WO2017190670A1 - 一种账号申诉处理方法及服务器 - Google Patents

一种账号申诉处理方法及服务器 Download PDF

Info

Publication number
WO2017190670A1
WO2017190670A1 PCT/CN2017/083050 CN2017083050W WO2017190670A1 WO 2017190670 A1 WO2017190670 A1 WO 2017190670A1 CN 2017083050 W CN2017083050 W CN 2017083050W WO 2017190670 A1 WO2017190670 A1 WO 2017190670A1
Authority
WO
WIPO (PCT)
Prior art keywords
account
appeal data
account appeal
data
feature content
Prior art date
Application number
PCT/CN2017/083050
Other languages
English (en)
French (fr)
Inventor
刘杰
Original Assignee
腾讯科技（深圳）有限公司
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 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Priority to EP17792496.6A priority Critical patent/EP3454503B1/en
Priority to JP2018557835A priority patent/JP6707147B2/ja
Priority to KR1020187034329A priority patent/KR102218506B1/ko
Publication of WO2017190670A1 publication Critical patent/WO2017190670A1/zh
Priority to US15/976,451 priority patent/US10567374B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/08—Network architectures or network communication protocols for network security for authentication of entities
    • H04L63/083—Network architectures or network communication protocols for network security for authentication of entities using passwords
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G06F18/24155—Bayesian classification
    • 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
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
    • H04L9/3226—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials using a predetermined code, e.g. password, passphrase or PIN
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2221/00—Indexing scheme relating to security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F2221/21—Indexing scheme relating to G06F21/00 and subgroups addressing additional information or applications relating to security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F2221/2131—Lost password, e.g. recovery of lost or forgotten passwords

Definitions

  • the present application relates to the field of network technologies, and in particular, to an account appeal processing method and a server.
  • Account appeal refers to a service that the account official provides for the account owner to retrieve the account in case the account is stolen or the account password is forgotten.
  • the account owner needs to submit the account appeal data.
  • the account may be determined to belong to the account owner, and the account owner may have the right to modify the account password to reach the account owner. The purpose of retrieving the account.
  • the submitted data When submitting account appeal data, the submitted data is often incomplete and may require the account owner to submit two or more account appeal data.
  • user data leakage is more common, so the hacker may steal the user information of the account owner when stealing the account owner's account.
  • the hacker In order to achieve the purpose of controlling the stolen account, the hacker often also makes an account appeal and submits the account appeal data.
  • the multiple account appeal data may be submitted by different natural persons.
  • the account appeal data submitted by the account owner may exist in the multiple account appeal data, and there is also a hacking number.
  • the account appeal data when reviewing the account appeal data, it is mainly to collect multiple account appeal data of the same account, and then distribute it to the customer service or audit system for review.
  • the multiple account appeal data of the same account collected is submitted by different natural persons, such as the account appeal data submitted by the account owner and the account appeal data submitted by the hacker, then multiple copies based on the same account.
  • the audit result may be inaccurate, and the account attribution may be judged to the non-account owner.
  • the embodiment of the present application provides a method for processing an account appeal, including:
  • the embodiment of the present application further provides a server, including:
  • the account appeal data obtaining module is configured to obtain first account appeal data of the first account submitted by the first user, and the account appeal data includes at least one feature content;
  • a correlation determining module configured to compare the feature content of the first account appeal data with the feature content of the second account appeal data when there is second account appeal data corresponding to the first account, and determine Determining the correlation between the first account appeal data and the second account appeal data;
  • the association determining module is configured to determine that the first account appeal data and the second account appeal data are associated when the correlation meets a preset related condition
  • the account attribution determining module determines, according to one or more of the first account appeal data and the second account appeal data, that the first account belongs to the first user;
  • a communication module configured to send a password of the first account to the first user or receive a password of the first account input by the first user, so that the first user operates the password by using the password An account.
  • the embodiment of the present application further provides a server, including a processor, a memory, a communication interface, and a communication bus, wherein the processor, the communication interface, and the memory communicate through the communication bus, the processor Execute the program stored in memory to execute:
  • the communication interface is configured to: when determining that the first account belongs to the first user, send a password of the first account to the first user or receive the first user input a password of an account to enable the first user to operate the first account by using the password.
  • the server may obtain the first account appeal data of the first account submitted by the first user, and compare the feature content with the second account appeal data of the recorded first account. Determining the correlation between the first account appeal data and the second account appeal data, the correlation indicating the possibility that the first account appeal data and the second account appeal data are associated, and presetting multiple accounts of the same account.
  • the preset relevant condition corresponding to the appeal data is associated, so that when the correlation meets the preset relevant condition, determining that the first account appeal data is associated with the second account appeal data, and implementing an account appeal belonging to the same natural person
  • the determination of the data in order to assist the account attribution judgment, improve the accuracy of the account attribution judgment result.
  • FIG. 1 is a structural block diagram of an account appeal processing system according to an embodiment of the present application.
  • FIG. 3 is a flowchart of a method for processing an account appeal according to an embodiment of the present application
  • FIG. 4 is a schematic diagram of a part of feature content corresponding to a social type account for an account appeal
  • FIG. 5 is another flowchart of an account appeal processing method according to an embodiment of the present application.
  • FIG. 6 is still another flowchart of an account appeal processing method according to an embodiment of the present application.
  • FIG. 7 is still another flowchart of a method for processing an account appeal according to an embodiment of the present application.
  • FIG. 8 is a flowchart of a pre-training method of a Bayesian classification classification model according to an embodiment of the present application.
  • FIG. 9 is a flowchart of determining correlation between account appeal data of each account according to an embodiment of the present application.
  • FIG. 10 is still another flowchart of the method for processing an account appeal according to an embodiment of the present application.
  • FIG. 11 is a structural block diagram of a server according to an embodiment of the present application.
  • FIG. 12 is a structural block diagram of a correlation determining module according to an embodiment of the present application.
  • FIG. 13 is another structural block diagram of a server according to an embodiment of the present application.
  • FIG. 14 is a structural block diagram of a correlation determination execution unit according to an embodiment of the present application.
  • FIG. 15 is a block diagram of still another structure of a server according to an embodiment of the present application.
  • 16 is another block diagram of a correlation determining execution unit provided by an embodiment of the present application.
  • FIG. 17 is still another structural block diagram of a server according to an embodiment of the present application.
  • FIG. 18 is a block diagram showing the hardware structure of a server according to an embodiment of the present application.
  • FIG. 1 is a structural block diagram of an account appeal processing system according to an embodiment of the present application.
  • the account appeal processing method provided by the embodiment of the present application can be implemented based on the system shown in FIG. 1.
  • an account appeal processing system provided by an embodiment of the present application may include a server 10 and at least one terminal 20.
  • the server 10 may be configured for processing the account appeal data submitted by the terminal for the network side. device.
  • the server 10 can be a single server or a server group composed of multiple servers.
  • the terminal 20 is a device for submitting account appeal data on the user side, such as a mobile phone, a tablet computer, a notebook computer, and the like.
  • the account appeal data submitted by the terminal 20 may be submitted by the account owner, or may be submitted by the hacker.
  • the server needs to process the account appeal data submitted by the terminal 20 for the same account, determine the associated account appeal data, and implement the judgment of the account appeal data submitted by the same natural person, so as to assist in determining the attribution of the account. , to improve the possibility of account ownership judgment to the account owner.
  • FIG. 2 shows a signaling flowchart of the account appeal processing method provided by the embodiment of the present application. As shown in FIG. 1 and FIG. 2, the flow may include the following steps S10-S15.
  • Step S10 The first user submits the first account appeal data of the first account to the server through the terminal.
  • the first account is an account that needs to be retrieved in the embodiment of the present application.
  • the user of the terminal may be the owner of the first account or may be the hacker.
  • the account appeal data has at least one feature content, and the feature content may be a specific appeal data in the account appeal data.
  • Step S11 The server acquires the first account appeal data.
  • the first account appeal data may have at least one feature content.
  • Step S12 When the server determines that the second account appeal data of the first account exists, compares the first account appeal data with the feature content of the second account appeal data, and determines the first account appeal data and the second account appeal data. Relevance.
  • the second account appeal data may be the account appeal data of the first account submitted by the second user of the server history.
  • the second account appeal data may be generated based on the account appeal data uploaded by the owner of the first account, or may be based on the hacker The uploaded account appeal data is generated. Therefore, the purpose of the embodiment of the present application is to determine whether the first account appeal data and the second account appeal data for the first account are associated, that is, whether the first user and the second user belong to the same natural person, so as to assist according to the judgment result. Determine the attribution of the first account.
  • the embodiment of the present application may separately compare the feature content of the first account appeal data and the second account appeal data, and determine the comparison result of the feature content, based on the first The comparison result of the feature content of the account appeal data and the second account appeal data, the embodiment of the present application may determine the correlation between the first account appeal data and the second account appeal data, and the correlation may be used to indicate the first account appeal data.
  • the possibility associated with the second account appeal data that is, the possibility that the first user submitting the first account appeal data and the second user submitting the second account appeal data belong to the same natural person.
  • Step S13 The server determines that the first account appeal data and the second account appeal data are associated when determining that the correlation meets a preset related condition.
  • the preset related condition may indicate that when multiple account appeal data of the same account are associated (if multiple account appeal data of the same account belong to the same natural person), the target correlation between the corresponding account appeal data;
  • the application embodiment may determine that the first account appeal data and the second account appeal data are associated when the correlation between the first account appeal data and the second account appeal data meets the preset relevant conditions, and implement the submitting the first account appeal data.
  • the determination of the first user and the second user submitting the second account appeal data belong to the same natural person.
  • Step S14 when the server determines that the first account appeal data and the second account appeal data are associated, determining, according to one or more of the first account appeal data and the second account appeal data, The first account belongs to the first user.
  • Step S15 The server sends the password of the first account to the first user or receives the password of the first account input by the first user, so that the first user operates by using the password. First account.
  • the server may obtain the first account appeal data of the first account, and compare the feature content with the second account appeal data of the first account that has been recorded, and determine the first account appeal. Correlation between the data and the second account appeal data, the correlation indicates the possibility that the first account appeal data and the second account appeal data are associated, and is corresponding to when multiple account appeal data of the same account are associated Presetting the relevant condition, so that when the correlation meets the preset relevant condition, determining that the first account appeal data is associated with the second account appeal data, that is, submitting the first user of the first account appeal data and submitting the first
  • the second user of the second account appeal data belongs to the same natural person, thereby assisting the attribution judgment of the account, and improving the accuracy of the account attribution judgment result.
  • the account appeal processing method provided by the embodiment of the present application is introduced from the perspective of the server.
  • the account appeal processing method described below may be cross-referenced with the signaling flow content described above.
  • FIG. 3 is a flowchart of a method for processing an account appeal according to an embodiment of the present application.
  • the method can be applied to the server.
  • the server can be an officially set server for performing an account appeal, and the server can collect the account appeal data submitted by the user.
  • the account appeal processing method provided by the embodiment of the present application may include the following steps S100-S140.
  • Step S100 Acquire first account appeal data of the first account submitted by the first user.
  • the first account is an account that needs to be retrieved in the embodiment of the present application.
  • the first account appeal data may be appeal data submitted by the first user to the server through the terminal for retrieving the first account.
  • the first user may be the owner of the first account or may be the hacker.
  • the account appeal data may include at least one feature content, where the feature content may be a specific appeal data in the account appeal data.
  • a feature content may correspond to a specific content filled in by the item when the account is appealed.
  • the terminal can display at least one item to be filled out.
  • fill in the items such as contact information (such as mobile phone, email address, etc.), personal information (such as name, ID card, address, etc.), usage information (such as used passwords, security issues, bundled phones, etc.).
  • the feature content may further include: the terminal automatically extracting the terminal identifier, the network IP, and the like carried in the account appeal data.
  • the items to be filled in different account types may be different.
  • the items to be filled in the social type account may have: contact information, personal information, usage information, social relationships (such as friend information, etc.);
  • the required items for the type account can have: contact information, personal information, usage information, game role name, etc.; therefore, when the account is appealed, the specific required items can be determined according to the actual account type and application scenario, and are not strict. limits.
  • FIG. 4 shows part of the feature content corresponding to the social type account when the account appeal is made.
  • Step S110 When there is the second account appeal data of the first account, compare the first account appeal data with the feature content of the second account appeal data, and determine the correlation between the first account appeal data and the second account appeal data. .
  • the second account appeal data may be the account appeal data of the first account of the server history.
  • the time when the second account appeal data is submitted to the server may be earlier than the time when the first account appeal data is submitted to the server, and the second account appeal data may also be generated based on the account appeal data submitted earlier than the first account appeal data.
  • the second account appeal data may be generated based on the account appeal data uploaded by the owner of the first account, or may be generated based on the account appeal data uploaded by the hacker. Therefore, the purpose of the embodiment of the present application is to determine whether the first account appeal data and the second account appeal data for the first account are associated, so as to assist in determining the attribution of the first account according to the judgment result.
  • the server may determine the first account appeal data and the second by processing the specific content of the first account appeal data and the second account appeal data.
  • the possibility of associated account appeal data called first Correlation between account appeal data and second account appeal data).
  • the feature content of the first account appeal data and the second account appeal data may be compared to determine a comparison result of the feature content. If the first account appeal data and the second account appeal data both have the feature content of the contact mode and the feature content of the personal information, the embodiment of the present application can compare the contact information of the first account appeal data and the second account appeal data respectively. The characteristic content, and the characteristic content of the personal information, thereby obtaining the comparison result of each feature content.
  • the correlation between the first account appeal data and the second account appeal data may be determined.
  • the correlation may be a probability associated with the first account appeal data and the second account appeal data.
  • Step S120 When the correlation meets the preset related condition, determine that the first account appeal data is associated with the second account appeal data.
  • the corresponding preset relevant conditions when the multiple account appeal data of the same account is associated, the corresponding preset relevant conditions (pre-set relevant conditions, such as when multiple account appeal data of the same account are associated, a target correlation between the corresponding account appeal data, the target relevance may represent a target possibility corresponding to the multiple account appeal data of the same account, so that the correlation is consistent with the preset correlation
  • the condition is determined, the possibility that the first account appeal data and the second account appeal data are associated is not lower than the target possibility, and the determination of the first account appeal data and the second account appeal data is determined.
  • the embodiment of the present application may set a target probability corresponding to multiple account appeal data of the same account, and thereby determine a probability associated with the first account appeal data and the second account appeal data, and the probability When the target probability is not lower than the target probability, it is determined that the first account appeal data is associated with the second account appeal data.
  • Step S130 according to the first account appeal data and the second account appeal data One or more of the first account is determined to belong to the first user.
  • Step S140 Send a password of the first account to the first user or receive a password of the first account input by the first user, so that the first user operates the password by using the password.
  • An account Send a password of the first account to the first user or receive a password of the first account input by the first user, so that the first user operates the password by using the password.
  • the account appeal processing method includes: the server acquiring the first account appeal data of the first account submitted by the first user, where the account appeal data includes at least one feature content; when the first account exists When the second account appeal data is used, the first account appeal data and the second account appeal data are compared, and the correlation between the first account appeal data and the second account appeal data is determined; when the correlation meets the preset Determining, by the relevant condition, the first account appeal data and the second account appeal data; determining, according to one or more of the first account appeal data and the second account appeal data, An account belongs to the first user and sends a password of the first account to the first user or receives a password of the first account input by the first user, so that the first user passes The password operates the first account.
  • the server may obtain the first account appeal data of the first account submitted by the first user, and compare the feature content with the second account appeal data of the first account that has been recorded, and determine the first Correlation between the account appeal data and the second account appeal data, the correlation indicates the possibility that the first account appeal data and the second account appeal data are associated, and is associated with multiple account appeal data of the same account. Corresponding preset related conditions, so that when the correlation meets the preset related condition, determining that the first account appeal data is associated with the second account appeal data, and determining the account appeal data belonging to the same natural person, This assists in the attribution of the account and improves the accuracy of the account attribution judgment result.
  • the first account appeal data and the second account appeal data may be integrated, so as to subsequently determine the first account in the audit.
  • the comprehensive first account appeal data and the second account can be obtained.
  • the appeal data improves the probability that the first account belongs to the account owner, and improves the accuracy of the first account attribution judgment.
  • the first account appeal data and the second account appeal data may be considered as not associated, and the subsequent account appeal data is At the same time, the first account appeal data and the second account appeal data can be distinguished and reviewed.
  • the embodiment of the present application may perform the difference comparison of the feature content to determine the feature content.
  • the difference level between the various characteristics is based on the difference level between the feature contents, and the difference between the first account appeal data and the second account appeal data is comprehensively analyzed to determine the correlation between the first account appeal data and the second account appeal data. .
  • the difference level is the same as the feature content, the feature content is similar, the feature content is not similar, the feature content is completely different, and the like.
  • the difference level type is preferably two types of the same feature content and similar feature content. Obviously, based on different account appeal situations, the selected difference level may not be limited to the same two types of feature content and similar feature content.
  • FIG. 5 is another flowchart of the account appeal processing method provided by the embodiment of the present application. Referring to FIG. 5, the method may include the following steps S200-S250.
  • Step S200 Acquire first account appeal data of the first account submitted by the first user.
  • Step S210 When there is the second account appeal data of the first account, the feature content of the first account appeal data and the second account appeal data are separately compared and the difference level of each feature content is determined.
  • Step S220 Determine, by using a difference level of the feature content, a correlation between the first account appeal data and the second account appeal data.
  • the correlation may represent the first account appeal data and the second account appeal number. According to the associated possibilities.
  • Step S230 When the correlation meets the preset related condition, determine that the first account appeal data is associated with the second account appeal data.
  • the determined difference level of each feature content may include: the feature content is the same and the feature content is similar.
  • the same content content can be considered that the first account appeal data and the second account appeal data have the same feature content of the item.
  • a feature content similarly may be considered to be the same as the feature content of the item in the first account appeal data and the second account appeal data, and the same part range is not lower than the set range.
  • the same part of the range is not lower than the set range, for example, the number of characters in the same part is not less than the set proportion in the total number of characters of the feature content, or the number of characters in the same part is not lower than the set The number and so on.
  • the specific requirements of the same part range not lower than the set range may be determined according to actual conditions, and are not limited to the description in this paragraph.
  • the feature content of the social relationship in the first account appeal data is a1, a2, a3, a4, a5, a6 (referring to the friend nickname, etc.) Information
  • the feature content of the social relationship required in the second account appeal data is also a1, a2, a3, a4, a5, a6, because the feature content of the social relationship of the first account appeal data and the second account appeal data is completely If the same, it is considered that the difference level of the feature content of the social relationship required in the first account appeal data and the second account appeal data is the same as the feature content.
  • the feature content of the social relationship in the first account appeal data is a1, a2, a3, a4, a5, a6, the feature content of the social relationship in the second account appeal data is a1, a2, a3, a4 , a7; then the first account appeal data and the second account appeal in the social relationship need to fill in the feature content, there are 4 friends information is the same, not lower than the set ratio or set the number of settings (specific values can be According to the actual situation, it can be considered that the difference level of the feature content of the social relationship required in the first account appeal data and the second account appeal data is similar to the feature content.
  • Step S240 Determine, according to one or more of the first account appeal data and the second account appeal data, that the first account belongs to the first user.
  • Step S250 sending a password of the first account to the first user or receiving a password of the first account input by the first user, so that the first user operates the password by using the password.
  • whether the first account appeal data and the second account appeal data are strongly correlated or weak are determined by the determined difference level between the first account appeal data and the second account appeal data.
  • the strong correlation may be considered as the first correlation between the determined first account appeal data and the second account appeal data
  • the weak correlation may be considered as the determined first account appeal data and the second account appeal data.
  • the correlation is in accordance with the second correlation, and the first correlation is higher than the second correlation; thus, when it is determined that the first account appeal data and the second account appeal data are strongly correlated, the first account appeal data and the second account appeal are determined.
  • the correlation of the data conforms to the preset relevant condition, and the first account appeal data is associated with the second account appeal data, that is, belongs to the same natural person.
  • FIG. 6 is a flowchart of the account appeal processing method provided by the embodiment of the present application. Referring to FIG. 6, the method may include the following steps S300-S360.
  • Step S300 Acquire first account appeal data of the first account submitted by the first user.
  • Step S310 When there is the second account appeal data of the first account, the difference between the feature content of the first account appeal data and the second account appeal data is compared, and the difference level of the feature content is determined; the difference level includes: The feature content is the same and the feature content is similar.
  • Step S320 Determine a correlation between the first account appeal data and the second account appeal data by using the difference level of the feature content.
  • Step S330 when the correlation meets the first correlation, determining that the correlation meets a preset related condition; the correlation conforming to the first relevance includes: each feature content is the same; or, a part of the feature content is the same And another part of the feature content is similar, the type of the feature content with the same feature content conforms to the preset first type, and the type of the feature content with the similar feature content conforms to the preset second type.
  • the part of the feature content is the same and the other part of the feature content is similar.
  • the feature content in the first account appeal data and the second account appeal data is divided into two parts: the same and similar parts; the type of the feature content with the same feature content is consistent with the preset first type.
  • the type of the feature content with the same feature content should include at least the preset first type, that is, the type of the feature content with the same feature content should have at least the preset first type, and there may be other types.
  • the type of feature content with similar feature content conforms to the preset second type. It can be considered that the type of feature content with similar feature content can only be in the range of the preset second type, and should not exceed the range.
  • the preset first type may be at least one feature content type selected from the plurality of feature content types to be filled in.
  • the social type account is used as an example, and the preset first type may be a contact mode or a terminal identifier. , IP, personal information, etc.
  • the preset second type may be at least one feature content type selected from a plurality of feature content types to be filled out, such as usage materials, social relationships, and the like.
  • the types of the preset first type and the preset second type may be different or partially the same.
  • the correlation between the first account appeal data and the second account appeal data conforms to the first correlation may be:
  • the IP is the same (the IP of the account appeal data is the same),
  • the terminal identifier is the same (the identifier of the terminal that submits the account appeal data is the same),
  • the usage data is similar (for example, fill in 6 passwords for the first time, 5 passwords for the second time, and four of them are the same)
  • the correlation between the first account appeal data and the second account appeal data conforms to the first correlation may also be:
  • the terminal identifier is the same.
  • the correlation between the first account appeal data and the second account appeal data is consistent with the first correlation, and may be related to the mandatory first account appeal data set by the embodiment of the present application and the second account appeal data.
  • the specific content of the first relevance may depend on the actual account type and application scenario.
  • Step S340 Determine that the first account appeal data and the second account appeal data are associated.
  • Step S350 Determine, according to one or more of the first account appeal data and the second account appeal data, that the first account belongs to the first user.
  • Step S360 sending a password of the first account to the first user or receiving a password of the first account input by the first user, so that the first user operates the password by using the password.
  • the embodiment of the present application may further determine through training.
  • the algorithm model of the probability associated with the first account appeal data and the second account appeal data indicating the correlation between the first account appeal data and the second account appeal by the determined probability, realizing the first account appeal data and the second Whether the account appeal data is associated with the judgment.
  • the algorithm model may use a Bayesian classification classification model, and the Bayesian classification model is generated based on a Bayesian classification algorithm, which is a statistical classification model, which can be classified by using probability and statistics knowledge.
  • Bayesian classification algorithm can be compared with decision tree and neural network classification algorithm.
  • the algorithm can be applied to large databases, and the method is simple, the classification accuracy is high, and the speed is fast. Therefore, the embodiment of the present application can calculate the probability associated with the first account appeal data and the second account appeal data by using a pre-trained Bayesian classification model, and implementing the first account appeal by using the Bayesian classification classification model. Determination of the relevance of the data and the second account appeal data.
  • FIG. 7 is still another flowchart of the account appeal processing method provided by the embodiment of the present application.
  • the method shown in FIG. 7 mainly adopts the Bayesian classification classification model for account appeal processing, which can be implemented independently with the method of FIG. 6 using the first correlation for account appeal processing.
  • the method may include the following steps S400-S460.
  • Step S400 Acquire first account appeal data of the first account submitted by the first user.
  • Step S410 When there is the second account appeal data of the first account, the difference between the feature content of the first account appeal data and the second account appeal data is separately compared, and the difference level of each feature content is determined.
  • Step S420 Perform classification processing on the difference level of the feature content according to the pre-trained Bayesian classification classification model, and obtain a probability that the first account appeal data and the second account appeal data are associated.
  • the probability may represent a correlation between the first account appeal data and the second account appeal data.
  • the difference level here includes at least: the feature content is the same and the feature content is similar.
  • Step S430 When the probability meets the preset probability condition, determining that the correlation between the first account appeal data and the second account appeal data conforms to a preset relevant condition.
  • Step S440 determining that the first account appeal data and the second account appeal data are associated.
  • Step S450 Determine, according to one or more of the first account appeal data and the second account appeal data, that the first account belongs to the first user.
  • Step S460 the password of the first account is sent to the first user or the password of the first account input by the first user is received, so that the first user operates the password by using the password.
  • An account An account.
  • FIG. 8 shows a pre-training process of the Bayesian classification classification model.
  • the process may include the following steps S500-S530.
  • Step S500 Collecting account appeal data of multiple accounts, and having one account corresponding to multiple account appeal data.
  • the embodiment of the present application can collect the account appeal data of the massive account to perform the training of the subsequent Bayesian classification model.
  • the account appeal data of one account collected may have multiple copies, and multiple account appeal data corresponding to one account may be submitted by the account owner, or may be submitted by the hacker.
  • Step S510 For each account, compare each feature content of each account appeal data, and determine a difference level of each feature content of the account appeal data of each account.
  • the account appeal data of each account is processed as a unit, and the account appeal data of each account is processed to determine the difference level of each feature content of the account appeal data of each account.
  • the embodiment of the present application can compare the characteristics of each account appeal data, identify the difference level of each feature content of the account appeal data, and determine the feature content of the account appeal data. The specific process of the difference level can be described in the corresponding section above.
  • the embodiment of the present application is directed to multiple account appeal data of the same account, and may be a pair of account appeal data for comparing the feature contents, and the account appeal data integrated after the pairwise comparison may be further The content of each feature of the account appeal data of the same account is compared.
  • Step S520 Determine a correlation between account appeal data of each account by using a difference level of each feature content of the account appeal data of each account; the correlation includes a first correlation and a second correlation.
  • the first correlation indicates that there is a strong correlation between the account appeal data of the same account
  • the second correlation indicates that the account appeal data of the same account is weakly correlated.
  • the difference level includes the same feature content and the feature content is similar.
  • the first correlation may indicate that the feature content is the same; or, the feature content is the same, the other feature content is similar, and the feature content is the same.
  • the type of the content conforms to the preset first type, and the type of the feature content with similar feature content should conform to the second type.
  • the second correlation may represent that the content of each of the plurality of account appeal data of the same account is similar.
  • the correlation between the account appeal data of the account is the second correlation may be:
  • IP similar eg 1.2.3.4 and 1.2.3.5
  • Personal information is similar (for example, the first bill of lading name is Liu Jie, and the second bill of lading name is Jay)
  • the usage data is low similar (using the similar data can reach the first value as the same password, such as 4 identical, using the data low similarity can be the same as the password, the character is lower than the first value, and higher than the second value, the first value Greater than the second value, such as 2 in the password character)
  • Step S530 According to the correlation between the account appeal data of each account, and the collected account appeal data and the unrelated account appeal data, the Bayesian classification algorithm is used to train and obtain the Bayesian classification classification model. .
  • the embodiment of the present application can perform training based on the extracted positive and negative sample data by using a Bayesian classification algorithm.
  • the positive sample data can be considered as the collected related account appeal data, and the negative sample data can be regarded as the collected unrelated account appeal data; the positive and negative sample data can be obtained mainly through manual selection and user complaint data.
  • the associated account appeal data and the unrelated account appeal data are collected, and the association between the associated account appeal data and the unrelated account appeal data are analyzed. Correlation, so as to analyze the correlation between the associated account appeal data and the association between the unrelated account appeal data, and use the Bayesian classification algorithm to account for each account.
  • the correlation between the appeal data is trained, and an algorithm formula that can calculate the probability associated with multiple account appeal data of the same account is obtained, and the Bayesian classification classification model is trained.
  • the embodiment of the present application may numerically represent the difference level of each feature content of the account appeal data.
  • the process of determining the correlation between the account appeal data of each account in the method shown in FIG. 8 may be as shown in FIG. 9, and includes the following steps S600-S610.
  • Step S600 For each account, according to the difference level of each feature content of the account appeal data, define a difference representation value of each feature content between the account appeal data, and obtain a difference representation value group corresponding to the account appeal data.
  • the difference characterization value group may be based on a difference level of each feature content of the two-two account grievance data, and the defined difference characterization values are combined; the difference characterization value includes a first value indicating that the feature content is the same. And a second value indicating that the feature content is similar, the first value being different from the second value.
  • the first value may be 1 and the second value may be 0.
  • the specific values of the first value and the second value may be determined according to actual conditions, the first value is 1, and the second value is 0. Only optional.
  • the embodiment of the present application may indicate the difference level of each feature content by 1 and 0, and if the feature content of the comparison is the same, A difference characterization value of 1 is defined for the feature content. If a feature content is similar, a difference characterization value of 0 can be defined for the feature content.
  • the difference characterization value group 11100 corresponding to the account appeal data of the account is obtained;
  • IP is the same (definition difference characterization value 1)
  • the terminal identifier is the same (definition difference characterization value 1),
  • Step S610 For each account, according to the difference representation value group corresponding to the account appeal data, establish a difference characterization value table corresponding to each account, and use the difference characterization value table to indicate the correlation between the account appeal data of each account.
  • the difference characterization value table of the one account may have at least one difference characterization value group, and the difference characterization value group indicates a combination of the difference characterization values of the feature contents of the two account appeal data of the account.
  • the form of the difference characterization value table of an account is as follows, wherein 1 0 0 0 0 is a difference characterization value group, and 1 1 0 1 0 is another difference characterization value group:
  • the embodiment of the present application can obtain a matrix group composed of multiple matrices, so as to correlate the data between the analyzed related accounts.
  • the relevance between the sexual appeal and the unrelated account appeal data is a classification reference.
  • the Bayesian classification algorithm is used to train the correlation between the account appeal data of each account, and obtain multiple accounts that can calculate the same account.
  • the algorithm formula of the correlation probability of the appeal data realizes the training of the Bayesian classification classification model.
  • the embodiment of the present application may be based on the difference representation table corresponding to each account, and the associated account appeal data and the unrelated account appeal data collected by Bayesian.
  • the classification algorithm is trained to obtain a Bayesian classification classification model.
  • the specific content of the first correlation is mainly obtained according to the actual account appealing experience, and may be generated according to the experience of the account appeal processing expert in the actual account appealing work, for forcibly determining the first account appeal data and the The second account appeal data is associated, so according to the first relevance, the first account appeal data and the second account appeal data are associated with each other, and the account account data of the submitted account is relatively standardized, and the account has a higher account number.
  • the accuracy of the attribution of the appeal data is judged, and when the submitted account appeal data is not standardized, the accuracy of the attribution judgment of the appeal data by the first relevance is not very satisfactory.
  • the Bayesian classification classification model is based on the account appeal data generation of the massive account, which can handle the account appeal data of various situations more flexibly, and can respond to the account appeal data of the submitted account appeal data which is more standardized and not standardized.
  • the attribution judgment therefore, the Bayesian classification classification model can be used to flexibly realize the attribution judgment of the account appeal data, and the accuracy of the judgment result is also high.
  • the embodiment of the present application may independently perform the attribution determination of the account appeal data based on the first correlation or the Bayesian classification classification model; or may combine the first correlation and the Bayesian classification classification model to implement the account appeal. Judging the attribution of the data, if it is first determined based on the first correlation whether the first account appeal data and the second account appeal data are related, and if not, the first account appeal data and the second account appeal are determined based on the Bayesian classification classification model. Whether the data is related, so as to consider the overall situation of the account appeal data, and improve the accuracy of the account appeal data attribution judgment.
  • FIG. 10 shows still another flowchart of the account appeal processing method provided by the embodiment of the present application.
  • the method may include the following steps S700-S760.
  • Step S700 Acquire first account appeal data of the first account submitted by the first user.
  • Step S710 When there is the second account appeal data of the first account, the difference between the feature content of the first account appeal data and the second account appeal data is separately compared, and the difference level of each feature content is determined;
  • the difference level includes: the feature content is the same and the feature content is similar.
  • Step S720 When the determined correlation between the first account appeal data and the second account appeal data does not meet the first correlation, according to the pre-trained Bayesian classification model, the difference level of the feature content is performed.
  • the classification process obtains a probability that the first account appeal data and the second account appeal data are associated, and the probability of the first account appeal data and the second account appeal data is represented by the probability.
  • the correlation does not meet the first correlation, and the feature content is different, or the feature content of the first type is different, or the feature content of the second type is not similar.
  • the content of each feature may be different: the contact information is different, the IP is different, the terminal identifier is different, the personal information is different, the usage data is different, and the social relationship is different.
  • the preset first type of feature content may be different, such as: different contact methods, or different IPs, or different terminal identifiers, or different personal information; as for the difference level between the use of materials and social relationships, may be considered, or may be considered .
  • the preset second type of feature content is not similar, such as: the usage data is not similar, or the social relationship is not similar; as for the contact information, IP, terminal identification, and personal information, the difference level may be considered or considered.
  • Step S730 When the probability meets a preset probability condition, determine that the correlation meets a preset related condition.
  • the preset probability condition may be implemented by setting a target probability, and when the probability is not lower than the target probability, the probability may be considered to meet a preset probability condition.
  • Step S740 determining that the first account appeal data and the second account appeal data are associated.
  • the first account appeal data and the second account appeal data may be considered as not associated.
  • Step S750 Determine, according to one or more of the first account appeal data and the second account appeal data, that the first account belongs to the first user.
  • the first account appeal data and the second account appeal data may be integrated to obtain the integrated account appeal data, so that The account owner of the first account is judged by the integrated account appeal data.
  • the method for synthesizing the first account appeal data and the second account appeal data may be that the parts with the same feature content are merged into one part for the feature content, and the parts with different feature contents are different. Retain and combine with the same feature content after the combination; for the feature content of the usage data, the password for the first account appeal data is 123456, a total of 6 characters, and the password for the second account appeal data is 72345, a total of 5 Characters, where 4 characters are the same, the same 2345 can be combined into one, and for different characters 1, 6, 7 can be retained and combined with 2345 combined into one, resulting in 1 or 7-2345-6 ;
  • the second account appeal data may be an account appeal data submitted by the history for the first account; the second account appeal data may also be an account appeal processing method provided by the embodiment of the present application, and the history submitted is directed to the first
  • the account appeal data generated by the multiple account appeal data of the account is integrated; if the server history obtains the A and B account appeal data for the first account, after determining that A and B are associated, A, B is integrated into the second account appeal data C, and after obtaining the first account appeal data D, the server can further determine whether C and D are associated;
  • the multiple account appeal data for the same account may be related to whether the account appeal data is related to each other, and if the two or two judgments are associated, the comprehensive account appeal data may be integrated. It can be judged whether the data is related to another account of the same account.
  • the server determines that the requester of the second account appeal data is the owner of the first account when acquiring the first account appeal data, determining the first account appeal data and the second After the account appeal data is associated, the requester of the first account appeal data can be directly determined as the owner of the first account.
  • a user has filed an account complaint one year ago, he submitted an account appeal data A and passed the review. The user forgot the account password after one year, and then filed an account appeal and submitted another account appeal data B. If the account appeal processing data provided by the embodiment of the present application determines that the account appeal data A and B are associated, the requester of the account appeal data B may be considered as the account owner, and the account password may be modified to avoid subsequent follow-up. The process of auditing the system or customer service review, and efficiently and accurately determine the attribution of the account.
  • the embodiment of the present application determines that the second account appeal data is historically based on the second account appeal data.
  • the requester is the owner of the first account, and the requester that determines the first account appeal data is also the owner of the first account.
  • the first account may be The appeal data is integrated with the second account appeal data, and the integrated account appeal data is obtained, so that the account owner of the first account is determined by the integrated account appeal data.
  • Step S760 the password of the first account is sent to the first user or the password of the first account input by the first user is received, so that the first user passes the password operation. Make the first account.
  • the account appeal processing method provided by the embodiment of the present application can implement the judgment of the related account appeal data, and can determine the account appeal data belonging to the same natural person from multiple account appeal data of the same account, thereby assisting the account.
  • the attribution judgment determines the accuracy of the account attribution judgment result.
  • the server described below may refer to the account appeal processing method described above in the server perspective.
  • FIG. 11 is a structural block diagram of a server according to an embodiment of the present application.
  • the server may include:
  • the account appeal data obtaining module 100 is configured to obtain first account appeal data of the first account submitted by the first user, and the account appeal data includes at least one feature content;
  • the correlation determining module 200 is configured to compare the first account appeal data with the second account appeal data when the second account appeal data corresponding to the first account is present, and determine the first account appeal Correlation between data and second account appeal data;
  • the association determining module 300 is configured to determine that the first account appeal data and the second account appeal data are associated when the correlation meets a preset related condition
  • the account attribution determining module 400 determines that the first account belongs to the first user according to one or more of the first account appeal data and the second account appeal data;
  • the communication module 500 sends a password of the first account to the first user or receives a password of the first account input by the first user, so that the first user performs the password operation by using the password.
  • First account sends a password of the first account to the first user or receives a password of the first account input by the first user, so that the first user performs the password operation by using the password.
  • the correlation between the first account appeal data and the second account appeal data may indicate a possibility that the first account appeal data and the second account appeal data are associated.
  • the relating to the preset related condition may include: the possibility is not lower than the possibility corresponding to the multiple account appeal data of the same account.
  • FIG. 12 shows an optional structure of the correlation determining module 200 provided by the embodiment of the present application.
  • the correlation determination module 200 may include:
  • the difference level determining unit 210 is configured to compare and compare the feature contents of the first account appeal data and the second account appeal data respectively, and determine a difference level of each feature content;
  • the correlation determining execution unit 220 is configured to determine, according to the difference level of the feature content, the correlation between the first account appeal data and the second account appeal data, and the first account appeal data and the second account are represented by the correlation The likelihood that the appeal data is associated.
  • the difference level may include: the feature content is the same and the feature content is similar; based on FIG. 12, correspondingly, FIG. 13 is another structural block diagram of the server provided by the embodiment of the present application, in combination with FIG. 11 and As shown in FIG. 13, the server may further include:
  • the first condition determining module 600 is configured to: when the determined correlation between the first account appeal data and the second account appeal data meets the first correlation, determine that the correlation meets a preset relevant condition.
  • the correlation conforms to the first correlation: the feature content is the same; or, the part of the feature content is the same and the other part of the feature content is similar, and the type of the feature content with the same feature content conforms to the preset first type, and the feature The type of feature content with similar content conforms to the preset second type.
  • FIG. 14 shows an optional structure of the correlation determination execution unit 220 provided by the embodiment of the present application.
  • the correlation determination execution unit 220 may include:
  • the association probability determination sub-unit 221 is configured to perform classification processing on the difference level of the feature content according to the pre-trained Bayesian classification classification model, to obtain the first account appeal data and the second account appeal data.
  • the associated probability, the correlation is expressed in terms of the probability.
  • FIG. 15 is a block diagram showing still another structure of the server provided by the embodiment of the present application.
  • the server may further include:
  • a second condition matching determination module 700 configured to: when the probability meets a preset probability condition When it is determined, the correlation is in accordance with a preset relevant condition.
  • the difference level includes: the feature content is the same and the feature content is similar; and correspondingly, FIG. 16 shows another optional structure of the correlation determination execution unit 220 provided by the embodiment of the present application.
  • the sex determination execution unit 220 may include:
  • the binding execution unit 222 is configured to: when the determined correlation between the first account appeal data and the second account appeal data does not meet the first correlation, according to the pre-trained Bayesian classification model, the features are The difference level of the content is classified, and the probability that the first account appeal data and the second account appeal data are associated is obtained, and the correlation is represented by the probability.
  • the correlation does not meet the first correlation, and the feature content is different, or the feature content of the first type is different, or the feature content of the second type is not similar.
  • the embodiment of the present application may determine that the correlation meets a preset related condition when the probability meets a preset probability condition.
  • FIG. 17 shows another structural block diagram of the server provided by the embodiment of the present application, and FIG. 17 shows a pre-training process of the Bayesian classification classification model.
  • the server further Can include:
  • the pre-training module 800 is configured to collect account appeal data of multiple accounts, and an account has multiple account appeal data; for each account, compare each feature content of each account appeal data, and determine an account number of each account.
  • the difference level of each feature content of the appeal data determining the correlation between the account appeal data of each account by using the difference level of each feature content of the account appeal data of each account; the correlation includes the first correlation and the first Second correlation; the first correlation indicates that the feature content is the same, or a part of the feature content is the same as the other feature content, and the feature content of the feature content is the same as the preset first type and the feature content is similar.
  • the type of the feature content conforms to the preset second type; the second relevance represents the same account
  • the characteristics of each account in the multiple account appeal data are similar; according to the correlation between the account appeal data of each account, and the associated account appeal data and the unrelated account appeal data collected, Bayesian classification
  • the algorithm is trained to obtain a Bayesian classification classification model.
  • the pre-training module 800 determines the correlation between the account appeal data of each account in the difference level of each feature content of the account appeal data of each account.
  • the pre-training module 800 can be used for:
  • the difference characterization value of each feature content between the account appeal data is defined, and the difference characterization value group corresponding to the account appeal data is obtained;
  • the difference characterization value includes a first value indicating that the feature content is the same, and a second value indicating that the feature content is similar, the first value is different from the second value;
  • each of the account values is represented according to the difference between the account appeal data, and each The difference characterization value table corresponding to the account, and the difference characterization value table indicates the correlation between the account appeal data of each account.
  • the pre-training module 800 performs training according to the relevance of the account appeal data of each account, the collected account appeal data and the unrelated account appeal data, and is trained by the Bayesian classification algorithm to obtain the shell.
  • the Yesi classification model it can be used to:
  • the Bayesian classification algorithm is used to train and obtain the Bayesian classification classification model.
  • the account attribution determining module 400 is configured to determine, according to the second account appeal data, that the first is determined after determining that the first account appeal data and the second account appeal data are associated The account belongs to the second user who submits the second account appeal data, and determines that the first account belongs to the first user.
  • the account attribution determining module 400 is further configured to: after determining that the first account appeal data and the second account appeal data are associated, the first account appeal data and the second account appeal data are In summary, the integrated account appeal data is obtained, and the integrated account appeal data is used to determine that the first account belongs to the first user.
  • the server provided in the embodiment of the present application can implement the judgment of the associated account appeal data, and provides the possibility of determining the accuracy of the account attribution determination result by determining the attribution of the auxiliary account.
  • the hardware structure of the server provided by the embodiment of the present application can be as shown in FIG. 19, and includes: a processor 1, a communication interface 2, a memory 3, and a communication bus 4.
  • the processor 1, the communication interface 2, and the memory 3 complete communication with each other through the communication bus 4;
  • the communication interface 2 can be an interface of the communication module, such as an interface of the GSM module;
  • a processor 1 for executing a program
  • a memory 3 for storing a program
  • the program can include program code, the program code including computer operating instructions.
  • the processor 1 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.
  • CPU central processing unit
  • ASIC Application Specific Integrated Circuit
  • the memory 3 may include a high speed RAM memory and may also include a non-volatile memory such as at least one disk memory.
  • the program can be specifically used to:
  • the first account appeal data and the second account appeal data are compared, and the first account appeal data and the second account appeal data are determined. Correlation;
  • the communication interface is configured to: when determining that the first account belongs to the first user, send a password of the first account to the first user or receive the first user input a password of an account to enable the first user to operate the first account by using the password.
  • the embodiment of the present application further provides an account appeal processing system, and the system may be configured as shown in FIG. 1 , including: a server and at least one terminal;
  • the terminal is configured to submit, to the server, first account appeal data of the first account submitted by the first user; and the account appeal data includes at least one feature content;
  • the server is configured to obtain the first account appeal data of the first account, and the account appeal data includes at least one feature content; when there is the second account appeal data corresponding to the first account, the first account appeal data is respectively Comparing with the feature content of the second account appeal data, determining the relevance of the first account appeal data and the second account appeal data; determining the first account appeal when the correlation meets the preset relevant conditions
  • the data is associated with the second account appeal data; determining, according to one or more of the first account appeal data and the second account appeal data, that the first account belongs to the first user; When it is determined that the first account belongs to the first user, sending a password of the first account to the first user or receiving a password of the first account input by the first user, so that The first user operates the first account by using the password.
  • the steps of a method or algorithm described in connection with the embodiments disclosed herein can be implemented directly in hardware, a software module executed by a processor, or a combination of both.
  • the software module can be placed in random access memory (RAM), memory, read only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or technical field. Any other form of storage medium known.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Business, Economics & Management (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Fuzzy Systems (AREA)
  • Computational Linguistics (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Bioinformatics & Cheminformatics (AREA)

Abstract

本申请实施例提供的账号申诉处理方法包括：获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；当存在第一账号的第二账号申诉数据时，将第一账号申诉数据和第二账号申诉数据的特征内容进行比对，确定第一账号申诉数据和第二账号申诉数据的相关性；当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；根据第一账号申诉数据和/或第二账号申诉数据判断所述第一账号归属于第一用户；将第一账号的密码发送给第一用户或接收第一用户输入的第一账号的密码，以使第一用户通过所述密码操作第一账号。本申请实施例提升了账号归属判断结果的准确性。

Description

一种账号申诉处理方法及服务器
本申请要求于2016年05月05日提交中国专利局、申请号为201610293474.9、发明名称为“一种账号申诉处理方法及服务器”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及网络技术领域，更具体地说，涉及一种账号申诉处理方法及服务器。
背景技术
账号申诉是指在账号被盗，或者账号密码忘记的情况下，账号官方为账号所有者提供的一种找回账号的服务。进行账号申诉时，账号所有者需要提交账号申诉数据，在该账号诉数据被审核通过后，可判断账号归属于该账号所有者，该账号所有者可有权限修改账号的密码，达到账号所有者找回账号的目的。
在提交账号申诉数据时，提交的数据往往不全面，可能需要账号所有者提交二次或多次的账号申诉数据。另外，由于网络安全问题，用户资料泄露的情况较为常见，因此盗号者在盗取账号所有者的账号时，可能同时盗取到账号所有者的用户资料。盗号者为了达到控制被盗账号的目的，往往也会进行账号申诉，提交账号申诉数据。
可见，同一账号的账号申诉数据可能有多份，且该多份账号申诉数据可能是不同的自然人提交的，如该多份账号申诉数据中可能存在账号所有者提交的账号申诉数据，也存在盗号者提交的账号申诉数据，账号所有 者与盗号者认为是不同的自然人。
目前在审核账号申诉数据时，主要是将同一账号的多份账号申诉数据进行收集，然后分配给客服或者审核系统进行审核。然而，如果所收集的同一账号的多份账号申诉数据是不同自然人提交的，如既有账号所有者提交的账号申诉数据，又有盗号者提交的账号申诉数据，那么基于该同一账号的多份账号申诉数据进行审核时，审核结果存在不准确的可能，且可能将账号的归属判断给非账号所有者的情况。
因此，如何从同一账号的多份账号申诉数据中，识别相关联的账号申诉数据，实现同一自然人提交的账号申诉数据的确定，从而辅助账号的归属判断，显得尤为必要。
发明内容
本申请实施例提供了一种账号申诉处理方法，包括：
获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
当存在第一账号的第二账号申诉数据时，将所述第一账号申诉数据的特征内容和所述第二账号申诉数据的特征内容进行比对，确定所述第一账号申诉数据和所述第二账号申诉数据的相关性；
当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；
根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户；以及
将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
本申请实施例还提供了一种服务器，包括：
账号申诉数据获取模块，用于获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
相关性确定模块，用于当存在与第一账号对应的第二账号申诉数据时，将所述第一账号申诉数据的特征内容和所述第二账号申诉数据的特征内容进行比对，确定所述第一账号申诉数据和所述第二账号申诉数据的相关性；
关联确定模块，用于当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；
账号归属确定模块，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户；以及
通信模块，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
本申请实施例还提供了一种服务器，包括处理器，存储器，通信接口和通信总线，其中，其中所述处理器、所述通信接口和所述存储器通过所述通信总线通信，所述处理器执行存储器中存储的程序以执行：
获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
当存在与第一账号对应的第二账号申诉数据时，将所述第一账号申诉数据的特征内容和所述第二账号申诉数据的特征内容进行比对，确定所述第一账号申诉数据和所述第二账号申诉数据的相关性；
当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；
根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多 者判断所述第一账号归属于所述第一用户；
所述通信接口用于，当判断所述第一账号归属于所述第一用户时，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
本申请实施例提供的账号申诉处理方法中，服务器可获取第一用户提交的第一账号的第一账号申诉数据，并与已记录的第一账号的第二账号申诉数据进行特征内容的比对，确定出第一账号申诉数据和第二账号申诉数据的相关性，由该相关性表示第一账号申诉数据和第二账号申诉数据相关联的可能性，并通过预置同一账号的多份账号申诉数据相关联时所对应的预设相关条件，从而当所述相关性符合所述预设相关条件时，确定第一账号申诉数据和第二账号申诉数据相关联，实现属于同一自然人的账号申诉数据的确定，以此来辅助账号的归属判断，提升账号归属判断结果的准确性。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据提供的附图获得其他的附图。
图1为本申请实施例提供的账号申诉处理系统的结构框图；
图2为本申请实施例提供的账号申诉处理方法的信令流程图；
图3为本申请实施例提供的账号申诉处理方法的流程图；
图4为社交类型账号进行账号申诉时对应的部分特征内容的示意图；
图5为本申请实施例提供的账号申诉处理方法的另一流程图；
图6为本申请实施例提供的账号申诉处理方法的再一流程图；
图7为本申请实施例提供的账号申诉处理方法的又一流程图；
图8为本申请实施例提供的贝叶斯分类分类模型的预训练方法流程图；
图9为本申请实施例提供的确定各账号的账号申诉数据间的相关性的流程图；
图10为本申请实施例提供的账号申诉处理方法的又另一流程图；
图11为本申请实施例提供的服务器的结构框图；
图12为本申请实施例提供的相关性确定模块的结构框图；
图13为本申请实施例提供的服务器的另一结构框图；
图14为本申请实施例提供的相关性确定执行单元的结构框图；
图15为本申请实施例提供的服务器的再一结构框图；
图16为本申请实施例提供的相关性确定执行单元的另一构框图；
图17为本申请实施例提供的服务器的又一结构框图；
图18为本申请实施例提供的服务器的硬件结构框图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都对应本申请保护的范围。
图1为本申请实施例提供的账号申诉处理系统的结构框图。本申请实施例提供的账号申诉处理方法可基于图1所示系统实现。参照图1，本申请实施例提供的账号申诉处理系统可以包括服务器10和至少一个终端20。
服务器10可以为网络侧设置的用于处理终端提交的账号申诉数据的 设备。服务器10可以为单台服务器，也可以为由多台服务器组成的服务器群组。
终端20为用户侧的用于提交账号申诉数据的设备，如手机、平板电脑、笔记本电脑等。终端20所提交的账号申诉数据可能是账号所有者填写提交的，也有可能是盗号者填写提交的。本申请实施例需要通过服务器对终端20所提交的针对同一账号的账号申诉数据进行处理，判断出相关联的账号申诉数据，实现同一自然人所提交的账号申诉数据的判断，以便辅助判断账号的归属，提升账号归属判断给账号所有者的可能性。
基于图1所示系统，图2示出了本申请实施例提供的账号申诉处理方法的信令流程图，结合图1和图2所示，该流程可以包括以下步骤S10-S15。
步骤S10、第一用户通过终端向服务器提交第一账号的第一账号申诉数据。
第一账号为本申请实施例需要找回的账号。
终端的使用者，即，第一用户，可能是第一账号的所有者，也可能是盗号者。账号申诉数据具有至少一项特征内容，该特征内容可以是账号申诉数据中一项具体的申诉资料。
步骤S11、服务器获取所述第一账号申诉数据。
第一账号申诉数据中可以具有至少一项特征内容。
步骤S12、当服务器确定存在第一账号的第二账号申诉数据时，分别将第一账号申诉数据和第二账号申诉数据的特征内容进行比对，确定第一账号申诉数据和第二账号申诉数据的相关性。
可选的，第二账号申诉数据可以是服务器历史记录的第二用户提交的第一账号的账号申诉数据。在本申请实施例中，第二账号申诉数据可能是基于第一账号的所有者上传的账号申诉资料生成，也可能是基于盗号者 上传的账号申诉资料生成。因此，本申请实施例的目的即是判断针对第一账号的第一账号申诉数据和第二账号申诉数据是否相关联，即，第一用户和第二用户是否属于同一自然人，以便根据判断结果辅助判断第一账号的归属。
在得到第一账号申诉数据和第二账号申诉数据后，本申请实施例可将第一账号申诉数据和第二账号申诉数据的特征内容分别进行比对，确定特征内容的比对结果，基于第一账号申诉数据和第二账号申诉数据的特征内容的比对结果，本申请实施例可确定第一账号申诉数据和第二账号申诉数据的相关性，该相关性可用于表示第一账号申诉数据和第二账号申诉数据相关联的可能性，即提交第一账号申诉数据的第一用户和提交第二账号申诉数据的第二用户属于同一自然人的可能性。
步骤S13、服务器当确定所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联。
预设相关条件可以表示，在同一账号的多份账号申诉数据相关联时(如同一账号的多份账号申诉数据属于同一自然人时)，所对应的账号申诉数据之间的目标相关性；从而本申请实施例可在第一账号申诉数据和第二账号申诉数据的相关性符合所述预设相关条件时，认定第一账号申诉数据和第二账号申诉数据相关联，实现提交第一账号申诉数据的第一用户和提交第二账号申诉数据的第二用户属于同一自然人的确定。
步骤S14，当服务器确定所述第一账号申诉数据和所述第二账号申诉数据相关联时，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户。
步骤S15，服务器将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
可以看出，本申请实施例中，服务器可获取第一账号的第一账号申诉数据，并与已记录的第一账号的第二账号申诉数据进行特征内容的比对，确定出第一账号申诉数据和第二账号申诉数据的相关性，由该相关性表示第一账号申诉数据和第二账号申诉数据相关联的可能性，并通过预置同一账号的多份账号申诉数据相关联时所对应的预设相关条件，从而当所述相关性符合所述预设相关条件时，确定第一账号申诉数据和第二账号申诉数据相关联，即提交第一账号申诉数据的第一用户和提交第二账号申诉数据的第二用户属于同一自然人，以此来辅助账号的归属判断，提升账号归属判断结果的准确性。
下面以服务器的角度对本申请实施例提供的账号申诉处理方法进行介绍，下文描述的账号申诉处理方法可与上文描述的信令流程内容相互参照。
图3为本申请实施例提供的账号申诉处理方法的流程图。该方法可应用于服务器，可选的，该服务器可以是账号官方设置的用于进行账号申诉的服务器，该服务器可收集用户提交的账号申诉数据。参照图3，本申请实施例提供的账号申诉处理方法可以包括以下步骤S100-S140。
步骤S100、获取第一用户提交的第一账号的第一账号申诉数据。
第一账号为本申请实施例需要找回的账号。第一账号申诉数据可以是第一用户通过终端提交给服务器的用于找回第一账号的申诉资料。该第一用户可能是第一账号的所有者，也可能是盗号者。
可选的，一账号申诉数据可以包括至少一项特征内容，该特征内容可以是账号申诉数据中一项具体的申诉资料。具体的，一项特征内容可以对应账号申诉时一项需填写项目所填写的具体内容。
在填写账号申诉数据时，终端可显示至少一项的需填写项目，可选的，需填写项目如联系方式(比如手机，邮箱等)，个人信息(比如姓名， 身份证，住址等)，使用资料(比如使用过的密码，安全问题，绑定过的手机等)等。可选的，特征内容还可以包括终端自动提取携带在账号申诉数据中的终端标识、网络IP等。
值得注意的是，不同的账号类型对应的需填写项目可能不同，如社交类型账号对应的需填写项目可以具有：联系方式，个人信息，使用资料，社交关系(比如好友信息等)；又如游戏类型账号对应的需填写项目可以具有：联系方式，个人信息，使用资料，游戏角色名称等；因此进行账号申诉时，具体的需填写项目可视实际的账号类型和应用场景而定，并没有严格的限制。
为便于理解，图4示出了社交类型账号进行账号申诉时对应的部分特征内容。
步骤S110、当存在第一账号的第二账号申诉数据时，将第一账号申诉数据和第二账号申诉数据的特征内容进行比对，确定第一账号申诉数据和第二账号申诉数据的相关性。
可选的，第二账号申诉数据可以是服务器历史记录的第一账号的账号申诉数据。第二账号申诉数据提交至服务器的时间可以早于第一账号申诉数据提交至服务器的时间，第二账号申诉数据也可以是基于早于第一账号申诉数据提交的账号申诉数据生成。
可选的，第二账号申诉数据可能是基于第一账号的所有者上传的账号申诉资料生成，也可能是基于盗号者上传的账号申诉资料生成。因此，本申请实施例的目的即是判断针对第一账号的第一账号申诉数据和第二账号申诉数据是否相关联，以便根据判断结果辅助判断第一账号的归属。
服务器在确定第一账号的第一账号申诉数据和第二账号申诉数据后，可通过对第一账号申诉数据和第二账号申诉数据的具体内容的处理，确定出第一账号申诉数据和第二账号申诉数据相关联的可能性(称为第一 账号申诉数据和第二账号申诉数据的相关性)。
具体的，本申请实施例可将第一账号申诉数据和第二账号申诉数据的特征内容进行比对，确定特征内容的比对结果。如第一账号申诉数据和第二账号申诉数据均具有联系方式的特征内容，和个人信息的特征内容，则本申请实施例可分别比对第一账号申诉数据和第二账号申诉数据的联系方式的特征内容，及个人信息的特征内容，从而得到各项特征内容的比对结果。
基于第一账号申诉数据和第二账号申诉数据的各项特征内容的比对结果，可确定第一账号申诉数据和第二账号申诉数据的相关性。
可选的，该相关性可以如第一账号申诉数据和第二账号申诉数据相关联的概率。
步骤S120、当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联。
可选的，本申请实施例可预置在同一账号的多份账号申诉数据相关联时，所对应的预设相关条件(预设相关条件如在同一账号的多份账号申诉数据相关联时，所对应的账号申诉数据之间的目标相关性，该目标相关性可表示同一账号的多份账号申诉数据相关联时所对应的目标可能性)，从而在所述相关性符合所述预设相关条件时，认定第一账号申诉数据和第二账号申诉数据相关联的可能性，不低于，该目标可能性，实现所述第一账号申诉数据和所述第二账号申诉数据相关联的确定。
具体的，本申请实施例可设置同一账号的多份账号申诉数据相关联时，所对应的目标概率，从而在确定出第一账号申诉数据和第二账号申诉数据相关联的概率，且该概率不低于所述目标概率时，确定第一账号申诉数据和第二账号申诉数据相关联。
步骤S130，根据所述第一账号申诉数据和所述第二账号申诉数据中 的一者或多者判断所述第一账号归属于所述第一用户。
步骤S140，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
可以看出，本申请实施例提供的账号申诉处理方法包括：服务器获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；当存在第一账号的第二账号申诉数据时，将第一账号申诉数据和第二账号申诉数据的特征内容进行比对，确定第一账号申诉数据和第二账号申诉数据的相关性；当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户并将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
在本申请实施例中，服务器可获取第一用户提交的第一账号的第一账号申诉数据，并与已记录的第一账号的第二账号申诉数据进行特征内容的比对，确定出第一账号申诉数据和第二账号申诉数据的相关性，由该相关性表示第一账号申诉数据和第二账号申诉数据相关联的可能性，并通过预置同一账号的多份账号申诉数据相关联时所对应的预设相关条件，从而当所述相关性符合所述预设相关条件时，确定第一账号申诉数据和第二账号申诉数据相关联，实现属于同一自然人的账号申诉数据的确定，以此来辅助账号的归属判断，提升账号归属判断结果的准确性。
可选的，在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，可将第一账号申诉数据和第二账号申诉数据相综合，以便后续在审核判断第一账号的归属时，可通过综合的第一账号申诉数据和第二账号 申诉数据，提升第一账号归属判断给其账号所有者的概率，提升第一账号归属判断的准确性。
可选的，如果第一账号申诉数据和第二账号申诉数据的相关性不符合预设相关条件，则可以认为第一账号申诉数据和第二账号申诉数据不相关联，后续在审核账号申诉数据时，可将第一账号申诉数据和第二账号申诉数据区分审核。
可选的，在将第一账号申诉数据和第二账号申诉数据进行各项特征内容的比对时，本申请实施例可进行各项特征内容的差异性比对，从而确定各项特征内容之间的差异等级，基于各项特征内容之间的差异等级，综合分析出第一账号申诉数据和第二账号申诉数据的差异程度，以确定第一账号申诉数据和第二账号申诉数据的相关性。
可选的，差异等级诸如特征内容相同，特征内容相似，特征内容不相似，特征内容完全不同等。优选的，在本申请实施例中，差异等级类型优选采用特征内容相同及特征内容相似这两类。显然，基于不同的账号申诉情况，所选用的差异等级也可不限于特征内容相同及特征内容相似这两类。
图5示出了本申请实施例提供的账号申诉处理方法的另一流程图，参照图5，该方法可以包括以下步骤S200-S250。
步骤S200、获取第一用户提交的第一账号的第一账号申诉数据。
步骤S210、当存在第一账号的第二账号申诉数据时，分别将第一账号申诉数据和第二账号申诉数据的各项特征内容进行差异性比对，确定各项特征内容的差异等级。
步骤S220、以所述各项特征内容的差异等级确定第一账号申诉数据和第二账号申诉数据的相关性。
可选的，所述相关性可以表示第一账号申诉数据和第二账号申诉数 据相关联的可能性。
步骤S230、当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联。
可选的，所确定的各项特征内容的差异等级可以包括：特征内容相同及特征内容相似。一项特征内容相同可以认为第一账号申诉数据和第二账号申诉数据中该项的特征内容完全相同。一项特征内容相似可以认为第一账号申诉数据和第二账号申诉数据中该项的特征内容部分相同，且相同部分范围不低于设定范围。
可选的，相同部分范围不低于设定范围如相同部分的字符数在该项特征内容的总字符数中所占比例不低于设定比例，或者相同部分的字符数不低于设定数目等。显然，相同部分范围不低于设定范围的具体要求可视实际情况而定，并不限于本段描述。
以社交类型账号的账号申诉数据中社交关系需填写项为例，如果第一账号申诉数据中社交关系需填写项的特征内容为a1、a2、a3、a4、a5、a6(指代好友昵称等信息)，第二账号申诉数据中社交关系需填写项的特征内容也为a1、a2、a3、a4、a5、a6，由于第一账号申诉数据和第二账号申诉数据的社交关系的特征内容完全相同，则认为第一账号申诉数据和第二账号申诉数据中社交关系需填写项的特征内容的差异等级为特征内容相同。
如果第一账号申诉数据中社交关系需填写项的特征内容为a1、a2、a3、a4、a5、a6，第二账号申诉数据中社交关系需填写项的特征内容为a1、a2、a3、a4、a7；则第一账号申诉数据和第二账号申诉中社交关系需填写项的特征内容中，有4个好友信息是相同的，不低于设定比例或设定设定数目(具体数值可视实际情况而定)，则可认为第一账号申诉数据和第二账号申诉数据中社交关系需填写项的特征内容的差异等级为特征内容相似。
显然，针对第一账号申诉数据和第二账号申诉数据中其他项的特征内容的差异等级的确定原理，与上述说明类似，可相参照。
步骤S240，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户。
步骤S250，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
可选的，本申请实施例可通过所确定的第一账号申诉数据和第二账号申诉数据中各项特征内容的差异等级，判断第一账号申诉数据和第二账号申诉数据是强相关还是弱相关，强相关可以认为是所确定的第一账号申诉数据和第二账号申诉数据的相关性符合第一相关性，弱相关可以认为是所确定的第一账号申诉数据和第二账号申诉数据的相关性符合第二相关性，第一相关性高于第二相关性；从而在判断出第一账号申诉数据和第二账号申诉数据为强相关时，认定第一账号申诉数据和第二账号申诉数据的相关性符合预设相关条件，第一账号申诉数据和所述第二账号申诉数据相关联，即属于同一自然人。
可选的，以所述差异等级包括特征内容相同及特征内容相似进行说明，图6示出了本申请实施例提供的账号申诉处理方法的再一流程图。参照图6，该方法可以包括以下步骤S300-S360。
步骤S300、获取第一用户提交的第一账号的第一账号申诉数据。
步骤S310、当存在第一账号的第二账号申诉数据时，将第一账号申诉数据和第二账号申诉数据的特征内容进行差异性比对，确定特征内容的差异等级；所述差异等级包括：特征内容相同及特征内容相似。
步骤S320、以所述各项特征内容的差异等级确定第一账号申诉数据和第二账号申诉数据的相关性。
步骤S330、当所述相关性符合第一相关性时，确定所述相关性符合预设相关条件；所述相关性符合第一相关性包括：各项特征内容均相同；或，一部分特征内容相同及另一部分特征内容相似，特征内容相同的特征内容的类型符合预设第一类型，且特征内容相似的特征内容的类型符合预设第二类型。
一部分特征内容相同及另一部分特征内容相似表示第一账号申诉数据和第二账号申诉数据中的特征内容分为相同和相似两部分；特征内容相同的特征内容的类型符合预设第一类型可以认为是，特征内容相同的特征内容的类型应至少包含预设第一类型，即特征内容相同的特征内容的类型中至少应具有预设第一类型，同时也可能有其他的类型。
特征内容相似的特征内容的类型符合预设第二类型可以认为是，特征内容相似的特征内容的类型只能在预设第二类型的范围中，而不应超出该范围。
可选的，预设第一类型可以是从需填写的多项特征内容类型中选定的至少一项特征内容类型，以社交类型账号为例，预设第一类型可以为联系方式、终端标识、IP、个人信息等，预设第二类型可以是从需填写的多项特征内容类型中选定的至少一项特征内容类型，如使用资料、社交关系等。
预设第一类型和预设第二类型中的各项类型可以不同，也可以是部分相同。
以社交类型账号为例，第一账号申诉数据和第二账号申诉数据的相关性符合第一相关性可以是：
联系方式相同，
IP相同(提交账号申诉数据的IP相同)，
终端标识相同(提交账号申诉数据的终端的标识相同)，
个人信息相同，
使用资料相似(比如第一次填6个密码，第二次填5个密码，中间有四个是相同的)，
社交关系相似(第一次邀请6个好友，第二次邀请5个好友，中间有4个是相同的)。
或者，第一账号申诉数据和第二账号申诉数据的相关性符合第一相关性还可以是：
联系方式相同，
IP相同，
终端标识相同，
个人信息相同，
使用资料相同，
社交关系相同。
值得注意的，第一账号申诉数据和第二账号申诉数据的相关性符合第一相关性，可以是本申请实施例所设置的强制认定第一账号申诉数据和所述第二账号申诉数据相关联的策略，第一相关性的具体内容可视实际账号类型及应用场景而定。
步骤S340、确定所述第一账号申诉数据和所述第二账号申诉数据相关联。
步骤S350，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户。
步骤S360，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
可选的，本申请实施例除了设置强制认定第一账号申诉数据和所述第二账号申诉数据相关联的第一相关性内容外，还可以通过训练能够确定 第一账号申诉数据和第二账号申诉数据相关联的概率的算法模型，通过所确定的概率表示第一账号申诉数据和第二账号申诉之间的相关性，实现第一账号申诉数据和第二账号申诉数据是否相关联的判断。
可选的，该算法模型可以选用贝叶斯分类分类模型，贝叶斯分类模型是基于贝叶斯分类算法生成的，其是统计学的一种分类模型，可以利用概率统计知识进行分类。在许多场合，贝叶斯分类算法可以与决策树和神经网络分类算法相媲美，该算法能运用到大型数据库中，而且方法简单、分类准确率高、速度快。因此，本申请实施例可以通过预训练贝叶斯分类分类模型，以该贝叶斯分类分类模型来计算第一账号申诉数据和所述第二账号申诉数据相关联的概率，实现第一账号申诉数据和第二账号申诉数据的相关性的确定。
相应的，图7示出了本申请实施例提供的账号申诉处理方法的又一流程图。图7所示方法主要采用贝叶斯分类分类模型进行账号申诉处理，其可与图6所示方法采用第一相关性进行账号申诉处理的方式，相互独立实施。参照图7，该方法可以包括以下步骤S400-S460。
步骤S400、获取第一用户提交的第一账号的第一账号申诉数据。
步骤S410、当存在第一账号的第二账号申诉数据时，分别将第一账号申诉数据和第二账号申诉数据的各项特征内容进行差异性比对，确定各项特征内容的差异等级。
步骤S420、根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率。
所述概率可以表示所述第一账号申诉数据和所述第二账号申诉数据的相关性。
此处的差异等级至少包括：特征内容相同及特征内容相似。
步骤S430、当所述概率符合预设概率条件时，确定第一账号申诉数据和第二账号申诉数据的相关性符合预设相关条件。
步骤S440、确定所述第一账号申诉数据和所述第二账号申诉数据相关联。
步骤S450，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户。
步骤S460，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
可选的，图8示出了贝叶斯分类分类模型的预训练过程。参照图8，该过程可以包括以下步骤S500-S530。
步骤S500、收集多个账号的账号申诉数据，一账号对应有多份账号申诉数据。
本申请实施例可收集海量账号的账号申诉数据，以进行后续贝叶斯分类分类模型的训练。所收集的一个账号的账号申诉数据可以具有多份，一账号对应的多份账号申诉数据中可以有账号所有者提交的，也可能有盗号者提交的。
步骤S510、对于各账号，分别将各账号申诉数据的各项特征内容进行比对，确定各账号的账号申诉数据的各项特征内容的差异等级。
本申请实施例以一账号的多份账号申诉数据为单位，对各个账号的账号申诉数据进行处理，确定出各账号的账号申诉数据的各项特征内容的差异等级。具体的，针对每一账号，本申请实施例可将各账号申诉数据的各项特征内进行比对，识别出账号申诉数据的各项特征内容的差异等级；确定账号申诉数据的各项特征内容的差异等级的具体过程可以参照上文相应部分描述。
可选的，本申请实施例针对同一账号的多份账号申诉数据，可以是两两账号申诉数据进行各项特征内容的比对，两两比对后所综合的账号申诉数据，可再与另一该同一账号的账号申诉数据的各项特征内容进行比对。
步骤S520、以各账号的账号申诉数据的各项特征内容的差异等级，确定各账号的账号申诉数据间的相关性；所述相关性包括第一相关性和第二相关性。
如上文所述第一相关性表示同一账号的账号申诉数据之间为强相关，第二相关性表示同一账号的账号申诉数据之间为弱相关。
可选的，以差异等级包括特征内容相同及特征内容相似为例，第一相关性可以表示各项特征内容均相同；或，一部分特征内容相同及另一部分特征内容相似，且特征内容相同的特征内容的类型符合预设第一类型，特征内容相似的特征内容的类型应符合第二类型。
可选的，所述第二相关性可以表示同一账号的多份账号申诉数据中各项特征内容相似。以社交类型账号为例，账号的账号申诉数据间的相关性为第二相关性可以是：
联系方式相似(比如abcd@qq.com，abcd2016@qq.com)
IP相似(比如1.2.3.4和1.2.3.5)
个人信息相似(比如第一次提单姓名是刘杰，第二次提单姓名是杰)
使用资料低相似(使用资料相似可以如密码相同的字符达到第一值，如4个相同，使用资料低相似可以如密码相同的字符低于第一值，且高于第二值，第一值大于第二值，如密码字符中2个相同)
社交关系相似。
步骤S530、根据各账号的账号申诉数据间的相关性，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型。
在对收集的海量账号的账号申诉数据进行处理，确定出各账号的账号申诉数据间的相关性后，本申请实施例可基于提取的正、负样本数据，以贝叶斯分类算法进行训练。
正样本数据可以认为是收集提取的相关联的账号申诉数据，负样本数据可以认为是收集提取的不相关联的账号申诉数据；正、负样本数据可主要通过人工选择、及用户投诉数据获取。
可选的，本申请实施例可通过所收集的相关联的账号申诉数据和不相关联的账号申诉数据，分析出相关联的账号申诉数据间的关联性，及不相关联的账号申诉数据间的关联性，从而以所分析出的相关联的账号申诉数据间的关联性，及不相关联的账号申诉数据间的关联性为分类参考，以贝叶斯分类算法对所述各账号的账号申诉数据间的相关性进行训练，得出能够计算同一账号的多份账号申诉数据相关联的概率的算法公式，实现贝叶斯分类分类模型的训练。
可选的，为便于贝叶斯分类分类模型的训练，本申请实施例可将账号申诉数据的各项特征内容的差异等级进行数值化表示。具体的，图8所示方法中确定各账号的账号申诉数据间的相关性的过程可以如图9所示，包括以下步骤S600-S610。
步骤S600、对于各账号，根据账号申诉数据的各项特征内容的差异等级，定义账号申诉数据间的各项特征内容的差异表征值，得到账号申诉数据间对应的差异表征值组。
可选的，差异表征值组可以是基于两两账号申诉数据的各项特征内容的差异等级，所定义的差异表征值组合而成；所述差异表征值包括表示特征内容相同的第一值，及表示特征内容相似的第二值，所述第一值与所述第二值不同。可选的，第一值可以为1，第二值可以为0。显然，第一值和第二值的具体数值可根据实际情况而定，第一值为1，第二值为0的形式 仅是可选的。
在对同一账号的多份账号申诉数据的各项特征内容进行比对后，本申请实施例可以1和0表示各项特征内容的差异等级，如果比对出的一项特征内容相同，则可为该项特征内容定义差异表征值1，如果比对出一项特征内容相似，则可为该项特征内容定义差异表征值0。
如针对下述比对结果，本申请实施例定义各项特征内容的差异表征值后，可得到该账号的账号申诉数据对应的差异表征值组11100；
IP相同(定义差异表征值1)，
终端标识相同(定义差异表征值1)，
个人信息相同(定义差异表征值1)，
使用资料相似(定义差异表征值0)，
社交关系相似(定义差异表征值0)。
步骤S610、对于各账号，根据账号申诉数据间对应的差异表征值组，建立各账号对应的差异表征值表，以所述差异表征值表表示各账号的账号申诉数据间的相关性。
可选的，一账号的差异表征值表可以具有至少一个差异表征值组，一个差异表征值组表示该账号的两个账号申诉数据间各项特征内容的差异表征值的组合；。
可选的，一账号的差异表征值表的形式如下，其中1 0 0 0 0为一个差异表征值组，1 1 0 1 0为另一个差异表征值组：
1 0 0 0 0
1 1 0 1 0
可以看出，一账号的差异表征值表实际上是一个矩阵的形式。
在得到各账号对应的差异表征值表后，本申请实施例可以得到由多个矩阵构成的矩阵组，从而以所分析出的相关联的账号申诉数据间的关联 性，及不相关联的账号申诉数据间的关联性为分类参考，以贝叶斯分类算法对所述各账号的账号申诉数据间的相关性进行训练，得出能够计算同一账号的多份账号申诉数据相关联概率的算法公式，实现贝叶斯分类分类模型的训练。
因此，本申请实施例在训练贝叶斯分类分类模型时，可以根据各账号对应的差异表征值表，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型。
可选的，第一相关性的具体内容主要根据实际的账号申诉经验获得，可以是依据账号申诉处理专家在实际的账号申诉工作中的经验生成，用于强制认定第一账号申诉数据和所述第二账号申诉数据相关联，因此依据第一相关性认定第一账号申诉数据和所述第二账号申诉数据相关联的方式，可在所提交的账号申诉数据较为规范时，具有较高的账号申诉数据的归属判断准确性，而在所提交的账号申诉数据不较为规范时，通过第一相关性认定账号申诉数据的归属判断的准确性则不是很理想。
而贝叶斯分类分类模型是基于海量账号的账号申诉数据生成，可以较为灵活的处理各种情况的账号申诉数据，可以应对所提交的账号申诉数据较为规范及不较为规范情况下的账号申诉数据的归属判断，因此采用贝叶斯分类分类模型可以较为灵活的实现账号申诉数据的归属判断，且判断结果的准确性也较高。
基于此，本申请实施例可基于第一相关性或贝叶斯分类分类模型独立的进行账号申诉数据的归属判断；也可以是将第一相关性和贝叶斯分类分类模型相结合实现账号申诉数据的归属判断，如先基于第一相关性判断第一账号申诉数据和第二账号申诉数据是否相关联，若否，再基于贝叶斯分类分类模型判断第一账号申诉数据和第二账号申诉数据是否相关联，从而考虑账号申诉数据的全面情况，提升账号申诉数据归属判断的准确性。
相应的，图10示出了本申请实施例提供的账号申诉处理方法的又另一流程图。参照图10，该方法可以包括以下步骤S700-S760。
步骤S700、获取第一用户提交的第一账号的第一账号申诉数据。
步骤S710、当存在第一账号的第二账号申诉数据时，分别将第一账号申诉数据和第二账号申诉数据的各项特征内容进行差异性比对，确定各项特征内容的差异等级；所述差异等级包括：特征内容相同及特征内容相似。
步骤S720、当所确定的第一账号申诉数据和第二账号申诉数据的相关性不符合第一相关性时，根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率，以所述概率表示第一账号申诉数据和第二账号申诉数据的相关性。
其中，所述相关性不符合第一相关性包括：各项特征内容均不相同，或，预设第一类型的特征内容不相同，或，预设第二类型的特征内容不相似。
可选的，各项特征内容均不相同可以如：联系方式不同，IP不同，终端标识不同，个人信息不同，使用资料不同，社交关系不同。
预设第一类型的特征内容不相同可以如：联系方式不同，或，IP不同，或，终端标识不同，或，个人信息不同；至于使用资料和社交关系的差异等级则可不考虑，也可考虑。
预设第二类型的特征内容不相似可以如：使用资料不相似，或，社交关系不相似；至于联系方式，IP，终端标识，个人信息的差异等级则可不考虑，也可考虑。
步骤S730、当所述概率符合预设概率条件时，确定所述相关性符合预设相关条件。
可选的，预设概率条件可以通过设置目标概率实现，当所述概率不低于所述目标概率，则可认为所述概率符合预设概率条件。
步骤S740、确定所述第一账号申诉数据和所述第二账号申诉数据相关联。
显然，如果所述概率不符合预设概率条件，则可认为所述第一账号申诉数据和所述第二账号申诉数据不相关联。
步骤S750，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户。
可选的，在确定第一账号申诉数据和第二账号申诉数据相关联后，可将所述第一账号申诉数据和所述第二账号申诉数据相综合，得到综合后的账号申诉数据，以便以综合后的账号申诉数据判断第一账号的账号所有者。
可选的，将所述第一账号申诉数据和所述第二账号申诉数据相综合的方式，可以是针对各项特征内容，将特征内容相同的部分合并成一份，而特征内容不相同的部分保留、并与合并后的相同的特征内容组合；如针对使用资料的特征内容，第一账号申诉数据的使用密码为123456，共6个字符，第二账号申诉数据的使用密码为72345，共5个字符，则其中有4个字符相同，可将相同的2345合并成一份，而对于不同的字符1、6、7可保留并与合并成一份的2345组合，得到如1或7-2345-6；
可选的，第二账号申诉数据可以是历史提交的针对第一账号的账号申诉数据；第二账号申诉数据也可能是基于本申请实施例提供的账号申诉处理方法，将历史提交的针对第一账号的相关联的多份账号申诉数据综合后生成的账号申诉数据；如服务器历史获得了A、B两份针对第一账号的账号申诉数据，在判断A、B相关联后，可将A、B综合成第二账号申诉数据C，服务器在获取第一账号申诉数据D后，可再判断C和D是否相关联；
可选的，在本申请实施例中针对同一账号的多份账号申诉数据，可以是两两判断账号申诉数据是否相关联，两两判断后若相关联，则可相综合，综合的账号申诉数据，可再与同一账号的另一账号申诉数据进行是否相关联的判断。
可选的，如果服务器在获取第一账号申诉数据时，已审核判断出第二账号申诉数据的请求者为第一账号的所有者，则在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，可直接确定第一账号申诉数据的请求者为第一账号的所有者。
如一个用户在一年前进行了账号申诉，提交了一份账号申诉数据A并审核通过，而该用户在一年后忘记了账号密码，又进行了账号申诉，提交了另一账号申诉数据B，如果通过本申请实施例提供的账号申诉处理方法认定账号申诉数据A和B相关联，则可以认为账号申诉数据B的请求者为账号所有者，可拥有账号密码的修改权限，免去后续的审核系统或客服审核的过程，高效准确的实现账号的归属判断。
可以看出，本申请实施例在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，如果基于所述第二账号申诉数据已历史判断出所述第二账号申诉数据的请求者为第一账号的所有者，则确定所述第一账号申诉数据的请求者也为第一账号的所有者。
可选的，如果未基于第二账号申诉数据确定出第一账号的所有者，则在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，可以将所述第一账号申诉数据和所述第二账号申诉数据相综合，得到综合后的账号申诉数据，以便后续以综合后的账号申诉数据判断第一账号的账号所有者。
步骤S760，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操 作所述第一账号。
本申请实施例提供的账号申诉处理方法，可实现相关联的账号申诉数据的判断，可从同一账号的多份账号申诉数据中，实现属于同一自然人的账号申诉数据的确定，以此来辅助账号的归属判断，提升账号归属判断结果的准确性。
下面对本申请实施例提供的服务器进行介绍，下文描述的服务器可与上文以服务器角度描述的账号申诉处理方法相互对应参照。
图11为本申请实施例提供的服务器的结构框图。参照图11，该服务器可以包括：
账号申诉数据获取模块100，用于获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
相关性确定模块200，用于当存在与第一账号对应的第二账号申诉数据时，分别将第一账号申诉数据和第二账号申诉数据的各项特征内容进行比对，确定第一账号申诉数据和第二账号申诉数据的相关性；
关联确定模块300，用于当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；
账号归属确定模块400，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户；以及
通信模块500，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
可选的，所述第一账号申诉数据和第二账号申诉数据的相关性可以表示，第一账号申诉数据和第二账号申诉数据相关联的可能性。
所述相关性符合预设相关条件可以包括：所述可能性不低于同一账号的多份账号申诉数据相关联时所对应的可能性。
可选的，图12示出了本申请实施例提供的相关性确定模块200的可选结构。参照图12，相关性确定模块200可以包括：
差异等级确定单元210，用于分别将第一账号申诉数据和第二账号申诉数据的各项特征内容进行差异性比对，确定各项特征内容的差异等级；
相关性确定执行单元220，用于以所述各项特征内容的差异等级确定第一账号申诉数据和第二账号申诉数据的相关性，以所述相关性表示第一账号申诉数据和第二账号申诉数据相关联的可能性。
可选的，所述差异等级可以包括：特征内容相同及特征内容相似；基于图12所示，相应的，图13示出了本申请实施例提供的服务器的另一结构框图，结合图11和图13所示，该服务器还可以包括：
相关条件第一符合确定模块600，用于当所确定的第一账号申诉数据和第二账号申诉数据的相关性符合第一相关性时，确定所述相关性符合预设相关条件。
其中，所述相关性符合第一相关性包括：各项特征内容均相同；或，一部分特征内容相同及另一部分特征内容相似，且特征内容相同的特征内容的类型符合预设第一类型，特征内容相似的特征内容的类型符合预设第二类型。
可选的，图14示出了本申请实施例提供的相关性确定执行单元220的可选结构。参照图14，相关性确定执行单元220可以包括：
关联概率确定子单元221，用于根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率，以所述概率表示所述相关性。
相应的，基于图14所示，图15示出了本申请实施例提供的服务器的再一结构框图。结合图11和图15所示，该服务器还可以包括：
相关条件第二符合确定模块700，用于当所述概率符合预设概率条件 时，确定所述相关性符合预设相关条件。
可选的，所述差异等级包括：特征内容相同及特征内容相似；相应的，图16示出了本申请实施例提供的相关性确定执行单元220的另一可选结构，参照图16，相关性确定执行单元220可以包括：
结合执行子单元222，用于当所确定的第一账号申诉数据和第二账号申诉数据的相关性不符合第一相关性时，根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率，以所述概率表示所述相关性。
其中，所述相关性不符合第一相关性包括：各项特征内容均不相同，或，预设第一类型的特征内容不相同，或，预设第二类型的特征内容不相似。
相应的，基于图16所示，本申请实施例可在所述概率符合预设概率条件时，确定所述相关性符合预设相关条件。
可选的，图17示出了本申请实施例提供的服务器的又一结构框图，图17示出了贝叶斯分类分类模型的预训练过程，结合图11和图17所示，该服务器还可以包括：
预训练模块800，用于收集多个账号的账号申诉数据，一账号对应有多份账号申诉数据；对于各账号，分别将各账号申诉数据的各项特征内容进行比对，确定各账号的账号申诉数据的各项特征内容的差异等级；以各账号的账号申诉数据的各项特征内容的差异等级，确定各账号的账号申诉数据间的相关性；所述相关性包括第一相关性和第二相关性；所述第一相关性表示各项特征内容均相同，或，一部分特征内容相同及另一部分特征内容相似、且特征内容相同的特征内容的类型符合预设第一类型、特征内容相似的特征内容的类型符合预设第二类型；所述第二相关性表示同一账 号的多份账号申诉数据中各项特征内容相似；根据各账号的账号申诉数据间的相关性，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型。
可选的，预训练模块800在各账号的账号申诉数据的各项特征内容的差异等级，确定各账号的账号申诉数据间的相关性时，具体可用于：
对于各账号，根据账号申诉数据的各项特征内容的差异等级，定义账号申诉数据间的各项特征内容的差异表征值，得到账号申诉数据间对应的差异表征值组；所述差异表征值包括表示特征内容相同的第一值，及表示特征内容相似的第二值，所述第一值与所述第二值不同；对于各账号，根据账号申诉数据间对应的差异表征值组，建立各账号对应的差异表征值表，以所述差异表征值表表示各账号的账号申诉数据间的相关性。
相应的，预训练模块800在根据各账号的账号申诉数据间的相关性，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型时，具体可用于：
根据各账号对应的差异表征值表，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型。
可选的，账号归属确定模块400，用于在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，如果基于所述第二账号申诉数据已历史判断出所述第一账号归属于提交所述第二账号申诉数据的第二用户，则确定所述第一账号归属于所述第一用户。
可选的，账号归属确定模块400还用于在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，将所述第一账号申诉数据和所述第二账号申诉数据相综合，得到综合后的账号申诉数据，并以综合后的账号申诉数据判断所述第一账号归属于所述第一用户。
本申请实施例提供的服务器可实现相关联的账号申诉数据的判断，为辅助账号的归属判断，提升账号归属判断结果的准确性提供可能。
本申请实施例提供的服务器的硬件结构可如图19所示，包括：处理器1，通信接口2，存储器3和通信总线4。
其中处理器1、通信接口2、存储器3通过通信总线4完成相互间的通信；
可选的，通信接口2可以为通信模块的接口，如GSM模块的接口；
处理器1，用于执行程序；
存储器3，用于存放程序；
程序可以包括程序代码，所述程序代码包括计算机操作指令。
处理器1可能是一个中央处理器CPU，或者是特定集成电路ASIC(Application Specific Integrated Circuit)，或者是被配置成实施本申请实施例的一个或多个集成电路。
存储器3可能包含高速RAM存储器，也可能还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。
其中，程序可具体用于：
获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
当存在与第一账号对应的第二账号申诉数据时，分别将第一账号申诉数据和第二账号申诉数据的各项特征内容进行比对，确定第一账号申诉数据和第二账号申诉数据的相关性；
当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；
根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多 者判断所述第一账号归属于所述第一用户；以及
所述通信接口用于，当判断所述第一账号归属于所述第一用户时，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
本申请实施例还提供一种账号申诉处理系统，该系统的结构可以如图1所示，包括：服务器及至少一个终端；
其中，所述终端用于，向服务器提交第一用户提交的第一账号的第一账号申诉数据；一账号申诉数据包括至少一项特征内容；
所述服务器用于，获取第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；当存在与第一账号对应的第二账号申诉数据时，分别将第一账号申诉数据和第二账号申诉数据的各项特征内容进行比对，确定第一账号申诉数据和第二账号申诉数据的相关性；当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户；以及当判断所述第一账号归属于所述第一用户时，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
本说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似部分互相参见即可。对于实施例公开的装置而言，由于其与实施例公开的方法相对应，所以描述的比较简单，相关之处参见方法部分说明即可。
专业人员还可以进一步意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照 功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
结合本文中所公开的实施例描述的方法或算法的步骤可以直接用硬件、处理器执行的软件模块，或者二者的结合来实施。软件模块可以置于随机存储器(RAM)、内存、只读存储器(ROM)、电可编程ROM、电可擦除可编程ROM、寄存器、硬盘、可移动磁盘、CD-ROM、或技术领域内所公知的任意其它形式的存储介质中。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本申请。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本申请的精神或范围的情况下，在其它实施例中实现。因此，本申请将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (16)

  1. 一种账号申诉处理方法，其特征在于，包括：
    获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
    当存在第一账号的第二账号申诉数据时，将所述第一账号申诉数据的特征内容和所述第二账号申诉数据的特征内容进行比对，确定所述第一账号申诉数据和所述第二账号申诉数据的相关性；
    当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；
    根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户；以及
    将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
  2. 根据权利要求1所述的账号申诉处理方法，其特征在于，当存在多项特征内容时，将所述第一账号申诉数据的特征内容和所述第二账号申诉数据的特征内容进行比对，确定所述第一账号申诉数据和所述第二账号申诉数据的相关性包括：
    分别将所述第一账号申诉数据的各项特征内容和所述第二账号申诉数据的各项特征内容进行差异性比对，确定各项特征内容的差异等级；
    以所述各项特征内容的差异等级确定所述第一账号申诉数据和所述第二账号申诉数据的相关性。
  3. 根据权利要求2所述的账号申诉处理方法，其特征在于，所述差异等级包括：特征内容相同及特征内容相似；所述方法还包括：
    当所述第一账号申诉数据和所述第二账号申诉数据的相关性符合第一相关性时，确定所述相关性符合预设相关条件；
    所述第一账号申诉数据和所述第二账号申诉数据的相关性符合第一相关性包括以下各项之一：
    各项特征内容均相同；以及
    一部分特征内容相同，另一部分特征内容相似，且特征内容相同的特征内容的类型符合预设第一类型，特征内容相似的特征内容的类型符合预设第二类型。
  4. 根据权利要求2所述的账号申诉处理方法，其特征在于，所述以所述各项特征内容的差异等级确定所述第一账号申诉数据和所述第二账号申诉数据的相关性包括：
    根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率，以所述概率表示所述相关性；
    所述方法还包括：
    当所述概率符合预设概率条件时，确定所述相关性符合预设相关条件。
  5. 根据权利要求2所述的申诉处理方法，其特征在于，所述差异等级包括：特征内容相同及特征内容相似；所述以所述各项特征内容的差异等级确定所述第一账号申诉数据和所述第二账号申诉数据的相关性包括：
    当所述第一账号申诉数据和所述第二账号申诉数据的相关性不符合第一相关性时，根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率，以所述概率表示所述相关性；
    其中，所述相关性不符合第一相关性包括以下各项中的一项或多项：
    各项特征内容均不相同，
    预设第一类型的特征内容不相同，以及
    预设第二类型的特征内容不相似，
    所述方法还包括：
    当所述概率符合预设概率条件时，确定所述相关性符合预设相关条件。
  6. 根据权利要求4或5所述的账号申诉处理方法，其特征在于，所述方法还包括：预训练所述贝叶斯分类分类模型，
    预训练所述贝叶斯分类分类模型包括：
    收集多个账号的多份账号申诉数据，一账号对应有多份账号申诉数据；
    对于各账号，分别将各账号申诉数据的各项特征内容进行比对，确定各账号的账号申诉数据的各项特征内容的差异等级；
    以各账号的账号申诉数据的各项特征内容的差异等级，确定各账号的账号申诉数据间的相关性；所述相关性包括第一相关性和第二相关性；所述第一相关性表示各项特征内容均相同，或，一部分特征内容相同及另一部分特征内容相似、且特征内容相同的特征内容的类型符合预设第一类型、特征内容相似的特征内容的类型符合预设第二类型；所述第二相关性表示同一账号的多份账号申诉数据中各项特征内容相似；
    根据各账号的账号申诉数据间的相关性，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型。
  7. 根据权利要求6所述的账号申诉处理方法，其特征在于，所述以各账号的账号申诉数据的各项特征内容的差异等级，确定各账号的账号申诉数据间的相关性包括：
    对于各账号，根据账号申诉数据的各项特征内容的差异等级，定义账号申诉数据间的各项特征内容的差异表征值，得到账号申诉数据间对应的差异表征值组；所述差异表征值包括表示特征内容相同的第一值，及表示特征内容相似的第二值，所述第一值与所述第二值不同；
    对于各账号，根据账号申诉数据间对应的差异表征值组，建立各账号对应的差异表征值表，以所述差异表征值表表示各账号的账号申诉数据间的相关性。
  8. 根据权利要求7所述的账号申诉处理方法，其特征在于，所述根据各账号的账号申诉数据间的相关性，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型包括：
    根据各账号对应的差异表征值表，及所收集的相关联的账号申诉数据和不相关联的账号申诉数据，以贝叶斯分类算法进行训练，得到贝叶斯分类分类模型。
  9. 根据权利要求1所述的账号申诉处理方法，其特征在于，根据所述 第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户包括：
    在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，如果基于所述第二账号申诉数据已历史判断出所述第一账号归属于提交所述第二账号申诉数据的第二用户，则判断所述第一账号归属于所述第一用户；
    或，在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，将所述第一账号申诉数据和所述第二账号申诉数据相综合，得到综合后的账号申诉数据，并以综合后的账号申诉数据判断所述第一账号归属于所述第一用户。
  10. 一种服务器，其特征在于，包括：
    账号申诉数据获取模块，用于获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
    相关性确定模块，用于当存在第一账号的第二账号申诉数据时，将所述第一账号申诉数据的特征内容和所述第二账号申诉数据的特征内容进行比对，确定所述第一账号申诉数据和所述第二账号申诉数据的相关性；
    关联确定模块，用于当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；
    账号归属确定模块，根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户；以及
    通信模块，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
  11. 根据权利要求10所述的服务器，其特征在于，所述相关性确定模块包括：
    差异等级确定单元，用于当存在多项特征内容时，分别将所述第一账号申诉数据的各项特征内容和所述第二账号申诉数据的各项特征内容进行差异性比对，确定各项特征内容的差异等级；
    相关性确定执行单元，用于以所述各项特征内容的差异等级确定第一账号申诉数据和第二账号申诉数据的相关性。
  12. 根据权利要求11所述的服务器，其特征在于，所述差异等级包括：特征内容相同及特征内容相似；所述服务器还包括：
    相关条件第一符合确定模块，用于当所述第一账号申诉数据和所述第二账号申诉数据的相关性符合第一相关性时，确定所述相关性符合预设相关条件；
    其中，所述第一账号申诉数据和所述第二账号申诉数据的相关性符合第一相关性包括以下各项之一：
    各项特征内容均相同；以及
    一部分特征内容相同，另一部分特征内容相似，且特征内容相同的特征内容的类型符合预设第一类型，特征内容相似的特征内容的类型符合预设第二类型。
  13. 根据权利要求11所述的服务器，其特征在于，所述相关性确定执行单元包括：
    关联概率确定子单元，用于根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率，以所述概率表示所述相关性；
    所述服务器还包括：
    相关条件第二符合确定模块，用于当所述概率符合预设概率条件时，确定所述相关性符合预设相关条件。
  14. 根据权利要求11所述的服务器，其特征在于，所述差异等级包括：特征内容相同及特征内容相似；所述相关性确定执行单元包括：
    结合执行子单元，用于当所述第一账号申诉数据和所述第二账号申诉数据的相关性不符合第一相关性时，根据预训练的贝叶斯分类分类模型，对所述各项特征内容的差异等级进行分类处理，得到所述第一账号申诉数据和所述第二账号申诉数据相关联的概率，以所述概率表示所述相关性；
    其中，所述相关性不符合第一相关性包括以下各项中的一项或多项：
    各项特征内容均不相同；
    预设第一类型的特征内容不相同；以及
    预设第二类型的特征内容不相似。
  15. 根据权利要求10所述的服务器，其特征在于，所述账号归属确定模块用于
    在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，如果基于所述第二账号申诉数据已历史判断出所述第一账号归属于提交所述第二账号申诉数据的第二用户，则确定所述第一账号归属于所述第一用户，或，
    在确定所述第一账号申诉数据和所述第二账号申诉数据相关联后，将所述第一账号申诉数据和所述第二账号申诉数据相综合，得到综合后的账号申诉数据，并以综合后的账号申诉数据判断所述第一账号归属于所述第一用户。
  16. 一种服务器，包括处理器，存储器，通信接口和通信总线，其中所述处理器、所述通信接口和所述存储器通过所述通信总线通信，所述处理器执行存储器中存储的程序以执行：
    获取第一用户提交的第一账号的第一账号申诉数据，一账号申诉数据包括至少一项特征内容；
    当存在第一账号的第二账号申诉数据时，将所述第一账号申诉数据的特征内容和所述第二账号申诉数据的特征内容进行比对，确定所述第一账号申诉数据和所述第二账号申诉数据的相关性；
    当所述相关性符合预设相关条件时，确定所述第一账号申诉数据和所述第二账号申诉数据相关联；以及
    根据所述第一账号申诉数据和所述第二账号申诉数据中的一者或多者判断所述第一账号归属于所述第一用户；
    所述通信接口用于，当判断所述第一账号归属于所述第一用户时，将所述第一账号的密码发送给所述第一用户或接收所述第一用户输入的所述第一账号的密码，以使所述第一用户通过所述密码操作所述第一账号。
PCT/CN2017/083050 2016-05-05 2017-05-04 一种账号申诉处理方法及服务器 WO2017190670A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP17792496.6A EP3454503B1 (en) 2016-05-05 2017-05-04 Account complaint processing method and server
JP2018557835A JP6707147B2 (ja) 2016-05-05 2017-05-04 アカウント申立て処理方法及びサーバ
KR1020187034329A KR102218506B1 (ko) 2016-05-05 2017-05-04 계정 컴플레인트 처리 방법 및 서버
US15/976,451 US10567374B2 (en) 2016-05-05 2018-05-10 Information processing method and server

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610293474.9A CN107346310B (zh) 2016-05-05 2016-05-05 一种账号申诉处理方法及服务器
CN201610293474.9 2016-05-05

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/976,451 Continuation US10567374B2 (en) 2016-05-05 2018-05-10 Information processing method and server

Publications (1)

Publication Number Publication Date
WO2017190670A1 true WO2017190670A1 (zh) 2017-11-09

Family

ID=60202759

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/083050 WO2017190670A1 (zh) 2016-05-05 2017-05-04 一种账号申诉处理方法及服务器

Country Status (6)

Country Link
US (1) US10567374B2 (zh)
EP (1) EP3454503B1 (zh)
JP (1) JP6707147B2 (zh)
KR (1) KR102218506B1 (zh)
CN (1) CN107346310B (zh)
WO (1) WO2017190670A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109741065A (zh) * 2019-01-28 2019-05-10 广州虎牙信息科技有限公司 一种支付风险识别方法、装置、设备及存储介质
US10728033B2 (en) * 2015-09-28 2020-07-28 Tencent Technology (Shenzhen) Company Limited Identity authentication method, apparatus, and storage medium

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107346310B (zh) * 2016-05-05 2020-10-27 腾讯科技（深圳）有限公司 一种账号申诉处理方法及服务器
CN111831286B (zh) * 2019-04-12 2023-11-14 中国移动通信集团河南有限公司 一种用户投诉处理方法和设备
US11449755B2 (en) * 2019-04-29 2022-09-20 Microsoft Technology Licensing, Llc Sensitivity classification neural network
CN111400174B (zh) * 2020-03-05 2022-08-12 支付宝(杭州)信息技术有限公司 数据源的应用效能的确定方法、装置和服务器

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103179098A (zh) * 2011-12-23 2013-06-26 阿里巴巴集团控股有限公司 一种网络账号的密码找回方法和装置
CN103188218A (zh) * 2011-12-28 2013-07-03 富泰华工业（深圳）有限公司 密码找回系统及密码找回方法
CN103281192A (zh) * 2013-05-31 2013-09-04 腾讯科技（深圳）有限公司 数据找回方法、装置和系统
WO2014139097A1 (en) * 2013-03-13 2014-09-18 Intel Corporation Systems and methods for account recovery using a platform attestation credential
CN104104656A (zh) * 2013-04-07 2014-10-15 腾讯科技（深圳）有限公司 找回帐号的方法及装置

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005182354A (ja) 2003-12-18 2005-07-07 Ricoh Co Ltd 認証サーバ、パスワード再発行通知方法及びプログラム
GB0624024D0 (en) * 2006-12-01 2007-01-10 Ibm Event correlation based trouble ticket resolution system incorporating adaptive rules optimization
US20090089876A1 (en) * 2007-09-28 2009-04-02 Jamie Lynn Finamore Apparatus system and method for validating users based on fuzzy logic
US20100306821A1 (en) * 2009-05-29 2010-12-02 Google, Inc. Account-recovery technique
US8863253B2 (en) * 2009-06-22 2014-10-14 Beyondtrust Software, Inc. Systems and methods for automatic discovery of systems and accounts
JP2013519162A (ja) * 2010-02-01 2013-05-23 ジャンプタップ，インコーポレイテッド 統合化された広告システム
US9626725B2 (en) * 2010-12-23 2017-04-18 Facebook, Inc. Using social graph for account recovery
CN102541918A (zh) * 2010-12-30 2012-07-04 阿里巴巴集团控股有限公司 重复信息识别方法和设备
EP2575315A1 (en) * 2011-09-30 2013-04-03 British Telecommunications Public Limited Company Controlled access
CN103118043B (zh) * 2011-11-16 2015-12-02 阿里巴巴集团控股有限公司 一种用户账号的识别方法及设备
US9589137B2 (en) 2013-03-01 2017-03-07 Hitachi, Ltd. Method for detecting unfair use and device for detecting unfair use
KR102002541B1 (ko) * 2013-03-08 2019-10-01 휴렛-팩커드 디벨롭먼트 컴퍼니, 엘.피. 클라우드 기반 어플리케이션의 로그인 관리 방법 및 이를 수행하기 위한 화상형성장치
US9197632B2 (en) * 2013-03-15 2015-11-24 Kaarya Llc System and method for account access
US20140344356A1 (en) * 2013-05-17 2014-11-20 Mohammad A.H. Ramadhan Social networking service
US9876804B2 (en) * 2013-10-20 2018-01-23 Cyber-Ark Software Ltd. Method and system for detecting unauthorized access to and use of network resources
CN105306425B (zh) * 2014-07-15 2020-01-10 腾讯科技（深圳）有限公司 一种对账号归属进行认证的方法及装置
CN107346310B (zh) * 2016-05-05 2020-10-27 腾讯科技（深圳）有限公司 一种账号申诉处理方法及服务器

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103179098A (zh) * 2011-12-23 2013-06-26 阿里巴巴集团控股有限公司 一种网络账号的密码找回方法和装置
CN103188218A (zh) * 2011-12-28 2013-07-03 富泰华工业（深圳）有限公司 密码找回系统及密码找回方法
WO2014139097A1 (en) * 2013-03-13 2014-09-18 Intel Corporation Systems and methods for account recovery using a platform attestation credential
CN104104656A (zh) * 2013-04-07 2014-10-15 腾讯科技（深圳）有限公司 找回帐号的方法及装置
CN103281192A (zh) * 2013-05-31 2013-09-04 腾讯科技（深圳）有限公司 数据找回方法、装置和系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3454503A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10728033B2 (en) * 2015-09-28 2020-07-28 Tencent Technology (Shenzhen) Company Limited Identity authentication method, apparatus, and storage medium
CN109741065A (zh) * 2019-01-28 2019-05-10 广州虎牙信息科技有限公司 一种支付风险识别方法、装置、设备及存储介质

Also Published As

Publication number Publication date
US20180262482A1 (en) 2018-09-13
CN107346310A (zh) 2017-11-14
KR102218506B1 (ko) 2021-02-19
JP2019515391A (ja) 2019-06-06
CN107346310B (zh) 2020-10-27
EP3454503A1 (en) 2019-03-13
EP3454503B1 (en) 2021-02-17
KR20190002593A (ko) 2019-01-08
US10567374B2 (en) 2020-02-18
EP3454503A4 (en) 2019-11-27
JP6707147B2 (ja) 2020-06-10

Similar Documents

Publication Publication Date Title
WO2017190670A1 (zh) 一种账号申诉处理方法及服务器
US20180300358A1 (en) Image Retrieval Method and System
US9077678B1 (en) Facilitating photo sharing
CN105022761B (zh) 群组查找方法和装置
TWI724552B (zh) 識別風險商家的方法及裝置
WO2016180267A1 (zh) 交互数据的处理方法及装置
CN104980402B (zh) 一种识别恶意操作的方法及装置
WO2018192348A1 (zh) 数据处理方法、装置及服务器
WO2020257991A1 (zh) 用户识别方法及相关产品
WO2018188588A1 (zh) 一种信息推送方法、可读介质及电子设备
US20220100839A1 (en) Open data biometric identity validation
WO2019238125A1 (zh) 信息处理方法、相关设备及计算机存储介质
WO2020186789A1 (zh) 用户反欺诈实现方法、装置、计算机设备及存储介质
US20230410221A1 (en) Information processing apparatus, control method, and program
CN110765760A (zh) 一种法律案件分配方法、装置、存储介质和服务器
WO2019033518A1 (zh) 信息获取方法、装置、计算机可读存储介质及终端设备
WO2018068664A1 (zh) 网络信息识别方法和装置
CN107786349B (zh) 一种针对用户账号的安全管理方法及装置
US11367311B2 (en) Face recognition method and apparatus, server, and storage medium
CN111027065B (zh) 一种勒索病毒识别方法、装置、电子设备及存储介质
US8918406B2 (en) Intelligent analysis queue construction
US11704392B2 (en) Fraud estimation system, fraud estimation method and program
US11295118B2 (en) Online user verification without prior knowledge of the user
TWM591664U (zh) 用以進行身分註冊程序的電子裝置
CN112992152B (zh) 一种单兵声纹识别系统、方法、存储介质及电子设备

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2018557835

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20187034329

Country of ref document: KR

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17792496

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017792496

Country of ref document: EP

Effective date: 20181205