CN106611321B - False mobile phone number identification method and device - Google Patents
False mobile phone number identification method and device Download PDFInfo
- Publication number
- CN106611321B CN106611321B CN201510695109.6A CN201510695109A CN106611321B CN 106611321 B CN106611321 B CN 106611321B CN 201510695109 A CN201510695109 A CN 201510695109A CN 106611321 B CN106611321 B CN 106611321B
- Authority
- CN
- China
- Prior art keywords
- mobile phone
- phone number
- order data
- identification
- feature
- 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.)
- Active
Links
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
- Telephone Function (AREA)
Abstract
The invention provides a false mobile phone number identification method and a false mobile phone number identification device, wherein the false mobile phone number identification method comprises the following steps: acquiring current order data and acquiring the mobile phone number of a user generating the current order data; performing feature extraction on the current order data to obtain a feature value corresponding to the current order data; acquiring a pre-generated identification model, identifying according to the identification model and the characteristic value, and judging whether the mobile phone number is a false mobile phone number identified according to the identification model, wherein the identification model is obtained according to historical order data and a mobile phone number corresponding to the historical order data; and if the mobile phone number is the false mobile phone number identified according to the identification model, verifying the mobile phone number, and determining whether the mobile phone number is the false mobile phone number according to a verification result. The method can identify false mobile phone numbers.
Description
Technical Field
The invention relates to the technical field of network anti-cheating, in particular to a false mobile phone number identification method and device.
Background
In the internet application, some illegal users register a plurality of false mobile phone numbers and perform profit activities through the false mobile phone numbers, such as obtaining red packages, vouchers and the like provided by providers. In order to solve such cheating problems of illegal users, false mobile phone numbers need to be identified.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, an object of the present invention is to provide a method for identifying a false mobile phone number, which can identify the false mobile phone number.
Another object of the present invention is to provide a device for identifying a false mobile phone number.
In order to achieve the above object, an embodiment of the present invention provides a method for identifying a false mobile phone number, including: acquiring current order data and acquiring a mobile phone number of a user generating the current order data; performing feature extraction on the current order data to obtain a feature value corresponding to the current order data; acquiring a pre-generated identification model, identifying according to the identification model and the characteristic value, and judging whether the mobile phone number is a false mobile phone number identified according to the identification model, wherein the identification model is obtained according to historical order data and a mobile phone number corresponding to the historical order data; and if the mobile phone number is the false mobile phone number identified according to the identification model, verifying the mobile phone number, and determining whether the mobile phone number is the false mobile phone number according to a verification result.
According to the method for identifying the false mobile phone number provided by the embodiment of the first aspect of the invention, the false mobile phone number can be identified through the pre-generated identification model and the verification process, so that the identification of the false mobile phone number is realized, and the normal rights and interests of all parties are protected.
In order to achieve the above object, an apparatus for identifying a false mobile phone number according to an embodiment of the second aspect of the present invention includes: the acquisition module is used for acquiring current order data and acquiring the mobile phone number of a user generating the current order data; the extraction module is used for extracting the characteristics of the current order data to obtain a characteristic value corresponding to the current order data; the identification module is used for acquiring a pre-generated identification model, identifying according to the identification model and the characteristic value and judging whether the mobile phone number is a false mobile phone number identified according to the identification model, wherein the identification model is obtained according to historical order data and a mobile phone number corresponding to the historical order data; and the verification module is used for verifying the mobile phone number if the mobile phone number is the false mobile phone number identified according to the identification model and determining whether the mobile phone number is the false mobile phone number according to a verification result.
The false mobile phone number recognition device provided by the embodiment of the second aspect of the invention can recognize the false mobile phone number through the pre-generated recognition model and the verification process, thereby realizing the recognition of the false mobile phone number and protecting the normal rights and interests of all parties.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart illustrating a method for identifying a false mobile phone number according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a method for identifying a false mobile phone number according to another embodiment of the present invention;
FIG. 3 is a schematic flow chart of determining features to be extracted according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of an architecture for recognition according to a recognition model in an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an apparatus for identifying a false mobile phone number according to another embodiment of the present invention;
fig. 6 is a schematic structural diagram of an apparatus for identifying a false mobile phone number according to another embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar modules or modules having the same or similar functionality throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention. On the contrary, the embodiments of the invention include all changes, modifications and equivalents coming within the spirit and terms of the claims appended hereto.
Fig. 1 is a schematic flow chart of a method for identifying a false mobile phone number according to an embodiment of the present invention, where the method includes:
s11: and acquiring current order data and acquiring the mobile phone number of the user generating the current order data.
The current order data is, for example, order data currently generated online by the user, and the order data includes, for example, information such as a name and a price of a commodity purchased by the user. The order data generated Online may specifically be Online order data in an Online to Offline (O2O) transaction.
In addition, when the user generates order data, the user can log in by using the registered mobile phone number so as to acquire the corresponding mobile phone number, or the user logs in by using the registered account number, and the account number and the mobile phone number are stored in the database in advance in a correlated manner so as to acquire the corresponding mobile phone number.
S12: and performing feature extraction on the current order data to obtain a feature value corresponding to the current order data.
Feature extraction may be performed on the current order data in one or more of the following dimensions:
user, group order, order.
The feature to be extracted in each dimension may be predetermined, the feature to be extracted in each dimension may include one or more kinds, and a plurality of features may constitute a feature vector.
For example, the features extracted in the user dimension may be classified into user attribute class features and user statistic class features, the user attribute type features may be as shown in table 1, and the user statistic class features may be as shown in table 2:
TABLE 1
| Characteristic name | Remarks for note |
| User registration mailbox email | String of characters, retaining only the content behind the @ symbol |
| User registration IP region | Character string |
| User name uname + | Character string |
| Registration time reg _ time of user | String of characters, accurate to hours |
| Registered city reg _ city of user | Character string |
| Mobile city of mobile phone | Character string |
TABLE 2
| Characteristic name | Remarks for note |
| User's half-year total order db _ order _ count | Integer number of |
| Half-year preferential order total db _ account _ order of user | Integer number of |
| User's half-year total amount db _ total _ money | Floating point number discretized by numerical range |
| Half-year actual payment amount db _ money for user | Floating point number discretized by numerical range |
| Number db _ mobile _ num of mobile phone numbers used by user in half year | Integer number of |
| Whether the user is a new guest db _ is _ new yesterday | 0/1 |
The features extracted in the single dimension of the clique may be as shown in table 3:
TABLE 3
| Characteristic name | Remarks for note |
| Cluster single-grade category1 | Character string |
| Cluster class catalog 2 | Character string |
| City deal _ city where the bill of things is located | Character string |
The features extracted in the order dimension may be as shown in table 4:
TABLE 4
Therefore, by performing feature extraction on the current order data according to the predetermined features to be extracted in one or more dimensions, the corresponding feature value, that is, the feature value corresponding to the current order data, can be obtained.
S13: and acquiring a pre-generated identification model, identifying according to the identification model and the characteristic value, and judging whether the mobile phone number is a false mobile phone number identified according to the identification model, wherein the identification model is obtained according to historical order data and the identification result of whether the mobile phone number corresponding to the historical order data is the false mobile phone number.
In some embodiments, the recognition model may be generated while offline, as shown in FIG. 2.
For example, referring to fig. 2, the process of generating the recognition model may include:
s21: historical order data is obtained.
For example, a preset amount of historical order data in a preset time period is selected as a sample to perform model training.
S22: and performing feature extraction on the historical order data to obtain a feature value corresponding to the historical order data.
Similar to the current feature extraction process of order data, the features to be extracted in one or more dimensions can be determined, and then corresponding features are extracted from historical order data to obtain corresponding feature values.
S23: and acquiring the identification result of whether the mobile phone number corresponding to the historical order data is a false mobile phone number.
After the historical order data is obtained, the corresponding mobile phone number can be obtained, and an identification result is obtained after the mobile phone number corresponding to the historical order data is identified historically according to a preset verification mode. The verification method is, for example, a manual or preset automatic method, such as a dial test.
S24: and performing model training according to the characteristic values corresponding to the historical order data and the corresponding recognition results to generate a recognition model.
After the characteristic value corresponding to the historical order data and the identification result of the mobile phone number corresponding to the historical order data are obtained, the characteristic value and the identification result can be used as training samples to perform model training to generate an identification model. In model training, for example, a maximum entropy model training mode is adopted.
Corresponding to historical order data, if the identification result of the mobile phone number corresponding to the historical order data is a false mobile phone number, the historical order can be called a cheating order, and if the historical order data corresponds to the false mobile phone numberIf the identification result of the mobile phone number is not a false mobile phone number, the historical order can be called a normal order. Model training is carried out according to the cheating order, the normal order and the corresponding characteristic values, so that the weight corresponding to each characteristic in the cheating order and the weight corresponding to each characteristic in the normal order can be obtained respectively, and the weights corresponding to the characteristics in the normal order are assumed to be respectively a1,a2,…,aNB represents the weight corresponding to the characteristics of the order cheating1,b2,…,bNAnd (4) showing.
After model training, a first group of weight values and/or a second group of weight values can be recorded in the generated recognition model, wherein the first group of weight values are weight values corresponding to a group of features extracted corresponding to a normal order, such as the above a1,a2,…,aNThe second group of weighted values is weighted values corresponding to a group of features extracted from the corresponding cheating order, such as b1,b2,…,bN。
When the identification model is the maximum entropy model, after the maximum entropy model is generated, as shown in fig. 2, when online prediction is performed, current order data may be obtained first (S25), feature extraction may be performed on the current order data to obtain a feature value corresponding to the current order data (S26), and then online prediction may be performed on the maximum entropy according to the maximum entropy model (S27) to identify whether a mobile phone number corresponding to the current order data is a false mobile phone number identified according to the identification model.
In the identification, X may be assumed according to a set of characteristic values corresponding to the current order data1,X2,…,XNAnd expressing, respectively calculating a first score value and/or a second score value according to a first group of weight values and/or a second group of weight values recorded in the identification model, and determining whether the mobile phone number corresponding to the current order data is the false mobile phone number identified according to the identification model according to the first score value and/or the second score value.
For example, the formula for calculating the first score value is:
the calculation formula of the second score value is:
wherein A is0＝a1×X1+a2×X2+…+aN×XN，
B0＝b1×X1+b2×X2+…+bN×XN。
And if the first score value is smaller than the preset value and/or the second score value is larger than the preset value, judging that the mobile phone number is the false mobile phone number identified according to the identification model, otherwise, judging that the mobile phone number is not the false mobile phone number.
S14: and if the mobile phone number is the false mobile phone number identified according to the identification model, verifying the mobile phone number, and determining whether the mobile phone number is the false mobile phone number according to a verification result.
For example, if the mobile phone number a is identified by the identification model as a false mobile phone number, then, referring to fig. 2, the mobile phone number a may be verified manually or in a preset automatic manner (S28), such as a dial test verification manner, to finally determine whether the mobile phone number a is a false mobile phone number.
In some embodiments, in the above process, when performing feature extraction on the current order data and the historical order data, the feature to be extracted may be determined first, and then a corresponding feature value is extracted from the current order data and used as the feature value corresponding to the current order data, or a corresponding feature value is extracted from the historical order data and used as the feature value corresponding to the historical order data. As described above, the features to be extracted include features in one or more of the following dimensions: user, group order, order.
Further, referring to fig. 3, determining the features to be extracted may include:
s31: and acquiring the information gain of the feature to be verified.
For example, the original system includes a feature a and a feature B, and when determining whether the feature C is a feature to be extracted, an information gain of the feature C may be obtained first, where the information gain of the feature C is a difference between an original entropy of the system (the system including the feature a and the feature B) and a conditional entropy after the feature C is added.
S32: and if the information gain is larger than a preset gain threshold value, taking the features to be verified as the features for generating the identification model and generating the identification model.
For example, if the information gain of the feature C is greater than a preset gain threshold, the feature C is also used as a feature sample and is trained together with the feature a and the feature B to generate the recognition model.
S33: and acquiring the identification accuracy rate of the generated identification model, and determining the feature to be verified as the feature to be extracted if the identification accuracy rate is greater than a preset accuracy rate threshold value.
For example, after the identification model is generated, the mobile phone numbers corresponding to the preset number of order data may be identified and the identification accuracy may be obtained, and if the identification accuracy is higher, the feature C is determined as the feature to be extracted.
Further, as shown in fig. 4, the architecture for identifying the mobile phone number according to the identification model may be divided into: a data layer 41, a model layer 42, and a service layer 43.
The data layer 41 provides training data for model training, such as historical order data and corresponding feature values, and specifically, the historical order data may be stored in a database (e.g., Mysql) of the data layer. The characteristic values corresponding to the historical order data can be recorded in different containers according to different dimensions, for example, the transaction amount, the total order amount, the preferential amount and the like are recorded in the database; the application log can record log behavior characteristics of the user, including retrieval times, click times, conversion rate and the like; in addition, other information may also be obtained from third party data services (e.g., Eried, LBS Soul) and the like to extend the features.
And the model layer 42 is used for performing model training to generate an identification model in an off-line process, and performing false mobile phone number identification on current order data according to the identification model in an on-line process. Wherein, the maximum entropy algorithm can be specifically adopted during model training.
The service layer 43 is used to provide configuration information for model training, such as feature configuration and model configuration, and to provide current order data for online identification through data interfacing.
In the embodiment, the false mobile phone number can be identified through the pre-generated identification model and the verification process, so that the identification of the false mobile phone number is realized, and the normal rights and interests of all parties are protected.
Fig. 5 is a schematic structural diagram of an apparatus for identifying a false mobile phone number according to another embodiment of the present invention, where the apparatus 50 includes: an acquisition module 51, an extraction module 52, an identification module 53 and a verification module 54.
An obtaining module 51, configured to obtain current order data and obtain a mobile phone number of a user who generates the current order data;
the current order data is, for example, order data currently generated online by the user, and the order data includes, for example, information such as a name and a price of a commodity purchased by the user.
In addition, when the user generates order data, the user can log in by using the registered mobile phone number so as to acquire the corresponding mobile phone number, or the user logs in by using the registered account number, and the account number and the mobile phone number are stored in the database in advance in a correlated manner so as to acquire the corresponding mobile phone number.
The extracting module 52 is configured to perform feature extraction on the current order data to obtain a feature value corresponding to the current order data;
feature extraction may be performed on the current order data in one or more of the following dimensions:
user, group order, order.
The feature to be extracted in each dimension may be predetermined, the feature to be extracted in each dimension may include one or more kinds, and a plurality of features may constitute a feature vector.
The features to be extracted can be shown in tables 1-4, and will not be described herein.
Therefore, by performing feature extraction on the current order data according to the predetermined features to be extracted in one or more dimensions, the corresponding feature value, that is, the feature value corresponding to the current order data, can be obtained.
The identification module 53 is configured to acquire a pre-generated identification model, perform identification according to the identification model and the feature value, and determine whether the mobile phone number is a false mobile phone number identified according to the identification model, where the identification model is obtained according to history order data and a result of identifying whether the mobile phone number corresponding to the history order data is a false mobile phone number;
in some embodiments, referring to fig. 6, the apparatus 50 further comprises:
a modeling module 55 for obtaining historical order data; extracting the characteristics of the historical order data to obtain characteristic values corresponding to the historical order data; acquiring an identification result of whether the mobile phone number corresponding to the historical order data is a false mobile phone number; and performing model training according to the characteristic values corresponding to the historical order data and the corresponding recognition results to generate a recognition model.
For example, a preset amount of historical order data in a preset time period is selected as a sample to perform model training.
Similar to the current feature extraction process of order data, the features to be extracted in one or more dimensions can be determined, and then corresponding features are extracted from historical order data to obtain corresponding feature values.
After the historical order data is obtained, the corresponding mobile phone number can be obtained, and an identification result is obtained after the mobile phone number corresponding to the historical order data is identified historically according to a preset verification mode. The verification method is, for example, a manual or preset automatic method, such as a dial test.
In some embodiments, the identification model records a first group of weight values and/or a second group of weight values, where the first group of weight values is corresponding to a group of features extracted from a normal order, and the second group of weight values is corresponding to a group of features extracted from a cheat order, and the identification module 53 is specifically configured to:
calculating a first score value according to the characteristic value and the first group of weight values; and/or calculating a second score value according to the characteristic value and the second group of weight values;
and if the first score value is smaller than the preset value and/or if the second score value is larger than the preset value, judging that the mobile phone number is the false mobile phone number identified according to the identification model.
After the characteristic value corresponding to the historical order data and the identification result of the mobile phone number corresponding to the historical order data are obtained, the characteristic value and the identification result can be used as training samples to perform model training to generate an identification model. In model training, for example, a maximum entropy model training mode is adopted.
Corresponding to historical order data, if the identification result of the mobile phone number corresponding to the historical order data is a false mobile phone number, the historical order can be called a cheating order, and if the identification result of the mobile phone number corresponding to the historical order data is not the false mobile phone number, the historical order can be called a normal order. Model training is carried out according to the cheating order, the normal order and the corresponding characteristic values, so that the weight corresponding to each characteristic in the cheating order and the weight corresponding to each characteristic in the normal order can be obtained respectively, and the weights corresponding to the characteristics in the normal order are assumed to be respectively a1,a2,…,aNB represents the weight corresponding to the characteristics of the order cheating1,b2,…,bNAnd (4) showing.
After model training, a first group of weight values and/or a second group of weight values can be recorded in the generated recognition model, wherein the first group of weight values are weight values corresponding to a group of features extracted corresponding to a normal order, such as the above a1,a2,…,aNThe second group of weighted values is weighted values corresponding to a group of features extracted from the corresponding cheating order, such as b1,b2,…,bN。
In the identification, X may be assumed according to a set of characteristic values corresponding to the current order data1,X2,…,XNAnd expressing, respectively calculating a first score value and/or a second score value according to a first group of weight values and/or a second group of weight values recorded in the identification model, and determining whether the mobile phone number corresponding to the current order data is the false mobile phone number identified according to the identification model according to the first score value and/or the second score value.
For example, the formula for calculating the first score value is:
the calculation formula of the second score value is:
wherein A is0＝a1×X1+a2×X2+…+aN×XN，
B0＝b1×X1+b2×X2+…+bN×XN。
And if the first score value is smaller than the preset value and/or the second score value is larger than the preset value, judging that the mobile phone number is the false mobile phone number identified according to the identification model, otherwise, judging that the mobile phone number is not the false mobile phone number.
And the verification module 54 is configured to verify the mobile phone number if the mobile phone number is the false mobile phone number identified according to the identification model, and determine whether the mobile phone number is the false mobile phone number according to a verification result.
For example, if the mobile phone number a is identified as a false mobile phone number by the identification model, then the mobile phone number a may be verified manually or in a preset automatic manner, for example, in a dial test verification manner, to finally determine whether the mobile phone number a is a false mobile phone number.
In some embodiments, referring to fig. 6, the apparatus 50 further comprises:
a determining module 56, configured to determine features to be extracted, so as to extract the features to be extracted from the order data to obtain a feature value, where the features to be extracted include features in one or more of the following dimensions: user, group order, order.
Optionally, the determining module 56 is specifically configured to:
obtaining information gain of the feature to be verified;
if the information gain is larger than a preset gain threshold value, the feature to be verified is used as the feature for generating the identification model and generating the identification model;
and acquiring the identification accuracy rate of the generated identification model, and determining the feature to be verified as the feature to be extracted if the identification accuracy rate is greater than a preset accuracy rate threshold value.
For example, the original system includes a feature a and a feature B, and when determining whether the feature C is a feature to be extracted, an information gain of the feature C may be obtained first, where the information gain of the feature C is a difference between an original entropy of the system (the system including the feature a and the feature B) and a conditional entropy after the feature C is added.
For example, if the information of the feature C is larger than the preset gain threshold, the feature C is also used as a feature sample and is trained together with the feature a and the feature B to generate the recognition model.
For example, after the identification model is generated, the mobile phone numbers corresponding to the preset number of order data may be identified and the identification accuracy may be obtained, and if the identification accuracy is higher, the feature C is determined as the feature to be extracted.
In the embodiment, the false mobile phone number can be identified through the pre-generated identification model and the verification process, so that the identification of the false mobile phone number is realized, and the normal rights and interests of all parties are protected.
It should be noted that the terms "first," "second," and the like in the description of the present invention are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present invention, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a separate product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (8)
1. A false mobile phone number identification method is characterized by comprising the following steps:
acquiring current order data and acquiring a mobile phone number of a user generating the current order data;
performing feature extraction on the current order data to obtain a feature value corresponding to the current order data;
acquiring a pre-generated identification model, identifying according to the identification model and the characteristic value, and judging whether the mobile phone number is a false mobile phone number identified according to the identification model, wherein the identification model is obtained according to the characteristic value corresponding to historical order data and the identification result of whether the mobile phone number corresponding to the historical order data is a false mobile phone number, the identification result is obtained after the mobile phone number corresponding to the historical order data is identified according to a preset verification mode, and model training is performed according to the characteristic value corresponding to the historical order data and the corresponding identification result to generate the identification model;
if the mobile phone number is the false mobile phone number identified according to the identification model, verifying the mobile phone number, and determining whether the mobile phone number is the false mobile phone number according to a verification result;
the method for identifying the cheat order comprises the following steps of recording a first group of weight values and/or a second group of weight values in the identification model, wherein the first group of weight values are weight values corresponding to a group of features extracted corresponding to the normal order, the second group of weight values are weight values corresponding to a group of features extracted corresponding to the cheat order, and whether the mobile phone number is a false mobile phone number identified according to the identification model or not is judged according to the identification model and the feature values, and the method comprises the following steps:
calculating a first score value according to the characteristic value and the first group of weight values; and/or calculating a second score value according to the characteristic value and the second group of weight values;
and if the first score value is smaller than a preset value and/or if the second score value is larger than the preset value, judging that the mobile phone number is a false mobile phone number identified according to the identification model.
2. The method of claim 1, further comprising:
acquiring historical order data;
extracting the characteristics of the historical order data to obtain characteristic values corresponding to the historical order data;
acquiring an identification result of whether the mobile phone number corresponding to the historical order data is a false mobile phone number;
and performing model training according to the characteristic values corresponding to the historical order data and the corresponding recognition results to generate a recognition model.
3. The method according to any one of claims 1-2, further comprising:
determining features needing to be extracted, wherein the features needing to be extracted comprise features in one or more of the following dimensions: users, group orders, orders;
the step of performing feature extraction on the order data to obtain a feature value comprises the following steps:
and extracting the features to be extracted from the order data to obtain a feature value.
4. The method of claim 3, wherein the determining features to be extracted comprises:
obtaining information gain of the feature to be verified;
if the information gain is larger than a preset gain threshold value, the feature to be verified is used as the feature for generating the identification model and generating the identification model;
and acquiring the identification accuracy rate of the generated identification model, and determining the feature to be verified as the feature to be extracted if the identification accuracy rate is greater than a preset accuracy rate threshold value.
5. An apparatus for identifying a false mobile phone number, comprising:
the acquisition module is used for acquiring current order data and acquiring the mobile phone number of a user generating the current order data;
the extraction module is used for extracting the characteristics of the current order data to obtain a characteristic value corresponding to the current order data;
the identification module is used for acquiring a pre-generated identification model, identifying according to the identification model and the characteristic value and judging whether the mobile phone number is a false mobile phone number identified according to the identification model, wherein the identification model is obtained according to the characteristic value corresponding to historical order data and the identification result of whether the mobile phone number corresponding to the historical order data is the false mobile phone number, the identification result is obtained after the mobile phone number corresponding to the historical order data is identified according to a preset verification mode, and model training is carried out according to the characteristic value corresponding to the historical order data and the corresponding identification result to generate the identification model;
the verification module is used for verifying the mobile phone number if the mobile phone number is the false mobile phone number identified according to the identification model and determining whether the mobile phone number is the false mobile phone number according to a verification result;
the identification module is configured to record a first group of weight values and/or a second group of weight values, where the first group of weight values are weight values corresponding to a group of features extracted corresponding to a normal order, the second group of weight values are weight values corresponding to a group of features extracted corresponding to a cheat order, and the identification module is specifically configured to:
calculating a first score value according to the characteristic value and the first group of weight values; and/or calculating a second score value according to the characteristic value and the second group of weight values;
and if the first score value is smaller than a preset value and/or if the second score value is larger than the preset value, judging that the mobile phone number is a false mobile phone number identified according to the identification model.
6. The apparatus of claim 5, further comprising:
the modeling module is used for acquiring historical order data; extracting the characteristics of the historical order data to obtain characteristic values corresponding to the historical order data; acquiring an identification result of whether the mobile phone number corresponding to the historical order data is a false mobile phone number; and performing model training according to the characteristic values corresponding to the historical order data and the corresponding recognition results to generate a recognition model.
7. The apparatus of any of claims 5-6, further comprising:
the determining module is used for determining the features to be extracted so as to extract the features to be extracted from the order data to obtain feature values, wherein the features to be extracted comprise features in one or more dimensions as follows: user, group order, order.
8. The apparatus of claim 7, wherein the determining module is specifically configured to:
obtaining information gain of the feature to be verified;
if the information gain is larger than a preset gain threshold value, the feature to be verified is used as the feature for generating the identification model and generating the identification model;
and acquiring the identification accuracy rate of the generated identification model, and determining the feature to be verified as the feature to be extracted if the identification accuracy rate is greater than a preset accuracy rate threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510695109.6A CN106611321B (en) | 2015-10-22 | 2015-10-22 | False mobile phone number identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510695109.6A CN106611321B (en) | 2015-10-22 | 2015-10-22 | False mobile phone number identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106611321A CN106611321A (en) | 2017-05-03 |
