor) 501, a memory (memory) 502 and a
此外，上述的存储器502中的逻辑指令可以通过软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：存储芯片、U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。In addition, the above-mentioned logic instructions in the
另一方面，本发明实施例还提供一种计算机程序产品，所述计算机程序产品包括存储在处理器可读存储介质上的计算机程序，所述计算机程序包括程序指令，当所述程序指令被计算机执行时，计算机能够执行上述各方法实施例所提供的电信异常用户的处理方法。该方法包括：获取待分析用户的业务域数据和运营域数据；将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，得到所述待分析用户中的人证分离用户；将所述人证分离用户的通讯特征数据输入到电信异常用户识别模型，得到异常用户清单；其中，所述异常用户清单包含电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别；根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理。On the other hand, an embodiment of the present invention also provides a computer program product, the computer program product includes a computer program stored on a processor-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by the computer When executing, the computer can execute the methods for handling abnormal telecommunications users provided by the above method embodiments. The method includes: acquiring the business domain data and the operation domain data of the user to be analyzed; inputting the business domain data and the operation domain data into the identification model for separation of witnesses and certificates for analysis, and obtaining the separation of witnesses and certificates of the users to be analyzed User; input the communication characteristic data of the user whose identity is separated into the telecom abnormal user identification model to obtain a list of abnormal users; wherein, the list of abnormal users includes the communication number of the abnormal telecom user and the abnormal warning level corresponding to the communication number ; Process the communication numbers of the abnormal telecommunications users according to the abnormal warning levels corresponding to the communication numbers in the abnormal user list and the preset hierarchical processing mechanism.
又一方面，本发明实施例还提供一种处理器可读存储介质，所述处理器可读存储介质上存储有计算机程序，该计算机程序被处理器执行时实现以执行上述各实施例提供的电信异常用户的处理方法。该方法包括：获取待分析用户的业务域数据和运营域数据；将所述业务域数据和所述运营域数据输入到人证分离识别模型进行分析，得到所述待分析用户中的人证分离用户；将所述人证分离用户的通讯特征数据输入到电信异常用户识别模型，得到异常用户清单；其中，所述异常用户清单包含电信异常用户的通讯号码以及所述通讯号码对应的异常预警级别；根据所述异常用户清单中所述通讯号码对应的异常预警级别和预设的分级处理机制，对所述电信异常用户的通讯号码进行处理。In yet another aspect, an embodiment of the present invention further provides a processor-readable storage medium, where a computer program is stored on the processor-readable storage medium, and the computer program is implemented when executed by a processor to perform the functions provided by the above-mentioned embodiments. Handling methods for abnormal telecom users. The method includes: acquiring the business domain data and the operation domain data of the user to be analyzed; inputting the business domain data and the operation domain data into the identification model for separation of witnesses and certificates for analysis, and obtaining the separation of witnesses and certificates of the users to be analyzed User; input the communication characteristic data of the user whose identity is separated into the telecom abnormal user identification model to obtain a list of abnormal users; wherein, the list of abnormal users includes the communication number of the abnormal telecom user and the abnormal warning level corresponding to the communication number ; Process the communication numbers of the abnormal telecommunications users according to the abnormal warning levels corresponding to the communication numbers in the abnormal user list and the preset hierarchical processing mechanism.
所述处理器可读存储介质可以是处理器能够存取的任何可用介质或数据存储设备，包括但不限于磁性存储器(例如软盘、硬盘、磁带、磁光盘(MO)等)、光学存储器(例如CD、DVD、BD、HVD等)、以及半导体存储器(例如ROM、EPROM、EEPROM、非易失性存储器(NANDFLASH)、固态硬盘(SSD))等。The processor-readable storage medium can be any available medium or data storage device that can be accessed by a processor, including but not limited to magnetic storage (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical storage (e.g., CD, DVD, BD, HVD, etc.), and semiconductor memory (such as ROM, EPROM, EEPROM, non-volatile memory (NANDFLASH), solid-state disk (SSD)), etc.
以上所描述的装置实施例仅仅是示意性的，其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。The device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without any creative efforts.
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。Through the above description of the implementations, those skilled in the art can clearly understand that each implementation can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic discs, optical discs, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.
最后应说明的是：以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.
Claims (10)
Translated fromChinesePriority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111232820.XACN116033428A (en) | 2021-10-22 | 2021-10-22 | Method and device for processing abnormal telecommunication users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111232820.XACN116033428A (en) | 2021-10-22 | 2021-10-22 | Method and device for processing abnormal telecommunication users |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116033428Atrue CN116033428A (en) | 2023-04-28 |
Family
ID=86079869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111232820.XAPendingCN116033428A (en) | 2021-10-22 | 2021-10-22 | Method and device for processing abnormal telecommunication users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116033428A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116963072A (en)* | 2023-07-27 | 2023-10-27 | 中国移动通信集团青海有限公司 | Fraud user early warning method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080140576A1 (en)* | 1997-07-28 | 2008-06-12 | Michael Lewis | Method and apparatus for evaluating fraud risk in an electronic commerce transaction |
| CN109769244A (en)* | 2019-01-08 | 2019-05-17 | 中国联合网络通信集团有限公司 | Method, device, device and readable storage medium for identifying abnormal telephone card |
| CN111275507A (en)* | 2018-12-04 | 2020-06-12 | 北京嘀嘀无限科技发展有限公司 | Order abnormity identification and order risk management and control method and system |
| CN111654866A (en)* | 2020-05-29 | 2020-09-11 | 北京合力思腾科技股份有限公司 | Method, device and computer storage medium for preventing mobile communication from fraud |
| CN112819362A (en)* | 2021-02-20 | 2021-05-18 | 中国信息通信研究院 | Method, system and medium for analyzing and monitoring real-name registration data |
| CN115409521A (en)* | 2021-05-27 | 2022-11-29 | 中国移动通信集团内蒙古有限公司 | Method, device, electronic equipment and storage medium for predicting telecommunications fraud users |
- 2021
- 2021-10-22CNCN202111232820.XApatent/CN116033428A/enactivePending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080140576A1 (en)* | 1997-07-28 | 2008-06-12 | Michael Lewis | Method and apparatus for evaluating fraud risk in an electronic commerce transaction |
| CN111275507A (en)* | 2018-12-04 | 2020-06-12 | 北京嘀嘀无限科技发展有限公司 | Order abnormity identification and order risk management and control method and system |
| CN109769244A (en)* | 2019-01-08 | 2019-05-17 | 中国联合网络通信集团有限公司 | Method, device, device and readable storage medium for identifying abnormal telephone card |
| CN111654866A (en)* | 2020-05-29 | 2020-09-11 | 北京合力思腾科技股份有限公司 | Method, device and computer storage medium for preventing mobile communication from fraud |
| CN112819362A (en)* | 2021-02-20 | 2021-05-18 | 中国信息通信研究院 | Method, system and medium for analyzing and monitoring real-name registration data |
| CN115409521A (en)* | 2021-05-27 | 2022-11-29 | 中国移动通信集团内蒙古有限公司 | Method, device, electronic equipment and storage medium for predicting telecommunications fraud users |
Non-Patent Citations (1)
| Title |
|---|
| 姚慧;马思研;: "人工智能在电信实名认证中的关键技术及应用", 电信科学, no. 05, 20 May 2019 (2019-05-20)* |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116963072A (en)* | 2023-07-27 | 2023-10-27 | 中国移动通信集团青海有限公司 | Fraud user early warning method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11710033B2 (en) | Unsupervised machine learning system to automate functions on a graph structure | |
| CN112581259B (en) | Account risk identification method and device, storage medium and electronic equipment | |
| CN117993919A (en) | Bank anti-electricity fraud data model construction method based on multi-feature fusion | |
| US20190378050A1 (en) | Machine learning system to identify and optimize features based on historical data, known patterns, or emerging patterns | |
| US20190378049A1 (en) | Ensemble of machine learning engines coupled to a graph structure that spreads heat | |
| US20190377819A1 (en) | Machine learning system to detect, label, and spread heat in a graph structure | |
| CN107038449B (en) | Method and device for identifying fraudulent user | |
| CN110837963A (en) | Risk control platform construction method based on data, model and strategy | |
| CN115908022B (en) | A method and system for warning abnormal transaction risks based on network modeling | |
| CN115409521A (en) | Method, device, electronic equipment and storage medium for predicting telecommunications fraud users | |
| CN110930218B (en) | Method and device for identifying fraudulent clients and electronic equipment | |
| CN116963072A (en) | Fraud user early warning method and device, electronic equipment and storage medium | |
| CN118674465A (en) | Fraud recognition method, fraud recognition device, electronic equipment and storage medium | |
| CN117522410A (en) | Training method of transaction risk prediction model and transaction risk prediction method | |
| CN118802303A (en) | User behavior abnormality processing method, device, equipment, medium and program product | |
| CN119762201A (en) | Financial account security monitoring method and device, electronic device and storage medium | |
| CN117575595A (en) | Payment risk identification method, device, computer equipment and storage medium | |
| CN112116358A (en) | Transaction fraud prediction method and device and electronic equipment | |
| CN118761888B (en) | Smart city service platform, method and equipment based on cloud computing and big data | |
| CN116033428A (en) | Method and device for processing abnormal telecommunication users | |
| CN119906701A (en) | A PaaS-based microservice full-link grayscale traffic management system and method | |
| CN119629636A (en) | Spam call identification method, device, computer equipment and storage medium | |
| CN118568513B (en) | A user portrait construction method, system, device and medium based on AI | |
| CN114189585B (en) | Method, device and computing equipment for detecting abnormal harassment calls | |
| CN119697318A (en) | Fraud control method, equipment and medium based on AI big model outbound call return visit |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |