CN113159782B - Minor anti-addiction processing method, device and equipment based on federal learning - Google Patents

Minor anti-addiction processing method, device and equipment based on federal learning Download PDF

Info

Publication number
CN113159782B
CN113159782B CN202110326043.9A CN202110326043A CN113159782B CN 113159782 B CN113159782 B CN 113159782B CN 202110326043 A CN202110326043 A CN 202110326043A CN 113159782 B CN113159782 B CN 113159782B
Authority
CN
China
Prior art keywords
entertainment
addiction
account
payment
data
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
Application number
CN202110326043.9A
Other languages
Chinese (zh)
Other versions
CN113159782A (en
Inventor
杨哲
杨一鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alipay Hangzhou Information Technology Co Ltd
Original Assignee
Alipay Hangzhou Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Alipay Hangzhou Information Technology Co Ltd filed Critical Alipay Hangzhou Information Technology Co Ltd
Priority to CN202110326043.9A priority Critical patent/CN113159782B/en
Publication of CN113159782A publication Critical patent/CN113159782A/en
Application granted granted Critical
Publication of CN113159782B publication Critical patent/CN113159782B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Accounting & Taxation (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Computer Security & Cryptography (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The embodiment of the specification discloses an anti-addiction treatment method for minors based on federal learning, which comprises the following steps: determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to the service data of the payment platform and the entertainment platform; collecting related data of an entertainment account on an entertainment platform; determining whether a user corresponding to the entertainment account has a payment account on the payment platform; if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model; if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor according to a second anti-addiction auxiliary system, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms; and if the user is determined to be a minor, performing corresponding anti-addiction treatment.

Description

Minor anti-addiction processing method, device and equipment based on federal learning
Technical Field
The specification relates to the field of machine learning, in particular to a method, a device and equipment for preventing minor addiction based on federal learning.
Background
The entertainment industries such as games, live broadcast and the like have relatively fast development in recent years, attract the attention of people through various rich activities and playing methods, and a plurality of users carry out a great amount of recharging behaviors in the entertainment industries. However, many minors use parent money to recharge on entertainment platforms, and these enthusiasm behaviors seriously affect the physical and mental health of the minors and need to be deterred.
Currently, some entertainment platforms are accessible by simple registration in order to attract more users. And subsequently, the juveniles cannot be effectively subjected to anti-addiction treatment.
Based on this, there is a need for a more effective anti-addiction treatment regimen for minors.
Disclosure of Invention
One or more embodiments of the present specification provide a method, an apparatus, and a device for anti-addiction processing for minors based on federal learning, which are used to solve the following technical problems: there is a need for a more effective anti-addiction treatment regimen for minors.
One or more embodiments of the present disclosure adopt the following technical solutions:
one or more embodiments of the present specification provide a method for anti-addiction treatment of minors based on federal learning, comprising:
an anti-addiction processing method for minors based on federal learning comprises the following steps:
determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform;
judging whether a user corresponding to the entertainment account has a payment account on the payment platform;
if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model;
if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary system, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms;
and if the user is determined to be a minor, performing corresponding anti-addiction treatment.
One or more embodiments of the present specification provide a federal learning-based anti-addiction processing apparatus for minor, including:
the system comprises a first model determining unit, a second model determining unit and a third model determining unit, wherein the first model determining unit determines a first anti-addiction auxiliary model, and the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
the system comprises a first data acquisition unit, a second data acquisition unit and a payment processing unit, wherein the first data acquisition unit is used for acquiring related data of an entertainment account on a specified entertainment platform, and the specified entertainment platform provides a payment channel for the entertainment account through the payment platform;
the payment account determining unit is used for judging whether the user corresponding to the entertainment account has a payment account on the payment platform;
if the first anti-addiction processing unit exists, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model, and if the user is judged to be a minor, performing corresponding anti-addiction processing;
and if the second model does not exist, determining a second anti-addiction auxiliary model, judging whether the user corresponding to the entertainment account is a minor or not by the second anti-addiction processing unit according to the related data of the entertainment account and the second anti-addiction auxiliary system, and if the user is judged to be a minor, performing corresponding anti-addiction processing, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms.
One or more embodiments of the present specification provide a federal learning-based anti-addiction processing apparatus for minor, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform;
judging whether a user corresponding to the entertainment account has a payment account on the payment platform;
if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model;
if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary system, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms;
and if the user is determined to be a minor, performing corresponding anti-addiction treatment.
One or more embodiments of the present specification provide a non-transitory computer storage medium storing computer-executable instructions configured to:
determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform;
judging whether a user corresponding to the entertainment account has a payment account on the payment platform;
if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model;
if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary system, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms;
and if the user is determined to be a minor, performing corresponding anti-addiction treatment.
The embodiment of the specification adopts at least one technical scheme which can achieve the following beneficial effects: by the federal learning process based on the payment platform and the entertainment platform, training data are expanded, so that the extraction capability of the first anti-addiction auxiliary model obtained by training on the characteristics of the entertainment account is further enhanced, and when a user operates on the entertainment platform, whether the user corresponding to the entertainment account is a minor can be more accurately identified, instead of only utilizing simple registration data on the entertainment platform, so that the method is more effective; moreover, the second anti-addiction auxiliary model obtained through transfer learning carries out differentiated identification on the users without the payment accounts of the payment platform, and the overall accurate identification capability of the scheme is effectively improved with extremely low additional cost.
Drawings
In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort. In the drawings:
fig. 1 is a schematic flow chart of a method for anti-addiction treatment of minors based on federal learning according to one or more embodiments of the present disclosure;
fig. 2 is a detailed flowchart of the method in fig. 1 in an application scenario provided in one or more embodiments of the present disclosure;
FIG. 3 is a schematic diagram of training a first anti-addiction auxiliary model provided in one or more embodiments of the present disclosure;
FIG. 4 is a schematic diagram illustrating a user paying for an adult corresponding to an entertainment account as provided by one or more embodiments of the present disclosure;
FIG. 5 is a schematic illustration of a user paying for minors corresponding to an entertainment account as provided by one or more embodiments of the present disclosure;
fig. 6 is a schematic structural diagram of an anti-addiction processing device for minors based on federal learning according to one or more embodiments of the present disclosure;
fig. 7 is a schematic structural diagram of an anti-addiction processing device for minors based on federal learning according to one or more embodiments of the present disclosure.
Detailed Description
The embodiment of the specification provides a method, a device, equipment and a storage medium for anti-addiction processing of minors based on federal learning.
In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present specification, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present specification without any creative effort shall fall within the protection scope of the present specification.
The anti-addiction in the embodiment of the present specification is mainly limited to the payment behavior of the minors. If the minor is not cared for, the minor may use the money of the parents to recharge the game or spend money to watch the maid broadcasters, and such an enthusiasm may affect the physical and mental health of the minor and be prevented.
Since minors have not yet stepped into society, the required expenses are mostly provided by parents, and under the condition of no strong cognition on money, huge consumption is easy to occur through an entertainment platform, thereby causing huge fund loss to families.
Meanwhile, many parents find out that after the children use the payment platform to complete payment, the parents may complain about the entertainment platform or the payment platform, which will damage the reputation of the entertainment platform and the payment platform.
In addition, the juveniles pay for the entertainment platform without restriction, which can lead the country to restrain the entertainment industry and is not beneficial to the development of the industry.
The technical solution provided in the present specification is described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flow diagram of a minor anti-addiction processing method based on federal learning according to one or more embodiments of the present disclosure, where the flow may be executed by a computing device in the field of entertainment business or payment business (for example, a server or an intelligent mobile terminal corresponding to a payment business, and a server or an intelligent terminal corresponding to an entertainment business) or by an execution unit of a minor anti-addiction processing system, and some input parameters or intermediate results in the flow allow manual intervention and adjustment to help improve accuracy.
The process in fig. 1 may include the following steps:
s102: and determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to the respective service data of the payment platform and one or more entertainment platforms.
In one or more embodiments of the present description, an entertainment platform is a game platform, a live broadcast platform, or other entertainment consumer platform. The payment platform can be a third-party guarantee platform in the transaction process of the buyer and the seller, and is an independent mechanism for guaranteeing the benefits of the two parties under the supervision of a bank, such as a third-party financial payment platform, a payment platform with instant messaging, a payment platform released by each bank and the like.
The entertainment platform provides virtual goods, physical goods, services and the like for the user, the payment platform provides services for the entertainment platform, and the user can pay through the payment platform when recharging and purchasing the virtual goods, the physical goods and the services on the entertainment platform.
In one or more embodiments of the present description, the business data for one or more entertainment platforms includes entertainment account information, behavior data, and tag data. The entertainment account information comprises identity information, wherein the identity information comprises gender, age, identification card number, living city and the like. The behavior data includes payment behavior, browsing behavior, account level, and the like. The tag data includes adults and minors, etc., and the adults or minors are further subdivided into various types according to specific behaviors.
It should be noted that, in order to increase the attraction of more users, some entertainment platforms may establish an entertainment platform account through simple account registration, and the identity information of the entertainment platform account is relatively less, or the entertainment platform does not verify the identity information of the account, so that it is impossible to accurately determine whether a user (a registrant or a current user of the entertainment account) corresponding to the entertainment account is a minor in the following. In addition, the minor can also register the account by using the identity of the parent, which further increases the difficulty of judgment, so that whether the user corresponding to the entertainment account is the minor cannot be accurately identified only by virtue of the registration data of the entertainment platform. If the method can be used with the help of more service data of the entertainment platform and the service data of the payment platform, the identification accuracy can be greatly improved.
In one or more embodiments of the present description, the service data of the payment platform includes payment account information and payment information. The payment account information includes identity data, wherein the identity data includes gender, age, identification number, city of residence, and the like. The payment information includes payment frequency, payment success rate and the like.
The data can reflect the behavior habit of the current entertainment account, and the behavior habit can further indirectly reflect more information of the current entertainment account. For example, the determination is performed according to the payment habit of the current payment account, if the current payment account rarely pays on the entertainment platform (the payment account and the entertainment account may be the same user or different users at this time), and the payment account suddenly makes a large amount of payments on the entertainment account, it may be determined that the entertainment account has a high probability of being an account registered by a minor; for another example, the determination is performed according to the browsing content of the current entertainment account, and if the browsing content of the entertainment account is mostly the content concerned by the minors, it may be determined that the entertainment account is most likely the account registered by the minors.
For another example, the determination is performed according to the payment behavior of the current entertainment account, since the minor can borrow the mobile phone of the parent in a short period, if the entertainment account performs a large amount of payment only in a period of time and does not perform such operation in other periods, it can be determined that the entertainment account is the account registered by the minor with a high probability; for another example, the determination is performed according to the browsing time of the current entertainment account, if the browsing time of the entertainment account is mostly non-student class time, it can be determined that the entertainment account is most likely to be an account registered by a minor, and more information is limited to a payment platform or an entertainment platform and may be difficult to be comprehensively mined.
In one or more embodiments of the present description, federated learning is the practice of splitting data sets in a vertical (i.e., feature dimension) under conditions of more user overlap and less user feature overlap for multiple platform (paymate and one or more entertainment platforms) data sets, and extracting the portion of data for which the paymate and entertainment platforms have the same user but not the same user features for training. Meanwhile, when federal learning is applied, encrypted samples of data of multiple platforms can be aligned, so that common users of two parties can be confirmed on the premise that one or more entertainment platforms and one or more payment platforms do not disclose respective data, users which do not overlap with each other are not exposed, modeling is carried out by combining characteristics of the users, and the data privacy protection of the parties is facilitated. The embodiment of the specification finally obtains the first anti-addiction auxiliary model by training the business data of one or more entertainment platforms and the business data of the payment platform through federal learning.
S104: the method comprises the steps of collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through a payment platform.
In one or more embodiments of the present description, the data related to the entertainment account on the designated entertainment platform may include entertainment account information, behavior data, tag data, and the like for the current entertainment account on the entertainment platform. The payment channel comprises a payment interface provided by the payment platform for the appointed entertainment platform so as to realize that the appointed entertainment platform completes transactions through the payment platform, the entertainment platform can be provided with a plurality of payment channels respectively provided by different payment platforms for users to select, and the selection of the payment channels by the users does not influence the judgment whether the users corresponding to the subsequent entertainment accounts are minors or not.
It should be noted that the designated entertainment platform may not be included in the entertainment platform mentioned in S102. By training with data contributed by a portion of the entertainment platforms, the resulting first anti-addiction auxiliary model may be applicable to more entertainment platforms, and may even be applicable to platforms in other fields, such as public service platforms, government affairs platforms, and the like.
S106: and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model, and if so, performing corresponding anti-addiction treatment.
In one or more embodiments of the present disclosure, the first anti-addiction auxiliary model is trained according to service data of one or more entertainment platforms and a payment platform, but when the first anti-addiction auxiliary model is applied, it may be determined whether a user corresponding to an entertainment account is a minor according to related data of the entertainment account on a designated entertainment platform, that is, according to the service data of the designated entertainment platform, or it may be determined whether the user corresponding to the entertainment account is a minor according to related data of the entertainment account on the designated entertainment platform and related data of the payment account on the payment platform, that is, according to the service data of the designated entertainment platform and the service data of the payment platform, that is, it is determined whether the user corresponding to the entertainment account is a minor. If not, the anti-addiction treatment is not required. Wherein, the anti-addiction treatment can be a case reminder or require the user to perform identity verification.
In one or more embodiments of the present description, what is obtained by the first anti-addiction auxiliary model may be an intermediate result of the auxiliary determination, not necessarily a final result, and the minor anti-addiction processing system may combine the intermediate result to determine whether the user to which the entertainment account corresponds is a minor. For example, the first anti-addiction auxiliary model identifies that the operation of the entertainment account on the specified entertainment platform is consistently inconsistent, and the user watches some content of interest of the minor, but may need more information to determine whether the user is a minor, and it is assumed that it is further necessary to know whether the subsequent irregular operation (such as watching a main broadcast, recharging, etc.) occurs, and if the irregular operation does not occur, it may be determined that the user corresponding to the entertainment account is an adult, and may be an operation performed by the minor under the accompany of parents, and the anti-addiction process for the minor is not required.
Corresponding to the embodiment of fig. 1, fig. 2 is a detailed flowchart of the method in fig. 1 in an application scenario provided by one or more embodiments of the present disclosure.
The flow in fig. 2 may include the following steps:
s202, judging whether a user corresponding to the entertainment account on the entertainment platform has a payment account on the payment platform, if so, executing S204; if not, go to step S212.
In one or more embodiments of the present disclosure, the corresponding user in S202 may refer to a registrant of the entertainment account, but the registrant may also be a counterfeit identity, for example, a child is registered with a parent, and the registrant may be considered to be still a parent. In this case, the subsequent determination may be to determine whether the current user of the entertainment account (which may be the parent or the child or another person) is a minor, because it is only the registration identity that identifies the minor as being immortal, and the present solution uses the registration identity primarily or for the purpose of attempting to obtain multi-platform association data, rather than brainless dependent registration identities.
S204: and determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to the respective service data of the payment platform and one or more entertainment platforms.
In one or more embodiments of the present disclosure, before performing this step, the embodiment of the present disclosure further needs to train a first anti-addiction auxiliary model, and the specific steps may include:
(1) the method comprises the steps of constructing a training data set and establishing an initial first anti-addiction auxiliary model, wherein the training data set comprises business data of one or more entertainment platforms and payment platforms respectively and label data marked in advance, and the business data of the one or more entertainment platforms and the payment platforms respectively are sample data.
Note that the tag data may include minor and adult. When the label data is marked, the label data can be marked according to actual conditions.
(2) Performing an encryption-based sample alignment process on the training data set.
Because the user groups of the payment platform and the one or more entertainment platforms are not completely overlapped, the common users of the payment platform and the one or more entertainment platforms can be confirmed by utilizing an encryption-based sample alignment technology on the premise that the payment platform and the one or more entertainment platforms do not disclose respective data, the users which are not overlapped with each other are not exposed, and finally modeling is carried out by combining the characteristics of the users. The sample alignment technology can adopt RSA (RSA algorithm), public keys generated by the collaborators are distributed to the payment platform and one or more entertainment platforms to encrypt the users, the encrypted users are decrypted by the collaborators through private keys, and the shared users of the payment platform and the one or more entertainment platforms are taken out.
The public key and the private key in the RSA algorithm are generated by collaborators, and the specific steps are as follows:
a. randomly finding two large prime numbers P and Q, and calculating n-P-Q;
b. calculating an euler function m ═ $ (n) for n;
c. randomly selecting a positive integer e such that 1< e < m, and e is coprime to m d;
d. d is obtained according to the expanded Euclidean algorithm, so that the remainder of e x d/m is 1;
e. the public key is (n, e), and the private key is (n, d);
f. the public key is used for encryption, and the ciphertext is the remainder of the plaintext divided by n to the power e; the private key is used for decryption, and the decrypted plaintext is the remainder of the ciphertext divided by n by the power of d.
It should be noted that, in order to ensure privacy of data, the embodiments of the present specification select a collaborator to manage distribution of a public key, where the collaborator may select an existing framework to reduce implementation cost of a scheme.
(3) And obtaining the corresponding gradient and loss of one or more entertainment platforms and the payment platform according to the processed training data set, and determining the parameters of the first anti-addiction auxiliary model, thereby obtaining the first anti-addiction auxiliary model meeting the conditions.
Further, in one or more embodiments of the present specification, obtaining a gradient and a loss of the payment platform corresponding to the one or more entertainment platforms according to the processed training data set, and determining a parameter of the first anti-addiction auxiliary model, so as to obtain a qualified first anti-addiction auxiliary model, may specifically include:
the collaborator distributes the public key to the payment platform and one or more entertainment platforms to encrypt the data to be exchanged in the training process;
the payment platform and the one or more entertainment platforms respectively calculate intermediate results corresponding to the service data of the payment platform and the service data of the one or more entertainment platforms, and carry out encryption interaction to obtain corresponding gradient and loss, wherein A is set as the payment platform, B is set as the one or more entertainment platforms, and the objective function is as follows:
Figure BDA0002994703810000111
wherein, Θ is a model parameter (weight), x is a feature, and y is a label.
Is provided with
Figure BDA0002994703810000112
Encrypted objective functionComprises the following steps:
Figure BDA0002994703810000113
is provided with
Figure BDA0002994703810000114
Then the gradient is:
Figure BDA0002994703810000115
Figure BDA0002994703810000116
where λ is a regular coefficient.
The payment platform and the one or more entertainment platforms respectively calculate based on the encrypted gradient values, and meanwhile the one or more entertainment platforms calculate loss according to the tag data and collect calculation results to the collaborators so that the collaborators can calculate total gradient through the collected calculation results and decrypt the total gradient; the collaborator respectively transmits the decrypted total gradient back to the payment platform and one or more entertainment platforms; updating the parameters of the respective models by the payment platform and one or more entertainment platforms according to the total gradient; and finishing the training of the first anti-addiction auxiliary model through an iterative mode until the loss function is converged, and obtaining the first anti-addiction auxiliary model meeting the conditions.
It should be noted that, the training of the first anti-addiction auxiliary model can be referred to fig. 3, and fig. 3 includes the following steps:
step 1: the payment platform and one or more entertainment platforms complete data preparation and respectively wash out characteristic data required by training.
Step 2: and completing the task of aligning the encrypted samples through the authentication module.
And step 3: and the collaborators decrypt the data first and perform fusion learning on the decrypted data.
And 4, step 4: and obtaining the model platform after the training is finished.
It should be noted that the first anti-addiction auxiliary model in the embodiment of the present specification may utilize data of the payment platform and one or more entertainment platforms at the same time, where the payment platform has rich payment account information and payment data, and the entertainment platform has a behavior of the entertainment account in the entertainment platform, so as to solve a data island problem, make the depiction of the user more perfect, and make the prediction result more accurate. Moreover, the first anti-addiction auxiliary model is not trained by the payment platform or one or more entertainment platforms, but utilizes more data, so that better effect can be achieved. Meanwhile, the embodiment of the specification can ensure that the data of the payment platform and one or more entertainment platforms cannot have the risk of privacy disclosure through the encryption technology of federal study.
S206, collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform.
In one or more embodiments of the present disclosure, the step is the same as S104 of the previous embodiment, and is not described again.
S208: detecting whether the entertainment account initiates a payment behavior through the payment channel on the appointed entertainment platform, if so, executing S210; if not, the process continues to S208.
In one or more embodiments of the present specification, the transaction initiated through the payment channel may be a payment behavior performed by the user through a payment platform at a specified entertainment platform, or a payment page that the user jumps to the payment platform at the specified entertainment platform.
S210: and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model, and if so, performing corresponding anti-addiction treatment.
In one or more embodiments of the present disclosure, the first anti-addiction auxiliary model is trained according to service data of one or more entertainment platforms and a payment platform, but when the first anti-addiction auxiliary model is applied, it may be determined whether a user corresponding to an entertainment account is a minor according to related data of the entertainment account on a designated entertainment platform, that is, according to the service data of the designated entertainment platform, or it may be determined whether the user corresponding to the entertainment account is a minor according to related data of the entertainment account on the designated entertainment platform and related data of the payment account on the payment platform, that is, according to the service data of the designated entertainment platform and the service data of the payment platform, that is, it is determined whether the user corresponding to the entertainment account is a minor. If not, the anti-addiction treatment is not required.
Further, for S210, referring to fig. 4 and 5, a schematic diagram of payment for an adult by a user corresponding to an entertainment account and a schematic diagram of payment for a minor by a user corresponding to the entertainment account are respectively shown, through S208, it is detected that the user initiates a payment behavior on the entertainment platform a, the user selects the payment platform B to make payment, and then transfers to the payment platform B to make payment, if it is determined that the user corresponding to the entertainment account is not a minor (is an adult), the user may directly click the confirmation payment of fig. 4; if the user corresponding to the entertainment account is judged to be a minor, the page of the picture 5 pops up after the payment is confirmed by clicking, and at the moment, the user needs to carry out face verification.
Further, in one or more embodiments of the present specification, S210 may specifically include: receiving, by the first anti-addiction auxiliary model, relevant data reflecting the entertainment account; extracting high-dimensional features reflecting expected irrational consumption by processing the relevant data at a hidden layer of the first anti-addiction auxiliary model; and judging whether the user corresponding to the entertainment account is a minor or not according to the high-dimensional characteristics. In this case, a determination is made as to whether the user to which the entertainment account corresponds is a minor, substantially including high dimensional features of anticipated irrational consumption.
Further, the high-dimensional characteristic of the expected irrational consumption is that the entertainment account is over-consumed on the entertainment platform, which may mean that the entertainment account determines whether the user corresponding to the entertainment account is a minor when the payment is not completed, partial payment is completed, or the payment is completed.
If the high-dimensional characteristic of the expected irrational consumption is that the entertainment account has not finished paying, whether the user corresponding to the entertainment account is a minor is judged, and therefore the payment behavior of the minor to the entertainment platform can be effectively prevented. For example, a certain entertainment account pays 1 ten thousand yuan to the entertainment platform through the payment platform, and after the action is detected, the user corresponding to the entertainment account can be judged to be a minor, and the payment action is stopped.
If the high-dimensional characteristic of the expected irrational consumption is that the entertainment account is paid, whether the user corresponding to the entertainment account is a minor is judged, and thus the minor is not stopped from paying the entertainment platform in time. For example, a certain entertainment account can pay 1 ten thousand yuan to the entertainment platform through the payment platform, and after the action is detected, the user corresponding to the entertainment account can be judged to be a minor.
If the anticipated irrational consumption is that the entertainment account has completed partial payment, and the payment behaviors of the preset times and/or the payment reaches the preset amount according to the preset time, whether the user corresponding to the entertainment account is a minor is judged, so that the payment behaviors of the minor to the entertainment platform can be partially stopped. For example, when a certain entertainment account makes 3 payment within 3 minutes, and the entertainment account makes the 4 th payment, it may be determined that the user corresponding to the entertainment account is a minor, and the entertainment account is prevented from making the 4 th payment.
Further, according to the high-dimensional feature, determining whether the user corresponding to the entertainment account is a minor specifically includes: determining whether the entertainment account is an adult of irrational consumption by one or more entertainment platforms based on the high-dimensional features; and when the entertainment account is judged not to be an adult of irrational consumption of one or more entertainment platforms, judging whether the user corresponding to the entertainment account is a minor or not according to the high-dimensional characteristics.
It should be noted that the adult who consumes irrational may include an adult who consumes without restriction, and may be understood as a krypton player such as "local tyrant", and the high-dimensional feature that expects irrational consumption may include the time when the entertainment account number is on line with the entertainment platform, the frequency of payment, the time of payment, and the like, for example, it is detected that the time when the entertainment account number is on line with the entertainment platform is mostly during school of students, the frequency of payment is maintained at 2-3 times a week, and the like, and due to the activity time limitation of minors, it may be determined that the user corresponding to the entertainment account is an adult who consumes irrational. Since the minor may use the mobile phone of the parent, the adult who is not rationally consumed may be eliminated first, and then it is further determined whether the minor is an adult who is rationally consumed or a minor. Since the adults who consume rationally mostly do not consume irrational, whether the user corresponding to the entertainment account is a minor can be judged according to the high-dimensional characteristics of expected irrational consumption.
S212: and determining a second anti-addiction auxiliary model, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the business data of one or more entertainment platforms.
In one or more embodiments of the present disclosure, before performing this step, the embodiment of the present disclosure further needs to train a second anti-addiction auxiliary model, and the specific steps may include:
(1) the method comprises the steps of obtaining a first data set which is constructed in advance and constructing a second data set, wherein the second data set comprises business data and tag data of one or more entertainment platforms, and the business data of the one or more entertainment platforms are sample data.
Note that the tag data may include minor and adult. When the label data is marked, the label data can be marked according to actual conditions.
(2) And establishing an initial second anti-addiction auxiliary model.
(3) And performing transfer training on the initial second anti-addiction auxiliary model according to the first anti-addiction auxiliary model and the second data set, determining parameters of the initial second anti-addiction auxiliary model, and accordingly obtaining a qualified second anti-addiction auxiliary model.
It should be noted that, in the step (3), second anti-addiction auxiliary model training may be performed according to a large amount of trained model data (data used in federal learning) and service data without a payment platform, the model training may select a tragaboost algorithm, and the base classifier may select supervised models such as XGBoost, lr, neural network, and the like. The specific training process of the model is as follows:
a. and setting the weight of the new and old data, and comprehensively considering the proportion of the new and old samples and the proportion of different classes in the new and old samples. Let m be the new sample number and n be the old sample number. In this scenario, the weights may be set to the new scene weights: 1/m; old scene weight: 1/n. (the setting of the initial weight is important, and grid search can be adopted to select the optimal weight);
b. setting an iteration round number N, for example, N is 200;
c. in each iteration, a base model ht is obtained according to the weight at the moment and a base classifier;
d. calculating the error of the model on the new sample
Figure BDA0002994703810000151
e. The new weight for each sample is re-determined based on the error. Let betat＝∈t/(1-∈t)；
Figure BDA0002994703810000152
f. Repeating c, d and e until the iteration number N is reached
g. And weighting according to the weight of each round and the basic model obtained by each round to obtain a second anti-addiction auxiliary model meeting the conditions.
S214: data related to an entertainment account on a designated entertainment platform is collected.
In one or more embodiments of the present disclosure, the step is the same as S104 of the previous embodiment, and is not described again.
S216: and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary system, and if so, performing corresponding anti-addiction treatment.
In one or more embodiments of the present specification, determining whether a user corresponding to the entertainment account is a minor according to the related data of the entertainment account and the second anti-addiction auxiliary system may specifically include: receiving, by the second anti-addiction auxiliary model, relevant data reflecting the entertainment account; extracting high-dimensional features reflecting expected irrational consumption and/or payment through payment accounts of other users by processing the relevant data at a hidden layer of the second anti-addiction auxiliary model; and judging whether the user corresponding to the entertainment account is a minor or not according to the high-dimensional characteristics.
The high-dimensional features of the expected irrational consumption are not described in detail in the same manner as S210 of the embodiment of the present specification.
The high dimensional features of making payments through the payment accounts of other users may be as follows:
the entertainment account does not have a payment account, and when payment is carried out by detecting the payment account of other users, whether the user corresponding to the entertainment account is a minor can be judged. For example, when a certain entertainment account is paid on the entertainment platform, the payment is made through the payment accounts of other users, and the user corresponding to the entertainment account can be determined to be a minor.
However, it is not possible to accurately determine whether the user corresponding to the entertainment account is a minor by only one of the high-dimensional features for which the non-rational consumption is expected to pay the payment account of the other user, so it is possible to determine whether the user corresponding to the entertainment account is a minor by combining the high-dimensional features for which the non-rational consumption is expected to pay the payment account of the other user, thereby improving the accuracy of the determination. For example, when the entertainment account does not have a payment account, the user is not necessarily a minor, or the payment platform selected by the user is not registered, and there is a preference when paying through the payment platform, but the user does not want to register, and instead pays through the payment account of another user.
It should be noted that, by the anti-addiction processing method, when the suspected immature is recharged and consumed, an effective interception effect can be achieved, specifically:
1. the juvenile is prevented from over-consuming, and the invalid expense of the family is reduced;
2. preventing unnecessary complaints caused by a large amount of recharge of minors;
3. the inhibition of the country to the game industry due to the massive recharge of immature fruits is prevented;
4. meanwhile, data of the payment platform and the entertainment platform are utilized, the payment platform has rich user information and payment data, and the entertainment platform has the behavior of the user in the entertainment platform, so that the problem of data island is solved, the depiction of the user is more perfect, and the prediction result is more accurate;
5. through the encryption technology of federal learning, the risk of privacy disclosure of a payment platform and an entertainment platform is avoided;
6. through transfer learning, prediction can be performed on non-payment platform users, and the identification coverage rate is improved;
7. the time and labor are not wasted, a large amount of new scene data are not accumulated, and the anti-addiction processing efficiency of the juveniles is improved.
In one or more embodiments of the present description, an interception means avoiding violence is also considered, since this means has the disadvantage of being palliative and not radical, and is liable to cause a counterpsychological reaction. Therefore, it is considered to prevent addiction by means having a radical cure effect.
In practical applications, the entertainment of the user is because the user can enjoy the entertainment, and if the enjoyment is reduced or even replaced by the pain, the user is naturally difficult to indulge in. Based on the above, the potential abilities of the first anti-addiction auxiliary model and the second anti-addiction auxiliary system are further mined to discover the fun points of the user with extra small cost, so that the small cost can be realized here because the first anti-addiction auxiliary model and the second anti-addiction auxiliary system are obtained by learning based on the related data of the entertainment account, the related data can truly reflect the game habits and behaviors of the user, and the game habits and behaviors can reflect the fun points of the user in the game, such as the nervous stimulation of the user who likes fighting, the recourse and decryption of the user who likes to be careful, the good watching of the public, and the like.
Further, the data related to the root entertainment account, and the first anti-addiction auxiliary model or the second anti-addiction auxiliary system, determine the type of the user's entertainment disorder, the type of the user's entertainment disorder corresponding to the fun points, and attempt to cut down the corresponding fun points with a certain type of entertainment disorder. Specifically, for example, according to the type of the entertainment barrier, a corresponding entertainment barrier is set for the user's entertainment item, and the entertainment barrier may include: handicapped opponent users (e.g., intentionally matching users with a super-high ranking of competing users to make the users difficult to win, thereby losing fun to combat), virtual handicapped viewers (e.g., frequently having the user drink a lottery with a virtual cheating viewer), virtual handicapped assistants (e.g., actively advancing the user's puzzle answers to the user's puzzle, causing the user to lose decryption fun), and the like.
Fig. 6 is a schematic structural diagram of an anti-addiction processing device for minors based on federal learning according to one or more embodiments of the present disclosure, where the device includes:
the first model determining unit 302 is used for determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained by federal learning training according to the respective service data of a payment platform and one or more entertainment platforms;
the first data acquisition unit 304 is used for acquiring related data of an entertainment account on a specified entertainment platform, and the specified entertainment platform provides a payment channel for the entertainment account through the payment platform;
the payment account determining unit 306 is used for judging whether the user corresponding to the entertainment account has a payment account on the payment platform;
if the first anti-addiction processing unit 308 exists, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model, and if the user is judged to be a minor, performing corresponding anti-addiction processing;
and if not, a second model determining unit 310 determines a second anti-addiction auxiliary model, and a second anti-addiction processing unit 312 determines whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary system, and performs corresponding anti-addiction processing if the user is a minor, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms.
There are also some optional units below, which are not shown in the figures.
Further, the apparatus further comprises:
the payment account determination unit 306, determining whether the registrant of the entertainment account has a payment account on the payment platform;
the first anti-addiction processing unit or the second anti-addiction processing unit determines whether the current user of the entertainment account is a minor.
Further, the first anti-addiction processing unit 308 or the second anti-addiction processing unit 312 determines the type of the user's entertainment disorder according to the data related to the entertainment account and the first anti-addiction auxiliary model or the second anti-addiction auxiliary system;
setting a corresponding entertainment barrier for the entertainment item of the user according to the type of the entertainment barrier, wherein the entertainment barrier comprises: obstacle opponent users, virtual obstacle spectators, virtual obstacle assistants.
Further, the apparatus further comprises:
a first model building unit 316, which builds a first training data set and builds an initial first anti-addiction auxiliary model, wherein the first training data set includes service data and label data of one or more entertainment platforms and payment platforms, respectively;
an alignment processing unit 318 that performs encryption-based sample alignment processing on the first training data set;
the first model training unit 320 obtains gradients and losses corresponding to one or more entertainment platforms and the payment platform according to the processed training data set, and determines parameters of the initial first anti-addiction auxiliary model, so as to obtain a qualified first anti-addiction auxiliary model.
Further, the apparatus further comprises:
the second model building unit 322 is configured to obtain a first data set which is pre-built, build a second data set, and build an initial second anti-addiction auxiliary model, where the second data set includes service data and tag data of one or more entertainment platforms;
the second model training unit 324 performs migration training on the initial second anti-addiction auxiliary model according to the first anti-addiction auxiliary model and the second data set, determines parameters of the initial second anti-addiction auxiliary model, and accordingly obtains a qualified second anti-addiction auxiliary model.
Further, the apparatus further comprises:
and a payment detection unit 326, which detects that the entertainment account initiates payment behavior through the payment channel at the specified entertainment platform.
Further, the first anti-addiction processing unit 306 specifically includes:
a first data receiving unit 328 for receiving data related to the entertainment account through the first anti-addiction auxiliary model;
a first feature extraction unit 330 that extracts high-dimensional features reflecting expected irrational consumption by processing the relevant data at a hidden layer of the first anti-addiction auxiliary model;
the first account determination unit 332 determines whether the user corresponding to the entertainment account is an minor or not according to the high-dimensional feature.
Further, the first account determination unit 332 specifically includes:
the first determination unit 334 is configured to determine whether the user corresponding to the entertainment account is a minor by analyzing whether the user corresponding to the entertainment account is an adult for irrational consumption of one or more entertainment platforms according to the high-dimensional features.
Further, the second anti-addiction processing unit 314 specifically includes:
a second data receiving unit 338, which receives the data related to the entertainment account through the second anti-addiction auxiliary model;
a second feature extraction unit 340 for extracting high-dimensional features reflecting expected irrational consumption and/or payment through payment accounts of other users by processing the related data at a hidden layer of the second anti-addiction auxiliary model;
the second account determination unit 342 determines whether the user corresponding to the entertainment account is an minor or not according to the high-dimensional feature.
Further, the service data of the payment platform comprises payment account information and payment information, and the service data of the one or more entertainment platforms comprises entertainment account information, behavior data and tag data; wherein the payment account information includes identity data; the payment information comprises one or more items of payment frequency and payment success rate; the behavior data comprises one or more of payment behavior, browsing behavior and account level; the tag data includes adults and minors.
Fig. 7 is a schematic structural diagram of an anti-addiction processing device for minors based on federal learning according to one or more embodiments of the present specification, where the device includes:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform;
judging whether a user corresponding to the entertainment account has a payment account on the payment platform;
if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model;
if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary system, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms;
and if the user is determined to be a minor, performing corresponding anti-addiction treatment.
One or more embodiments of the present specification provide a non-transitory computer storage medium storing computer-executable instructions configured to:
determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform;
judging whether a user corresponding to the entertainment account has a payment account on the payment platform;
if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model;
if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary system, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms;
and if the user is determined to be a minor, performing corresponding anti-addiction treatment.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functions of the various elements may be implemented in the same one or more software and/or hardware implementations of the present description.
As will be appreciated by one skilled in the art, the present specification embodiments may be provided as a method, system, or computer program product. Accordingly, embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.
The description has been presented with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the description. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, 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 also be noted that 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 phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the apparatus, the device, and the nonvolatile computer storage medium, since they are substantially similar to the embodiments of the method, the description is simple, and for the relevant points, reference may be made to the partial description of the embodiments of the method.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The above description is merely one or more embodiments of the present disclosure and is not intended to limit the present disclosure. Various modifications and alterations to one or more embodiments of the present description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of one or more embodiments of the present specification should be included in the scope of the claims of the present specification.

Claims (21)

1. A minor anti-addiction processing method based on federal learning comprises the following steps:
determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform;
judging whether a user corresponding to the entertainment account has a payment account on the payment platform;
if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model;
if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary model, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms;
and if the user is determined to be a minor, performing corresponding anti-addiction treatment.
2. The method according to claim 1, wherein the determining whether the user corresponding to the entertainment account has a payment account on the payment platform specifically includes:
determining whether a registrant of the entertainment account has a payment account on the payment platform;
the determining whether the user corresponding to the entertainment account is a minor specifically includes:
determining whether a current user of the entertainment account is a minor.
3. The method of claim 1, further comprising:
determining the type of the entertainment disorder of the user according to the related data of the entertainment account and the first anti-addiction auxiliary model or the second anti-addiction auxiliary model;
setting a corresponding entertainment barrier for the entertainment item of the user according to the type of the entertainment barrier, wherein the entertainment barrier comprises: obstacle opponent users, virtual obstacle spectators, virtual obstacle assistants.
4. The method of claim 1, prior to determining the first anti-addiction auxiliary model, the method further comprising:
constructing a first training data set and establishing an initial first anti-addiction auxiliary model, wherein the first training data set comprises business data and label data of one or more entertainment platforms and payment platforms respectively;
performing encryption-based sample alignment processing on the first set of training data;
and obtaining the corresponding gradient and loss of one or more entertainment platforms and the payment platform according to the processed training data set, and determining the parameters of the initial first anti-addiction auxiliary model, thereby obtaining a first anti-addiction auxiliary model meeting the conditions.
5. The method of claim 3, prior to determining the first anti-addiction auxiliary model, the method further comprising:
acquiring a first data set constructed in advance and constructing a second data set, wherein the second data set comprises business data and tag data of one or more entertainment platforms;
establishing an initial second anti-addiction auxiliary model;
and performing transfer training on the initial second anti-addiction auxiliary model according to the first anti-addiction auxiliary model and the second data set, determining parameters of the initial second anti-addiction auxiliary model, and accordingly obtaining a qualified second anti-addiction auxiliary model.
6. The method of claim 1, the determining whether the user to which the entertainment account corresponds is a minor, the method further comprising:
and detecting that the entertainment account initiates payment behavior at the designated entertainment platform through the payment channel.
7. The method according to claim 1, wherein the determining whether the user corresponding to the entertainment account is a minor according to the data related to the entertainment account and the first anti-addiction auxiliary model specifically includes:
receiving, by the first anti-addiction auxiliary model, data related to the entertainment account;
extracting high-dimensional features reflecting expected irrational consumption by processing the relevant data at a hidden layer of the first anti-addiction auxiliary model;
and judging whether the user corresponding to the entertainment account is a minor or not according to the high-dimensional characteristics.
8. The method according to claim 7, wherein the determining whether the user corresponding to the entertainment account is a minor according to the high-dimensional feature specifically comprises:
and according to the high-dimensional characteristics, judging whether the user corresponding to the entertainment account is a minor person or not by analyzing whether the user corresponding to the entertainment account is an adult person who is consumed by one or more entertainment platforms irrational.
9. The method according to claim 3, wherein the determining whether the user corresponding to the entertainment account is a minor according to the data related to the entertainment account and the second anti-addiction auxiliary model specifically includes:
receiving, by the second anti-addiction auxiliary model, data related to the entertainment account;
extracting high-dimensional features reflecting expected irrational consumption and/or payment through payment accounts of other users by processing the relevant data at a hidden layer of the second anti-addiction auxiliary model;
and judging whether the user corresponding to the entertainment account is a minor or not according to the high-dimensional characteristics.
10. The method according to any one of claims 1 to 9, wherein the service data of the payment platform comprises payment account information and payment information, and the service data of the one or more entertainment platforms comprises entertainment account information, behavior data and tag data; wherein the content of the first and second substances,
the payment account information includes identity data;
the payment information comprises one or more items of payment frequency and payment success rate;
the behavior data comprises one or more of payment behavior, browsing behavior and account level;
the tag data includes adults and minors.
11. An anti-addiction processing device for minors based on federal learning, comprising:
the system comprises a first model determining unit, a second model determining unit and a third model determining unit, wherein the first model determining unit determines a first anti-addiction auxiliary model, and the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
the system comprises a first data acquisition unit, a second data acquisition unit and a payment processing unit, wherein the first data acquisition unit is used for acquiring related data of an entertainment account on a specified entertainment platform, and the specified entertainment platform provides a payment channel for the entertainment account through the payment platform;
the payment account determining unit is used for judging whether the user corresponding to the entertainment account has a payment account on the payment platform;
if the first anti-addiction processing unit exists, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model, and if the user is judged to be a minor, performing corresponding anti-addiction processing;
and if not, determining a second anti-addiction auxiliary model, judging whether a user corresponding to the entertainment account is a minor or not by the second anti-addiction processing unit according to the related data of the entertainment account and the second anti-addiction auxiliary model, and if so, performing corresponding anti-addiction processing, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms.
12. The apparatus of claim 11, the payment account determination unit to determine whether a registrant of the entertainment account has a payment account on the payment platform;
the first anti-addiction processing unit or the second anti-addiction processing unit determines whether the current user of the entertainment account is a minor.
13. The apparatus of claim 11, the first anti-addiction processing unit or the second anti-addiction processing unit determining the user's type of entertainment disorder based on the data associated with the entertainment account and the first anti-addiction auxiliary model or the second anti-addiction auxiliary model;
setting a corresponding entertainment barrier for the entertainment item of the user according to the type of the entertainment barrier, wherein the entertainment barrier comprises: obstacle opponent users, virtual obstacle spectators, virtual obstacle assistants.
14. The apparatus of claim 11, the apparatus further comprising:
the system comprises a first model building unit, a second model building unit and a third model building unit, wherein the first model building unit builds a first training data set and builds an initial first anti-addiction auxiliary model, and the first training data set comprises business data and label data of one or more entertainment platforms and payment platforms respectively;
an alignment processing unit which performs sample alignment processing based on encryption on the first training data set;
and the first model training unit is used for solving the corresponding gradient and loss of one or more entertainment platforms and the payment platform according to the processed training data set, determining the parameters of the initial first anti-addiction auxiliary model and accordingly obtaining the first anti-addiction auxiliary model meeting the conditions.
15. The apparatus of claim 13, the apparatus further comprising:
the second model building unit is used for obtaining a first data set which is built in advance, building a second data set and building an initial second anti-addiction auxiliary model, wherein the second data set comprises business data and label data of one or more entertainment platforms;
and the second model training unit is used for carrying out transfer training on the initial second anti-addiction auxiliary model according to the first anti-addiction auxiliary model and the second data set, determining parameters of the initial second anti-addiction auxiliary model and accordingly obtaining a second anti-addiction auxiliary model meeting conditions.
16. The apparatus of claim 11, the apparatus further comprising:
and the payment detection unit is used for detecting that the entertainment account initiates payment behavior at the appointed entertainment platform through the payment channel.
17. The device of claim 11, wherein the first anti-addiction processing unit comprises:
the first data receiving unit is used for receiving the related data of the entertainment account through the first anti-addiction auxiliary model;
a first feature extraction unit that extracts high-dimensional features reflecting expected irrational consumption by processing the relevant data at a hidden layer of the first anti-addiction auxiliary model;
and the first account judging unit is used for judging whether the user corresponding to the entertainment account is a minor or not according to the high-dimensional characteristics.
18. The apparatus according to claim 17, wherein the first account determination unit specifically includes:
and the first judging unit judges whether the user corresponding to the entertainment account is a minor or not by analyzing whether the user corresponding to the entertainment account is an adult for irrational consumption of one or more entertainment platforms according to the high-dimensional characteristics.
19. The device of claim 13, wherein the second anti-addiction processing unit comprises:
the second data receiving unit is used for receiving the related data of the entertainment account through the second anti-addiction auxiliary model;
a second feature extraction unit which extracts high-dimensional features reflecting expected irrational consumption and/or payment through payment accounts of other users by processing the relevant data at a hidden layer of the second anti-addiction auxiliary model;
and the second account judging unit is used for judging whether the user corresponding to the entertainment account is a minor or not according to the high-dimensional characteristics.
20. The device of any one of claims 11 to 19, wherein the service data of the payment platform comprises payment account information and payment information, and the service data of the one or more entertainment platforms comprises entertainment account information, behavior data and tag data; wherein the content of the first and second substances,
the payment account information includes identity data;
the payment information comprises one or more items of payment frequency and payment success rate;
the behavior data comprises one or more of payment behavior, browsing behavior and account level;
the tag data includes adults and minors.
21. A federal learning-based anti-addiction processing device for minor, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
determining a first anti-addiction auxiliary model, wherein the first anti-addiction auxiliary model is obtained through federal learning training according to business data of a payment platform and one or more entertainment platforms;
collecting related data of an entertainment account on a designated entertainment platform, wherein the designated entertainment platform provides a payment channel for the entertainment account through the payment platform;
judging whether a user corresponding to the entertainment account has a payment account on the payment platform;
if yes, judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the first anti-addiction auxiliary model;
if not, determining a second anti-addiction auxiliary model, and judging whether the user corresponding to the entertainment account is a minor or not according to the related data of the entertainment account and the second anti-addiction auxiliary model, wherein the second anti-addiction auxiliary model is obtained through transfer learning training according to the first anti-addiction auxiliary model and the service data of one or more entertainment platforms;
and if the user is determined to be a minor, performing corresponding anti-addiction treatment.
CN202110326043.9A 2021-03-26 2021-03-26 Minor anti-addiction processing method, device and equipment based on federal learning Active CN113159782B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110326043.9A CN113159782B (en) 2021-03-26 2021-03-26 Minor anti-addiction processing method, device and equipment based on federal learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110326043.9A CN113159782B (en) 2021-03-26 2021-03-26 Minor anti-addiction processing method, device and equipment based on federal learning

Publications (2)

Publication Number Publication Date
CN113159782A CN113159782A (en) 2021-07-23
CN113159782B true CN113159782B (en) 2022-04-15

Family

ID=76884925

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110326043.9A Active CN113159782B (en) 2021-03-26 2021-03-26 Minor anti-addiction processing method, device and equipment based on federal learning

Country Status (1)

Country Link
CN (1) CN113159782B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113694537B (en) * 2021-08-06 2023-05-16 完美世界(北京)软件科技发展有限公司 Game account control method, device and equipment

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN207520591U (en) * 2017-11-28 2018-06-22 信利光电股份有限公司 A kind of indulging system and smart machine
CN109214444A (en) * 2018-08-24 2019-01-15 小沃科技有限公司 Game Anti-addiction decision-making system and method based on twin neural network and GMM
CN109492420A (en) * 2018-12-28 2019-03-19 深圳前海微众银行股份有限公司 Model parameter training method, terminal, system and medium based on federation's study
CN109522069A (en) * 2018-10-26 2019-03-26 深圳壹账通智能科技有限公司 The method, apparatus and electronic equipment of Anti-addiction game
CN111126609A (en) * 2019-12-20 2020-05-08 深圳前海微众银行股份有限公司 Knowledge migration method, device, equipment and medium based on federal learning
CN112150221A (en) * 2020-11-25 2020-12-29 支付宝(杭州)信息技术有限公司 Live broadcast room service processing method, device and equipment based on federal learning

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150039517A1 (en) * 2013-08-05 2015-02-05 Mozido, Inc. Cloud entertainment platform

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN207520591U (en) * 2017-11-28 2018-06-22 信利光电股份有限公司 A kind of indulging system and smart machine
CN109214444A (en) * 2018-08-24 2019-01-15 小沃科技有限公司 Game Anti-addiction decision-making system and method based on twin neural network and GMM
CN109522069A (en) * 2018-10-26 2019-03-26 深圳壹账通智能科技有限公司 The method, apparatus and electronic equipment of Anti-addiction game
CN109492420A (en) * 2018-12-28 2019-03-19 深圳前海微众银行股份有限公司 Model parameter training method, terminal, system and medium based on federation's study
CN111126609A (en) * 2019-12-20 2020-05-08 深圳前海微众银行股份有限公司 Knowledge migration method, device, equipment and medium based on federal learning
CN112150221A (en) * 2020-11-25 2020-12-29 支付宝(杭州)信息技术有限公司 Live broadcast room service processing method, device and equipment based on federal learning

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Record and Reward Federated Learning;Ismael Martinez;《IEEE》;20191231;全文 *
基于进程行为的防沉迷模型;杨一涛;《南昌大学学报(理科版)》;20130625;全文 *

Also Published As

Publication number Publication date
CN113159782A (en) 2021-07-23

Similar Documents

Publication Publication Date Title
Budish The economic limits of bitcoin and the blockchain
US20170250796A1 (en) Trans Vernam Cryptography: Round One
CN112150221B (en) Live broadcast room service processing method, device and equipment based on federal learning
Stark Making sense of ethereum’s layer 2 scaling solutions: State channels, plasma, and truebit
CN111460511A (en) Federal learning and virtual object distribution method and device based on privacy protection
CN111784345B (en) Payment processing method, device, equipment and system
CN109118053B (en) Method and device for identifying card stealing risk transaction
US20230070586A1 (en) Methods for Evolution of Tokenized Artwork, Content Evolution Techniques, Non-Fungible Token Peeling, User-Specific Evolution Spawning and Peeling, and Graphical User Interface for Complex Token Development and Simulation
Skinner ValueWeb: How fintech firms are using bitcoin blockchain and mobile technologies to create the Internet of value
CN110046932A (en) A kind of data processing method, device and equipment
CN104423945A (en) Information processing method and electronic device
Saleh et al. Islamic approach toward purification of transaction with cryptocurrency
CN113159782B (en) Minor anti-addiction processing method, device and equipment based on federal learning
Benetka et al. Understanding context for tasks and activities
Tokarčíková et al. Diffusion of innovation: The case of the Slovak mobile communication market
Lee et al. Self-knowledge distillation via dropout
Verhoeven et al. Australian films at large: expanding the evidence about Australian cinema performance
Chen et al. Understanding big data: Data calculus in the digital era
CN106302367B (en) Transaction methods and system
Thomason et al. Advancements in the New World of Web 3: A Look Toward the Decentralized Future: A Look Toward the Decentralized Future
CN114240121A (en) Resource allocation processing method and device
Sandbukt Top-Up with Driver: Digital Money, Transactional Aspirations, and Peerhood in Yogyakarta, Indonesia
CN110163482A (en) Determine method, terminal device and the server of the safety approach data of activity scheme
Howard Democratic Futures and the Internet of Things: How Information Infrastructure Will Become a Political Constitution
Liow et al. Customer Outcome Framework for Blockchain-Based Mobile Phone Applications

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