| CN106611321B true CN106611321B (en) | 2020-09-25 |
Family
ID=58612613
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510695109.6A Active CN106611321B (en) | 2015-10-22 | 2015-10-22 | False mobile phone number identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106611321B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107256488B (en) * | 2017-06-07 | 2020-10-13 | 北京星选科技有限公司 | Order cheating detection method and device |
| CN109598566B (en) * | 2017-09-30 | 2021-07-09 | 北京嘀嘀无限科技发展有限公司 | Ordering prediction method, ordering prediction device, computer equipment and computer readable storage medium |
| CN107809762B (en) * | 2017-11-01 | 2021-05-25 | 南京欣网互联网络科技有限公司 | Security risk control method for card-raising identification by utilizing big data and equipment fingerprints |
| CN110322573A (en) * | 2018-03-30 | 2019-10-11 | 北京红马传媒文化发展有限公司 | User authentication method, user authentication device and electronic equipment |
| CN109219051B (en) * | 2018-11-28 | 2023-02-14 | 上海大汉三通通信股份有限公司 | False number determination method, device, equipment and readable storage medium |
| CN110297848B (en) * | 2019-07-09 | 2024-02-23 | 深圳前海微众银行股份有限公司 | Recommendation model training method, terminal and storage medium based on federal learning |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103853948B (en) * | 2012-11-28 | 2018-01-19 | 阿里巴巴集团控股有限公司 | The identification of user identity, the filtering of information and searching method and server |
| CN103064987B (en) * | 2013-01-31 | 2016-09-21 | 五八同城信息技术有限公司 | A kind of wash sale information identifying method |
| CN103279868B (en) * | 2013-05-22 | 2016-08-17 | 兰亭集势有限公司 | A kind of method and apparatus of automatic identification swindle order |
| CN104574126B (en) * | 2013-10-17 | 2018-10-23 | 阿里巴巴集团控股有限公司 | A kind of user characteristics recognition methods and device |
| CN104092601B (en) * | 2014-07-28 | 2017-12-05 | 北京微众文化传媒有限公司 | The recognition methods of social networks account and device |
| CN104915423B (en) * | 2015-06-10 | 2018-06-26 | 深圳市腾讯计算机系统有限公司 | The method and apparatus for obtaining target user |
-
2015
- 2015-10-22 CN CN201510695109.6A patent/CN106611321B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106611321A (en) | 2017-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106611321B (en) | False mobile phone number identification method and device | |
| CN106022834B (en) | Advertisement anti-cheating method and device | |
| US9596356B2 (en) | Analyzing voice characteristics to detect fraudulent call activity and take corrective action without using recording, transcription or caller ID | |
| CN106355431B (en) | Cheating flow detection method and device and terminal | |
| US20140351109A1 (en) | Method and apparatus for automatically identifying a fraudulent order | |
| CN106570718B (en) | Information delivery method and delivery system | |
| US20090070289A1 (en) | Methods, Systems, and Computer Program Products for Estimating Accuracy of Linking of Customer Relationships | |
| CN110060053B (en) | Identification method, equipment and computer readable medium | |
| CN108876188B (en) | Inter-connected service provider risk assessment method and device | |
| CN108876105B (en) | Transaction risk control method and device | |
| CN109993392A (en) | Business complaint risk predictor method, calculates equipment and storage medium at device | |
| Knežević | The characteristics of forensic audit and differences in relation to external audit | |
| CN112819476A (en) | Risk identification method and device, nonvolatile storage medium and processor | |
| CN103250376A (en) | Method and system for carrying out predictive analysis relating to nodes of a communication network | |
| KR101021400B1 (en) | System and method for determining value of data registered free | |
| CN107330709B (en) | Method and device for determining target object | |
| US20220284429A1 (en) | System, method, and recording medium having recorded thereon a program | |
| CN109214634A (en) | A kind of information processing method, device and information processing readable medium | |
| CN107862599B (en) | Bank risk data processing method and device, computer equipment and storage medium | |
| CN110609969A (en) | Information processing method and device | |
| CN111104628A (en) | User identification method and device, electronic equipment and storage medium | |
| CN111179023B (en) | Order identification method and device | |
| CN110348983B (en) | Transaction information management method and device, electronic equipment and non-transitory storage medium | |
| CN109711984B (en) | Pre-loan risk monitoring method and device based on collection urging | |
| CN116976664A (en) | Risk merchant prediction method, system, computer and readable storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |