Method and device for identifying stolen accounts
By generating an anti-account theft preparation library and using user behavior characteristics to match, more stolen accounts are identified, which solves the problem of the number of suspected stolen accounts in the existing technology and improves the accuracy of account identification.
Patent Information
- Application Number
- CN202011541662.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-12-23
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2040-12-23
Smart Images

Figure CN112667706B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of Internet, and in particular to a method and device for identifying stolen accounts. Background Art
[0002] With the development of the Internet, more and more social software has entered people's lives, and people record and share their lives on various social software. However, with the popularity of social software, the problem of social software account theft has also emerged in an endless stream. Many criminals steal other people's accounts to publish bad advertisements or for illegal activities (such as stealing funds from other people's accounts and using other people's accounts to commit fraud, etc.).
[0003] When identifying a suspected stolen account, the relevant technology will first verify the account's password, and if the password is correct, further verify the device information corresponding to the user's account login within a certain period of time.
[0004] However, this method of identifying stolen accounts may result in the number of suspected stolen accounts being too low, which in turn causes some suspected stolen accounts to remain undiscovered. Summary of the invention
[0005] The embodiment of the present application provides a method and device for identifying stolen accounts, which can solve the problem that the number of suspected stolen accounts mined is too low, resulting in some suspected stolen accounts still not being mined. In order to achieve the above purpose, the embodiment of the present application adopts the following technical solutions:
[0006] In a first aspect, a method for identifying a stolen account is provided, the method comprising:
[0007] Get the target account that meets the specified policy;
[0008] Matching the target account with accounts in a pre-generated anti-account theft preparation database, the anti-account theft preparation database including suspected stolen accounts mined based on user behavior characteristics;
[0009] In the case that there is an account matching the target account in the anti-account theft preparation database, the target account is determined to be a stolen account.
[0010] In a second aspect, a device for identifying a stolen account is provided, the device comprising:
[0011] The acquisition module is used to obtain the target account that meets the specified policy;
[0012] A matching module, used to match the target account with accounts in a pre-generated anti-account theft preparation library, wherein the anti-account theft preparation library includes suspected stolen accounts mined based on user behavior characteristics;
[0013] The determination module is used to determine that the target account is a stolen account when there is an account matching the target account in the anti-account theft preparation database.
[0014] In an embodiment of the present application, after obtaining a target account that meets the specified policy, the target account is matched with a suspected stolen account in an anti-account theft preparation library, and the anti-account theft preparation library includes suspected stolen accounts mined based on user behavior characteristics; if there is an account matching the target account in the anti-account theft preparation library, the target account is determined to be a stolen account. Since the anti-account theft preparation library includes a large number of suspected stolen accounts mined based on user behavior characteristics, the number of suspected stolen accounts is greatly expanded. Therefore, in the process of matching the target account with the suspected stolen accounts in the anti-account theft preparation library, the hit rate of the target account can be improved, thereby determining more stolen accounts, solving the problem that the number of suspected stolen accounts mined by related technologies is too low, which causes some suspected stolen accounts to still not be mined. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0016] Figure 1 This is a flowchart of a method for identifying a stolen account provided in an embodiment of the present application.
[0017] Figure 2 It is a flow chart of a process of generating an anti-account theft preparation library provided by an embodiment of the present application.
[0018] Figure 3 This is a flowchart of a method for determining a suspected stolen account provided by an embodiment of the present application.
[0019] Figure 4 This is a flowchart of another method for determining a suspected stolen account provided by an embodiment of the present application.
[0020] Figure 5 This is a flowchart of another method for identifying a stolen account provided in an embodiment of the present application.
[0021] Figure 6 It is a structural block diagram of a device for identifying a stolen account provided in an embodiment of the present application. DETAILED DESCRIPTION
[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0023] In response to the above-mentioned problem of how to solve the existing method of identifying stolen accounts because the number of suspected stolen accounts dug out is too low, resulting in some suspected stolen accounts still not being dug out, the present application proposes a solution, aiming to provide a method for identifying stolen accounts that can solve the problem of some suspected stolen accounts still not being dug out because the number of suspected stolen accounts dug out is too low.
[0024] Figure 1 1 is a flow chart of a method for identifying a stolen account provided by an embodiment of the present application. Figure 1 As shown, the method for identifying a stolen account provided by the embodiment of the present application may include:
[0025] Step 110: Obtain target accounts that meet the specified policy.
[0026] It is understandable that the specified strategy is a basic strategy for determining whether a user account may be a suspected stolen account. For example, an account that satisfies the specified strategy is an account that has passed password verification. The specified strategy may be part of the existing strategy for identifying suspected stolen accounts, or may be other newly established strategies.
[0027] Step 120, matching the target account with accounts in a pre-generated anti-account theft preparation database, wherein the anti-account theft preparation database includes suspected stolen accounts mined based on user behavior characteristics.
[0028] Step 130: if there is an account matching the target account in the anti-account theft preparation database, determine that the target account is a stolen account.
[0029] The method for identifying stolen accounts provided in the embodiment of the present application greatly expands the number of suspected stolen accounts because the anti-account theft preparation library includes a large number of suspected stolen accounts mined out based on user behavior characteristics. Therefore, in the process of matching the target account with the suspected stolen accounts in the anti-account theft preparation library, the hit rate of the target account can be improved, thereby determining more stolen accounts, solving the problem that the number of suspected stolen accounts mined out by related technologies is too low, which causes some suspected stolen accounts to still not be mined out.
