CN115392916A - Abnormal consumption control method and device, electronic equipment and storage medium - Google Patents
Abnormal consumption control method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115392916A CN115392916A CN202211045508.4A CN202211045508A CN115392916A CN 115392916 A CN115392916 A CN 115392916A CN 202211045508 A CN202211045508 A CN 202211045508A CN 115392916 A CN115392916 A CN 115392916A
- Authority
- CN
- China
- Prior art keywords
- target user
- historical
- consumption behavior
- historical consumption
- abnormal
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/405—Establishing or using transaction specific rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
- G06Q20/40145—Biometric identity checks
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Computer Security & Cryptography (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a control method and device for abnormal consumption, electronic equipment and a storage medium, which can be applied to the field of artificial intelligence or finance, wherein the method comprises the following steps: receiving a transaction request of current consumption behavior initiated by a target user; acquiring a current consumption attribute corresponding to a current consumption behavior; constructing current consumption behavior characteristics corresponding to the target user by utilizing the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the target user and the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the family group of the target user; inputting the current consumption behavior characteristics corresponding to the target user into the abnormal recognition model to obtain an abnormal value corresponding to the current consumption behavior of the target user; and if the abnormal value corresponding to the current consumption behavior of the target user is larger than the preset threshold value, verifying the target user by adopting a complex verification mode.
Description
Technical Field
The present application relates to the field of transaction processing technologies, and in particular, to a method and an apparatus for controlling abnormal consumption, an electronic device, and a storage medium.
Background
With the rapid development of the internet, online consumption modes are more and more popular, wherein a user can complete online consumption by binding a mobile phone bank or other third parties with a bank, but because information in the internet is complicated, a lot of money deductions can be completed under the condition that the user is unaware of the information, so that the user brings economic loss, and therefore it is very important to effectively identify the potential online abnormal consumption behaviors of the user.
In the existing mode, the consumption behavior of the user is mainly intercepted by setting a certain quota rule, namely when the consumption quota of the user meets the quota rule, the system can automatically intercept the current consumption behavior of the user and prompt that the current consumption behavior of the user is possibly abnormal, so that the abnormal consumption behavior of the user is identified.
However, in the prior art, the consumption behavior of the user is intercepted by setting a quota rule, so that the large-amount transaction behavior initiated by the user autonomously is intercepted indiscriminately, the autonomous behavior of the user cannot be effectively identified, and the method is not flexible and convenient.
Disclosure of Invention
Based on the defects of the prior art, the application provides a control method and device for abnormal consumption, an electronic device and a storage medium, so as to solve the problem that the autonomous behavior of a user cannot be effectively identified and the user cannot be flexible and convenient in the prior art.
In order to achieve the above object, the present application provides the following technical solutions:
the application provides a control method of abnormal consumption in a first aspect, which comprises the following steps:
receiving a transaction request of current consumption behavior initiated by a target user; wherein the transaction request at least comprises a consumption initiation mode and a transaction time;
acquiring a current consumption attribute corresponding to the current consumption behavior;
constructing current consumption behavior characteristics corresponding to the target user by using the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the target user and the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the family group of the target user; the historical consumption behavior baseline of the target user is obtained by clustering in advance by using historical consumption attributes corresponding to the historical consumption behaviors of the target user in a historical time period; the historical consumption behavior baseline of the family group of the target user is obtained by clustering the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user in a historical time period in advance;
inputting the current consumption behavior characteristics corresponding to the target user into an abnormal recognition model to obtain an abnormal value corresponding to the current consumption behavior of the target user; the abnormal recognition model is obtained by training in advance by utilizing historical consumption behavior characteristics of a plurality of sample users;
and if the abnormal value corresponding to the current consumption behavior of the target user is larger than the preset threshold value, verifying the target user by adopting a complex verification mode.
Optionally, in the above method for controlling abnormal consumption, the method for training the abnormal recognition model includes:
receiving a plurality of transaction requests of historical consumption behaviors initiated by sample users;
respectively aiming at each sample user, acquiring historical consumption attributes corresponding to the historical consumption behaviors of the sample users;
constructing historical consumption behavior characteristics corresponding to the sample user by utilizing the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the sample user and the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the family group of the sample user;
respectively inputting the historical consumption behavior characteristics corresponding to each sample user into the abnormal recognition model trained in advance, and obtaining abnormal values corresponding to the historical consumption behaviors of the sample users through the abnormal recognition model;
judging whether a loss function of the abnormal recognition model is converged or not based on an abnormal value corresponding to the historical consumption behavior of each sample user;
if the loss function of the abnormal recognition model is judged to be converged, determining the abnormal recognition model as a trained abnormal recognition model;
and if the loss function of the abnormal recognition model is judged not to be converged, adjusting the parameters of the abnormal recognition model, returning to execute the steps of inputting the historical consumption behavior characteristics corresponding to each sample user into the abnormal recognition model trained in advance, and obtaining the abnormal value corresponding to the historical consumption behavior of the sample user through the abnormal recognition model.
Optionally, in the above method for controlling abnormal consumption, the method for obtaining the historical consumption behavior baseline of the target user includes:
acquiring each historical consumption behavior of the target user in a historical time period;
extracting attribute features of each historical consumption behavior of the target user in a historical time period to obtain historical consumption attributes corresponding to each historical consumption behavior of the target user;
clustering historical consumption attributes corresponding to the historical consumption behaviors of the target user to obtain a first clustering result;
and obtaining a historical consumption behavior baseline of the target user based on the first clustering result.
Optionally, in the above method for controlling abnormal consumption, the method for obtaining a historical consumption behavior baseline of the family group of the target user includes:
acquiring each historical consumption behavior of the family group of the target user in a historical time period;
performing attribute feature extraction on each historical consumption behavior of the family group of the target user in a historical time period to obtain historical consumption attributes corresponding to each historical consumption behavior of the family group of the target user;
clustering historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user to obtain a second clustering result;
and obtaining a historical consumption behavior baseline of the family group of the target user based on the second clustering result.
Optionally, in the above method for controlling abnormal consumption, the verifying the target user in a complex verification manner includes:
judging whether the mobile phone authentication and the transaction password input by the target user are both correct or not;
if the mobile phone verification and the transaction password input by the target user are judged to be correct, carrying out face recognition on the target user;
if the target user passes face recognition, determining that the target user passes verification;
and if the mobile phone verification and the transaction password input by the target user are not correct, or if the target user does not pass face recognition, determining that the target user does not pass the verification.
A second aspect of the present application provides an abnormal consumption control apparatus, including:
the first receiving unit is used for receiving a transaction request of current consumption behavior initiated by a target user; wherein, the transaction request at least comprises a consumption initiation mode and a transaction time;
the first acquisition unit is used for acquiring the current consumption attribute corresponding to the current consumption behavior;
the first construction unit is used for constructing the current consumption behavior characteristics corresponding to the target user by utilizing the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the target user and the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the family group of the target user; the historical consumption behavior baseline of the target user is obtained by clustering in advance by using historical consumption attributes corresponding to the historical consumption behaviors of the target user in a historical time period; the historical consumption behavior baseline of the family group of the target user is obtained by clustering the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user in a historical time period in advance;
the first input unit is used for inputting the current consumption behavior characteristics corresponding to the target user into an abnormal recognition model to obtain an abnormal value corresponding to the current consumption behavior of the target user; the abnormal recognition model is obtained by training in advance by utilizing historical consumption behavior characteristics of a plurality of sample users;
and the checking unit is used for checking the target user by adopting a complex checking mode if the abnormal value corresponding to the current consumption behavior of the target user is greater than the preset threshold value.
Optionally, the above control device for abnormal consumption further comprises:
the second receiving unit is used for receiving transaction requests of historical consumption behaviors initiated by a plurality of sample users;
the second acquisition unit is used for acquiring historical consumption attributes corresponding to the historical consumption behaviors of the sample users respectively aiming at the sample users;
the second construction unit constructs the historical consumption behavior characteristics corresponding to the sample user by utilizing the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the sample user and the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the family group of the sample user;
the second input unit is used for respectively inputting the historical consumption behavior characteristics corresponding to each sample user into the pre-trained abnormal recognition model, and obtaining abnormal values corresponding to the historical consumption behaviors of the sample users through the abnormal recognition model;
a first judging unit, configured to judge whether a loss function of the anomaly identification model converges based on an anomaly value corresponding to a historical consumption behavior of each of the sample users;
a first determining unit, configured to determine the abnormal recognition model as a trained abnormal recognition model if it is determined that the loss function of the abnormal recognition model is converged;
and the adjusting unit is used for adjusting the parameters of the abnormal recognition model if the loss function of the abnormal recognition model is judged not to be converged, returning to execute the steps of inputting the historical consumption behavior characteristics corresponding to each sample user into the abnormal recognition model trained in advance, and obtaining the abnormal value corresponding to the historical consumption behavior of the sample user through the abnormal recognition model.
Optionally, the above abnormal consumption control device further comprises:
the third acquisition unit is used for acquiring each historical consumption behavior of the target user in a historical time period;
the first extraction unit is used for extracting attribute features of each historical consumption behavior of the target user in a historical time period to obtain historical consumption attributes corresponding to each historical consumption behavior of the target user;
the first clustering unit is used for clustering historical consumption attributes corresponding to the historical consumption behaviors of the target user to obtain a first clustering result;
and the first obtaining unit is used for obtaining the historical consumption behavior baseline of the target user based on the first clustering result.
Optionally, the above control device for abnormal consumption further comprises:
a fourth obtaining unit, configured to obtain each historical consumption behavior of the family group of the target user in a historical time period;
the second extraction unit is used for extracting the attribute characteristics of each historical consumption behavior of the family group of the target user in a historical time period to obtain the historical consumption attribute corresponding to each historical consumption behavior of the family group of the target user;
the second clustering unit is used for clustering the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user to obtain a second clustering result;
and the second obtaining unit is used for obtaining the historical consumption behavior baseline of the family group of the target user based on the second clustering result.
Optionally, in the above control device for abnormal consumption, the verification unit includes:
the second judgment unit is used for judging whether the mobile phone verification and the transaction password input by the target user are correct or not;
the face recognition unit is used for carrying out face recognition on the target user if the mobile phone verification and the transaction password input by the target user are judged to be correct;
the second determining unit is used for determining that the target user passes verification if the target user passes face recognition;
and the third determining unit is used for determining that the target user fails to pass the verification if the mobile phone verification and the transaction password input by the target user are not correct or if the target user fails to pass the face recognition.
A third aspect of the present application provides an electronic device, comprising:
a memory and a processor;
wherein the memory is used for storing programs;
the processor is configured to execute the program, and when the program is executed, the program is specifically configured to implement the method for controlling abnormal consumption according to any one of claims 1 to 5.
A fourth aspect of the present application provides a computer storage medium storing a computer program for implementing a method of controlling abnormal consumption according to any one of claims 1 to 5 when the computer program is executed.
The method for controlling abnormal consumption comprises the steps of receiving a transaction request of current consumption behavior initiated by a target user, obtaining current consumption attributes corresponding to the current consumption behavior, establishing current consumption behavior characteristics corresponding to the target user by using the distance between the consumption attributes corresponding to the current consumption behavior and a historical consumption behavior baseline of the target user and the distance between the consumption attributes corresponding to the current consumption behavior and the historical consumption behavior baseline of a family group of the target user, inputting the current consumption behavior characteristics corresponding to the target user into an abnormal recognition model to obtain abnormal values corresponding to the current consumption behavior of the target user, wherein the abnormal recognition model is obtained by training historical consumption behavior characteristics of a plurality of sample users in advance, and finally, if the abnormal values corresponding to the current consumption behavior of the target user are larger than a preset threshold value, the target user is verified in a complex verification mode. Therefore, the consumption behaviors of the user are not intercepted by setting a quota rule, but abnormal values of the consumption behaviors of the user are obtained through an abnormal recognition model to perform abnormal behavior early warning, and the property safety of the user is effectively guaranteed.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only the embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a method for controlling abnormal consumption according to an embodiment of the present application;
fig. 2 is a flowchart of a method for obtaining a historical consumption behavior baseline of a target user according to an embodiment of the present application;
fig. 3 is a flowchart of a method for obtaining a historical consumption behavior baseline of a family group of a target user according to an embodiment of the present application;
FIG. 4 is a flowchart of a training method for an anomaly recognition model according to an embodiment of the present disclosure;
FIG. 5 is a flowchart of a method for performing a complex check according to an embodiment of the present disclosure;
fig. 6 is a schematic structural diagram of an abnormal consumption control apparatus according to another embodiment of the present application;
fig. 7 is a schematic structural diagram of an electronic device according to another embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In this application, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.
The embodiment of the application provides a control method for abnormal consumption, which specifically comprises the following steps as shown in fig. 1:
s101, receiving a transaction request of the current consumption behavior initiated by a target user.
The transaction request may include a consumption initiating manner and transaction time, and may further include a consumption device, transaction time, transaction amount, a name of a transaction opponent, a payee account number, a payee account opening bank, a current-day transaction amount, a transaction type, a current-month single-transaction maximum amount, a current-month average transaction amount, a common consumption type, and the like.
And S102, acquiring the current consumption attribute corresponding to the current consumption behavior.
Specifically, attribute feature extraction needs to be performed on the current consumption behavior initiated by the target user in advance, data dimensionality is reduced, and existing data features are sorted, so that the current consumption attribute corresponding to the current consumption behavior is obtained, and the abnormal risk behavior can be conveniently and effectively identified in the follow-up process.
S103, constructing the current consumption behavior characteristics corresponding to the target user by utilizing the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the target user and the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the family group of the target user.
The historical consumption behavior baseline of the target user is obtained by clustering in advance by using historical consumption attributes corresponding to the historical consumption behaviors of the target user in a historical time period, and the historical consumption behavior baseline of the family group of the target user is obtained by clustering in advance by using the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user in the historical time period. It should be noted that, in the embodiment of the present application, the clustering center is used as a historical consumption behavior baseline of the user and the family group thereof, so as to calculate the similarity between consumption behaviors and accurately predict the potential risk.
Specifically, it is further required to jointly construct a current consumption behavior feature [ d1, w × d2] corresponding to the target user by using the baseline weight w of the historical consumption behavior of the home group of the target user, the distance d1 between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the target user, and the distance d2 between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the home group of the target user.
Optionally, an embodiment of the present application provides a method for obtaining a historical consumption behavior baseline of a target user, as shown in fig. 2, including the following steps:
s201, obtaining each historical consumption behavior of the target user in the historical time period.
It should be noted that the consumption of the user has a diversified characteristic, but generally has a certain regularity, and in order to timely and accurately identify the risk abnormal consumption behavior and give an early warning to the user, step S201 needs to be executed for subsequent data analysis.
Alternatively, the historical time period may be the time of the past year, and of course, the specific time period may be set according to requirements.
S202, extracting attribute features of each historical consumption behavior of the target user in a historical time period to obtain historical consumption attributes corresponding to each historical consumption behavior of the target user.
It should be noted that, in the specific implementation of step S202, reference may be made to step S102, which is not described herein again.
S203, clustering the historical consumption attributes corresponding to the historical consumption behaviors of the target user to obtain a first clustering result.
It should be noted that, in order to obtain the consumption behavior baseline of the target user subsequently, and in order to trend the target user to a common data in terms of various features, and to exclude some unnecessary features, that is, some features that do not belong to normalized transactions, the behavior baseline can be obtained in a clustering manner in the subsequent step.
And S204, obtaining a historical consumption behavior baseline of the target user based on the first clustering result.
Specifically, the clustering centers of the clusters in the first clustering result may be combined to obtain the behavior baseline.
Optionally, an embodiment of the present application provides a method for obtaining a historical consumption behavior baseline of a family group of a target user, as shown in fig. 3, including the following steps:
s301, obtaining each historical consumption behavior of the family group of the target user in the historical time period.
It should be noted that, in addition to the consumption behavior of the user himself, the consumption behavior of the family group of the user may also have an indirect influence on the consumption behavior of the user, and therefore, it is also necessary to acquire each historical consumption behavior of the family group of the target user in a historical time period for analyzing the consumption behavior analysis of the user.
S302, extracting the attribute characteristics of each historical consumption behavior of the family group of the target user in a historical time period to obtain the historical consumption attribute corresponding to each historical consumption behavior of the family group of the target user.
It should be noted that, in the specific implementation of step S302, reference may be made to step S102, which is not described herein again.
S303, clustering the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user to obtain a second clustering result.
It should be noted that, in the specific implementation of step S303, reference may be made to step S203, which is not described herein again.
And S304, obtaining the historical consumption behavior baseline of the family group of the target user based on the second clustering result.
And S104, inputting the current consumption behavior characteristics corresponding to the target user into the abnormal recognition model to obtain an abnormal value corresponding to the current consumption behavior of the target user.
The abnormal recognition model is obtained by training in advance by utilizing historical consumption behavior characteristics of a plurality of sample users.
Optionally, the anomaly recognition model may be constructed based on an unsupervised algorithm of a single classification algorithm (One Class SVM), so in the embodiment of the present application, the recommendation model is trained based on an unsupervised learning mechanism to obtain a training result.
Optionally, an embodiment of the present application provides a training method for an anomaly recognition model, as shown in fig. 4, including the following steps:
s401, receiving transaction requests of historical consumption behaviors initiated by a plurality of sample users.
S402, respectively aiming at each sample user, obtaining historical consumption attributes corresponding to the historical consumption behaviors of the sample user.
It should be noted that, the step S102 may be referred to in the embodiment of the step S402, and details are not repeated herein.
S403, establishing the historical consumption behavior characteristics corresponding to the sample user by utilizing the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the sample user and the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the family group of the sample user.
It should be noted that, in the specific implementation of step S403, reference may be made to step S103, which is not described herein again.
S404, inputting the historical consumption behavior characteristics corresponding to each sample user into a pre-trained abnormal recognition model, and obtaining abnormal values corresponding to the historical consumption behaviors of the sample users through the abnormal recognition model.
Optionally, the anomaly recognition model may be constructed based on an unsupervised algorithm of a single classification algorithm (One Class SVM), so in the embodiment of the present application, the recommendation model is trained based on an unsupervised learning mechanism to obtain a training result.
S405, judging whether the loss function of the abnormal recognition model is converged or not based on the abnormal value corresponding to the historical consumption behavior of each sample user.
It should be noted that, the smaller the loss function of the model is, the better the robustness of the representative model is, so in the embodiment of the present application, it is necessary to determine whether the loss function of the abnormal recognition model converges based on the abnormal value corresponding to the historical consumption behavior of each sample user, and if it is determined that the loss function of the abnormal recognition model converges, it is necessary to perform step S406 because the training result of the model meets the expected requirement. If it is determined that the loss function of the abnormal recognition model is not converged, which indicates that iterative training of the abnormal recognition model is required, step S407 is performed.
And S406, determining the abnormal recognition model as a trained abnormal recognition model.
S407, adjusting parameters of the abnormal recognition model.
It should be noted that, when it is determined that the loss function of the abnormality recognition model does not converge, the step S404 needs to be executed again until the loss function of the abnormality recognition model achieves the convergence effect.
And S105, judging whether the abnormal value corresponding to the current consumption behavior of the target user is larger than a preset threshold value or not.
It should be noted that, in the embodiment of the present application, the current consumption behavior initiated by the user is detected in the field, and a potential online abnormal consumption behavior of the user can be effectively identified, so a risk value is set in advance accordingly, and a further determination is performed on the abnormal value obtained in step S104, that is, it is determined whether the abnormal value corresponding to the current consumption behavior of the target user is greater than a preset threshold, and if it is determined that the abnormal value corresponding to the current consumption behavior of the target user is greater than the preset threshold, it is indicated that the risk coefficient of the consumption behavior of the target user is very high, and multiple checks need to be performed on the identity of the target user, so step S106 needs to be performed. If the abnormal value corresponding to the current consumption behavior of the target user is judged to be not larger than the preset threshold value, the risk coefficient existing in the consumption behavior of the target user is low, and only a conventional verification mode needs to be adopted for the target user, for example, a mobile phone verification code is sent or a mode of judging whether a transaction password is input correctly or not.
And S106, verifying the target user by adopting a complex verification mode.
Specifically, if the target user passes the complex verification, the deduction is performed from the account of the target user.
Optionally, in another embodiment of the present application, a specific implementation manner of step S106, as shown in fig. 5, includes the following steps:
s501, judging whether the mobile phone verification and the transaction password input by the target user are both correct.
Specifically, in order to prevent the property safety of the user from being damaged when the user himself or the minor conducts a transaction under the condition that the guardian is unknown, it is necessary to determine whether the mobile phone authentication and the transaction password input by the target user are both correct, and if the mobile phone authentication and the transaction password input by the target user are both correct, it is further necessary to determine whether the user himself or herself needs to be prevented, and prevent others from conducting a transaction by using the account of the user, so that step S502 needs to be executed, and if the mobile phone authentication and the transaction password input by the target user are both incorrect, which indicates that the transaction may be risky, step S505 is executed.
And S502, carrying out face recognition on the target user.
S503, judging whether the target user passes face recognition.
Specifically, if it is determined that the target user passes the face recognition, step S504 is executed, and if it is determined that the target user does not pass the face recognition, step S505 is executed.
S504, determining that the target user passes the verification.
And S505, determining that the target user does not pass the verification.
It should be noted that, when it is determined that the target user fails to pass the verification, the target user may be notified that the current consumption behavior of the target user is at risk in a short message manner.
The method for controlling abnormal consumption comprises the steps of receiving a transaction request of current consumption behavior initiated by a target user, obtaining current consumption attributes corresponding to the current consumption behavior, establishing current consumption behavior characteristics corresponding to the target user by using the distance between the consumption attributes corresponding to the current consumption behavior and a historical consumption behavior baseline of the target user and the distance between the consumption attributes corresponding to the current consumption behavior and the historical consumption behavior baseline of a family group of the target user, inputting the current consumption behavior characteristics corresponding to the target user into an abnormal recognition model to obtain abnormal values corresponding to the current consumption behavior of the target user, wherein the abnormal recognition model is obtained by training historical consumption behavior characteristics of a plurality of sample users in advance, and finally, if the abnormal values corresponding to the current consumption behavior of the target user are larger than a preset threshold value, the target user is verified in a complex verification mode. Therefore, the consumption behaviors of the user are not intercepted by setting a quota rule, but abnormal values of the consumption behaviors of the user are obtained through an abnormal recognition model to perform abnormal behavior early warning, and the property safety of the user is effectively guaranteed.
Another embodiment of the present application provides a control apparatus for abnormal consumption, as shown in fig. 6, including the following units:
the first receiving unit 601 is configured to receive a transaction request of a current consumption behavior initiated by a target user.
Wherein the transaction request at least comprises a consumption initiation mode and a transaction time.
A first obtaining unit 602, configured to obtain a current consumption attribute corresponding to a current consumption behavior.
The first constructing unit 603 is configured to construct a current consumption behavior feature corresponding to the target user by using a distance between a consumption attribute corresponding to the current consumption behavior and a historical consumption behavior baseline of the target user, and a distance between a consumption attribute corresponding to the current consumption behavior and a historical consumption behavior baseline of a family group of the target user.
The historical consumption behavior baseline of the target user is obtained by clustering in advance by using historical consumption attributes corresponding to the historical consumption behaviors of the target user in a historical time period, and the historical consumption behavior baseline of the family group of the target user is obtained by clustering in advance by using the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user in the historical time period.
The first input unit 604 is configured to input the current consumption behavior characteristics corresponding to the target user into the anomaly identification model, so as to obtain an anomaly value corresponding to the current consumption behavior of the target user.
The abnormal recognition model is obtained by training in advance by utilizing historical consumption behavior characteristics of a plurality of sample users.
The verification unit 605 is configured to verify the target user in a complex verification manner if the abnormal value corresponding to the current consumption behavior of the target user is greater than the preset threshold.
It should be noted that, for the specific working process of the foregoing units in the embodiment of the present application, reference may be made to step S101 to step S106 in the foregoing method embodiment, which is not described herein again.
Optionally, another embodiment of the present application provides a control device for abnormal consumption, further including the following units:
and the second receiving unit is used for receiving a plurality of transaction requests of historical consumption behaviors initiated by the sample users.
And the second acquisition unit is used for acquiring historical consumption attributes corresponding to the historical consumption behaviors of the sample users respectively aiming at each sample user.
The second construction unit constructs the historical consumption behavior characteristics corresponding to the sample user by using the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the sample user and the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the family group of the sample user.
And the second input unit is used for respectively inputting the historical consumption behavior characteristics corresponding to each sample user into a pre-trained abnormal recognition model, and obtaining abnormal values corresponding to the historical consumption behaviors of the sample users through the abnormal recognition model.
And the first judgment unit is used for judging whether the loss function of the abnormal recognition model is converged or not based on the abnormal value corresponding to the historical consumption behavior of each sample user.
And a first determining unit, configured to determine the abnormal recognition model as a trained abnormal recognition model if it is determined that the loss function of the abnormal recognition model is converged.
And the adjusting unit is used for adjusting the parameters of the abnormal recognition model if the loss function of the abnormal recognition model is judged not to be converged, returning to execute the step of inputting the historical consumption behavior characteristics corresponding to each sample user into the abnormal recognition model trained in advance, and obtaining the abnormal value corresponding to the historical consumption behavior of the sample user through the abnormal recognition model.
It should be noted that, for the specific working processes of each unit provided in the foregoing embodiments of the present application, corresponding steps in the foregoing method embodiments may be referred to accordingly, and are not described herein again.
Optionally, another embodiment of the present application provides a control device for abnormal consumption, further including the following units:
and the third acquisition unit is used for acquiring each historical consumption behavior of the target user in the historical time period.
The first extraction unit is used for extracting the attribute characteristics of each historical consumption behavior of the target user in the historical time period to obtain the historical consumption attribute corresponding to each historical consumption behavior of the target user.
And the first clustering unit is used for clustering the historical consumption attributes corresponding to the historical consumption behaviors of the target user to obtain a first clustering result.
And the first obtaining unit is used for obtaining the historical consumption behavior baseline of the target user based on the first clustering result.
It should be noted that, for the specific working processes of each unit provided in the foregoing embodiments of the present application, corresponding steps in the foregoing method embodiments may be referred to accordingly, and are not described herein again.
Optionally, another embodiment of the present application provides a control device for abnormal consumption, further including the following units:
and the fourth acquisition unit is used for acquiring each historical consumption behavior of the family group of the target user in the historical time period.
And the second extraction unit is used for extracting the attribute characteristics of each historical consumption behavior of the family group of the target user in the historical time period to obtain the historical consumption attribute corresponding to each historical consumption behavior of the family group of the target user.
And the second clustering unit is used for clustering the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user to obtain a second clustering result.
And the second obtaining unit is used for obtaining the historical consumption behavior baseline of the family group of the target user based on the second clustering result.
It should be noted that, for the specific working processes of each unit provided in the foregoing embodiments of the present application, corresponding steps in the foregoing method embodiments may be referred to accordingly, and are not described herein again.
Optionally, in a control apparatus for abnormal consumption provided in another embodiment of the present application, a verification unit includes:
and the second judgment unit is used for judging whether the mobile phone verification and the transaction password input by the target user are both correct.
And the face recognition unit is used for carrying out face recognition on the target user if the mobile phone authentication and the transaction password input by the target user are judged to be correct.
And the second determination unit is used for determining that the target user passes verification if the target user passes face recognition.
And the third determining unit is used for determining that the target user fails to pass the verification if the mobile phone verification and the transaction password input by the target user are judged to be incorrect or if the target user fails to pass the face recognition.
It should be noted that, for the specific working processes of each unit provided in the foregoing embodiments of the present application, corresponding steps in the foregoing method embodiments may be correspondingly referred to, and details are not described here again.
Another embodiment of the present application provides an electronic device, as shown in fig. 7, including:
a memory 701 and a processor 702.
The memory 701 is used for storing programs.
The processor 702 is configured to execute a program, and when the program is executed, the program is specifically configured to implement a method for controlling exception consumption according to any one of the above embodiments.
Another embodiment of the present application provides a computer storage medium for storing a computer program, and when the computer program is executed, the computer program is used to implement a method for controlling abnormal consumption according to any one of the above embodiments.
Computer storage media, including persistent and non-persistent, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should be noted that the control method and device, the electronic device, and the storage medium for abnormal consumption provided by the present invention can be used in the field of artificial intelligence or the financial field. The foregoing is merely an example, and does not limit the application fields of the method and apparatus for controlling abnormal consumption, the electronic device, and the storage medium provided in the present invention.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for controlling abnormal consumption, comprising:
receiving a transaction request of current consumption behavior initiated by a target user; wherein the transaction request at least comprises a consumption initiation mode and a transaction time;
acquiring a current consumption attribute corresponding to the current consumption behavior;
constructing current consumption behavior characteristics corresponding to the target user by using the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the target user and the distance between the consumption attribute corresponding to the current consumption behavior and the historical consumption behavior baseline of the family group of the target user; the historical consumption behavior baseline of the target user is obtained by clustering in advance by using historical consumption attributes corresponding to the historical consumption behaviors of the target user in a historical time period; the historical consumption behavior baseline of the family group of the target user is obtained by clustering the historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user in a historical time period in advance;
inputting the current consumption behavior characteristics corresponding to the target user into an abnormal recognition model to obtain an abnormal value corresponding to the current consumption behavior of the target user; the abnormal recognition model is obtained by utilizing historical consumption behavior characteristics of a plurality of sample users in advance through training;
and if the abnormal value corresponding to the current consumption behavior of the target user is larger than the preset threshold value, verifying the target user by adopting a complex verification mode.
2. The method of claim 1, wherein the training method of the anomaly recognition model comprises:
receiving a plurality of transaction requests of historical consumption behaviors initiated by sample users;
respectively aiming at each sample user, obtaining historical consumption attributes corresponding to the historical consumption behaviors of the sample users;
constructing historical consumption behavior characteristics corresponding to the sample user by utilizing the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the sample user and the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the family group of the sample user;
respectively inputting the historical consumption behavior characteristics corresponding to each sample user into the pre-trained abnormal recognition model, and obtaining abnormal values corresponding to the historical consumption behaviors of the sample users through the abnormal recognition model;
judging whether a loss function of the abnormal recognition model is converged or not based on an abnormal value corresponding to the historical consumption behavior of each sample user;
if the loss function of the abnormal recognition model is judged to be converged, determining the abnormal recognition model as a trained abnormal recognition model;
and if the loss function of the abnormal recognition model is judged not to be converged, adjusting the parameters of the abnormal recognition model, returning to execute the steps of inputting the historical consumption behavior characteristics corresponding to each sample user into the abnormal recognition model trained in advance, and obtaining the abnormal value corresponding to the historical consumption behavior of the sample user through the abnormal recognition model.
3. The method of claim 1, wherein the method for obtaining the historical consumption behavior baseline of the target user comprises:
acquiring each historical consumption behavior of the target user in a historical time period;
extracting attribute features of each historical consumption behavior of the target user in a historical time period to obtain historical consumption attributes corresponding to each historical consumption behavior of the target user;
clustering historical consumption attributes corresponding to the historical consumption behaviors of the target user to obtain a first clustering result;
and obtaining the historical consumption behavior baseline of the target user based on the first clustering result.
4. The method of claim 1, wherein the method for obtaining the historical consumption behavior baseline of the family group of the target user comprises:
acquiring each historical consumption behavior of the family group of the target user in a historical time period;
performing attribute feature extraction on each historical consumption behavior of the family group of the target user in a historical time period to obtain historical consumption attributes corresponding to each historical consumption behavior of the family group of the target user;
clustering historical consumption attributes corresponding to the historical consumption behaviors of the family group of the target user to obtain a second clustering result;
and obtaining a historical consumption behavior baseline of the family group of the target user based on the second clustering result.
5. The method according to claim 1, wherein the verifying the target user by using the complex verification method includes:
judging whether the mobile phone authentication and the transaction password input by the target user are both correct or not;
if the mobile phone authentication and the transaction password input by the target user are judged to be correct, carrying out face recognition on the target user;
if the target user passes face recognition, determining that the target user passes verification;
if the mobile phone verification and the transaction password input by the target user are judged to be incorrect, or if the target user does not pass the face recognition, determining that the target user does not pass the verification.
6. An abnormal consumption control apparatus, comprising:
the first receiving unit is used for receiving a transaction request of current consumption behavior initiated by a target user; wherein the transaction request at least comprises a consumption initiation mode and a transaction time;
the first acquisition unit is used for acquiring the current consumption attribute corresponding to the current consumption behavior;
a first constructing unit, configured to construct a current consumption behavior feature corresponding to the target user by using a distance between a consumption attribute corresponding to the current consumption behavior and a historical consumption behavior baseline of the target user, and a distance between a consumption attribute corresponding to the current consumption behavior and a historical consumption behavior baseline of a family group of the target user; the historical consumption behavior baseline of the target user is obtained by clustering in advance by using historical consumption attributes corresponding to the historical consumption behaviors of the target user in a historical time period; the historical consumption behavior baseline of the family group of the target user is obtained by clustering in advance by utilizing the historical consumption attribute corresponding to the historical consumption behavior of the family group of the target user in a historical time period;
the first input unit is used for inputting the current consumption behavior characteristics corresponding to the target user into an abnormal recognition model to obtain an abnormal value corresponding to the current consumption behavior of the target user; the abnormal recognition model is obtained by training in advance by utilizing historical consumption behavior characteristics of a plurality of sample users;
and the verification unit is used for verifying the target user by adopting a complex verification mode if the abnormal value corresponding to the current consumption behavior of the target user is greater than the preset threshold value.
7. The apparatus of claim 6, further comprising:
the second receiving unit is used for receiving transaction requests of historical consumption behaviors initiated by a plurality of sample users;
the second acquisition unit is used for acquiring historical consumption attributes corresponding to the historical consumption behaviors of the sample users aiming at the sample users respectively;
the second construction unit constructs the historical consumption behavior characteristics corresponding to the sample user by utilizing the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the sample user and the distance between the historical consumption attribute corresponding to the historical consumption behavior of the sample user and the historical consumption behavior baseline of the family group of the sample user;
the second input unit is used for respectively inputting the historical consumption behavior characteristics corresponding to each sample user into the abnormal recognition model trained in advance, and obtaining abnormal values corresponding to the historical consumption behaviors of the sample users through the abnormal recognition model;
a first judging unit, configured to judge whether a loss function of the anomaly identification model converges based on an anomaly value corresponding to a historical consumption behavior of each of the sample users;
a first determining unit, configured to determine the abnormal recognition model as a trained abnormal recognition model if it is determined that a loss function of the abnormal recognition model is converged;
and the adjusting unit is used for adjusting the parameters of the abnormal recognition model if the loss function of the abnormal recognition model is judged not to be converged, returning to execute the steps of inputting the historical consumption behavior characteristics corresponding to the sample users into the abnormal recognition model trained in advance, and obtaining abnormal values corresponding to the historical consumption behaviors of the sample users through the abnormal recognition model.
8. The apparatus of claim 6, further comprising:
the third acquisition unit is used for acquiring each historical consumption behavior of the target user in a historical time period;
the first extraction unit is used for extracting attribute features of each historical consumption behavior of the target user in a historical time period to obtain historical consumption attributes corresponding to each historical consumption behavior of the target user;
the first clustering unit is used for clustering historical consumption attributes corresponding to the historical consumption behaviors of the target user to obtain a first clustering result;
and the first obtaining unit is used for obtaining the historical consumption behavior baseline of the target user based on the first clustering result.
9. An electronic device, comprising:
a memory and a processor;
wherein the memory is used for storing programs;
the processor is configured to execute the program, and when the program is executed, the program is specifically configured to implement the method for controlling abnormal consumption according to any one of claims 1 to 5.
10. A computer storage medium storing a computer program which, when executed, implements a method of controlling consumption exceptions according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211045508.4A CN115392916A (en) | 2022-08-30 | 2022-08-30 | Abnormal consumption control method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211045508.4A CN115392916A (en) | 2022-08-30 | 2022-08-30 | Abnormal consumption control method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115392916A true CN115392916A (en) | 2022-11-25 |
Family
ID=84122046
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211045508.4A Pending CN115392916A (en) | 2022-08-30 | 2022-08-30 | Abnormal consumption control method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115392916A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116611844A (en) * | 2023-07-21 | 2023-08-18 | 江苏金农股份有限公司 | Local financial consumer equity protection system based on blockchain |
| CN118365334A (en) * | 2024-06-17 | 2024-07-19 | 广州合利宝支付科技有限公司 | Payment safety identification method and system of payment terminal |
-
2022
- 2022-08-30 CN CN202211045508.4A patent/CN115392916A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116611844A (en) * | 2023-07-21 | 2023-08-18 | 江苏金农股份有限公司 | Local financial consumer equity protection system based on blockchain |
| CN118365334A (en) * | 2024-06-17 | 2024-07-19 | 广州合利宝支付科技有限公司 | Payment safety identification method and system of payment terminal |
| CN118365334B (en) * | 2024-06-17 | 2024-08-30 | 广州合利宝支付科技有限公司 | Payment safety identification method and system of payment terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3719678B1 (en) | Identity verification method and apparatus | |
| US11507645B1 (en) | Behavioral profiling method and system to authenticate a user | |
| US10708776B2 (en) | Method for automatic possession-factor authentication | |
| US9514293B1 (en) | Behavioral profiling method and system to authenticate a user | |
| US20210398129A1 (en) | Software architecture for machine learning feature generation | |
| CN112581259B (en) | Account risk identification method and device, storage medium and electronic equipment | |
| US11030624B2 (en) | Techniques to perform computational analyses on transaction information for automatic teller machines | |
| CN115392916A (en) | Abnormal consumption control method and device, electronic equipment and storage medium | |
| US11823197B2 (en) | Authenticating based on user behavioral transaction patterns | |
| CN112016929B (en) | Method and device for online payment, electronic equipment and computer storage medium | |
| CN105740667A (en) | User behavior based information identification method and apparatus | |
| TW202009834A (en) | Core-body product push and core-body method and system | |
| CN110874743A (en) | Method and device for determining account transaction risk | |
| CN109583731B (en) | Risk identification method, device and equipment | |
| US20240135379A1 (en) | Authenticating Based on Behavioral Transactional Patterns | |
| US20240338706A1 (en) | Flexible authentication | |
| CN118134634A (en) | Internet credit integrated management system | |
| US11947643B2 (en) | Fraud detection system, fraud detection method, and program | |
| US20240095327A1 (en) | Computer authentication using knowledge of former devices | |
| CN114782192B (en) | Bank account opening risk control method and system | |
| US11218493B2 (en) | Identity verification | |
| US20240211574A1 (en) | Learning model creating system, learning model creating method, and program | |
| US20240202743A1 (en) | Learning model evaluation system, learning model evaluation method, and program | |
| US20220036219A1 (en) | Systems and methods for fraud detection using game theory | |
| CN108876386B (en) | Object authentication method and device, and transaction method and device based on object authentication |
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 |