[0030] Optionally, before matching the target account with the account in the pre-generated anti-account theft preparation database in step 130, refer to Figure 2 The generation process of the anti-account theft preparation library in the method for identifying stolen accounts provided in the embodiment of the present application may also include the following steps:
[0031] Step 1310, obtaining behavior characteristic information of each user, wherein the behavior characteristic information includes uplink operation information and user information;
[0032] Step 1320, determining a suspected stolen account based on the behavior characteristic information;
[0033] Step 1330, putting the mined suspected stolen accounts into the anti-account theft preparation database.
[0034] Optionally, the uplink operation information in the above step 1310 may include one or more of user posting information, comment information, forwarding information, profile information, like information, etc. The user information may include one or more of account information, device information, location information, etc.
[0035] It is understandable that the user's uplink operation information and user information can reflect the user's behavioral characteristics, and whether the user's account is a suspected stolen account can be determined from the user's behavioral characteristic information. For example, when the user's behavioral characteristics suddenly change significantly, that is, they are inconsistent with the user's usual behavioral characteristics, it can be determined that the person who logged into the user account may not be the user himself, that is, the user's account may be stolen. After obtaining the suspected stolen account based on the user's behavioral characteristic information, the suspected stolen account is placed in the anti-theft reserve library.
[0036] Optionally, before the suspected stolen accounts are put into the anti-account theft preparation database, the anti-account theft preparation database may include data of suspected stolen accounts mined based on traditional methods, and may also include data of suspected stolen accounts obtained through other strategies.
[0037] Optionally, in an embodiment of the present application, the uplink operation information may also include stolen data reported by users, and the step 1310 of obtaining the behavior characteristic information of each user may include obtaining the stolen data reported by each user on the social software. Accordingly, the step 1320 of determining a suspected stolen account based on the behavior characteristic information may include: determining a suspected stolen account based on the stolen data reported by users on the social software.
[0038] The method for determining suspected stolen accounts based on the stolen data fed back by users on the social software can timely determine suspected stolen accounts based on the information fed back by the users themselves, and because the stolen data fed back by the users themselves are relatively real, the accuracy of determining suspected stolen accounts based on the stolen data fed back by the users is relatively high.
[0039] In an embodiment of the present application, a user can feedback stolen data through the user profile of a social software. Of course, a user can also directly feedback stolen data through the content of a post on a social software. For example, a user can send a microblog to feedback that the account has been stolen, etc. That is to say, in an embodiment of the present application, based on the stolen data reported by users on the social software, determining a suspected stolen account may include: for any user: determining whether the user's profile content contains the stolen data reported by the user; if the user's profile content contains the stolen data reported by the user, determining that the user's account is a suspected stolen account; if the user's profile content does not contain the stolen data reported by the user, filtering out the user's non-original post content; and determining whether the user's post content contains the stolen data reported by the user; if the user's post content contains the stolen data reported by the user, determining that the user's account is a suspected stolen account.
[0040] The stolen data reported by the user in the profile and post content may be words such as "stolen", "account stolen" and "account hacking" in the profile and post content, indicating that the user account may be stolen. The post content may be original post content.
[0041] It is understandable that when a user discovers that his account has been stolen, he will usually indicate in his profile that his account has been stolen, or inform his friends through a post.
[0042] The user's profile and post content can often directly reflect the information the user currently wants to express. Therefore, based on whether the user's profile and original post content contain stolen data reported by the user, it is possible to quickly and accurately determine whether the user's account is a suspected stolen account.
[0043] Optionally, the stolen data reported by the user in the post content can also be the official list mentioned by the user in the post content. The mention can be the user mentioning other accounts on the social software through the @ (@XXX means, tell the user with the nickname XXX, the user with the nickname XXX can see the post content with "@XXX") character, and the official list can include official management accounts such as official customer service, security center and official secretary.
[0044] It is understandable that when a user finds that his account has been stolen, he will usually promptly inform the official management account that his account has been stolen and request the official management account to provide him with help.
[0045] Figure 3 1 is a flow chart of a method for determining a suspected stolen account provided by an embodiment of the present application. Figure 3As shown, in an optional embodiment, determining a suspected stolen account based on the stolen data feedback by users on the social software can be achieved through the following steps:
[0046] Judge whether the content of the user profile contains "stolen";
[0047] If the content of the user profile contains "stolen", then determine the user's account as a suspected stolen account;
[0048] If the content of the user profile does not contain "stolen", then after filtering out the non-original posts in the user's posts, judge whether the content of the user's original posts contains "stolen";
[0049] If the content of the user's original posts contains "stolen", then judge whether the content of the user's original posts @s the official list;
[0050] If the content of the user's original posts does not contain "stolen", then discard the user's account;
[0051] If the content of the user's original posts @s the official list, then determine the user's account as a suspected stolen account;
[0052] If the content of the user's original posts does not @ the official list, then discard the user's account.
[0053] Optionally, if the content of the user's original posts does not @ the official list, the following steps can also be performed:
[0054] After filtering out the reverse keyword A in the content of the user's original posts, judge whether the content of the user's original posts contains "Weibo stolen";
[0055] If the content of the user's original posts contains "Weibo stolen" and does not contain the reverse keyword B, then judge whether the content of the user's original posts contains "suspect";
[0056] If the content of the user's original posts contains "suspect", then further judge whether the content of the user's original posts contains "suspect myself" or "suspect oneself";
[0057] If the content of the user's original posts contains "suspect myself" or "suspect oneself", and does not @ anyone, or @s a certain account and the content of the original posts contains "small account" or "new account", then determine the user's account as a suspected stolen account;
[0058] If the content of the user's original posts does not contain "suspect", and does not @ anyone, or @s a certain account and the content of the original posts contains "small account" or "new account", then determine the user's account as a suspected stolen account.
[0059] It is understandable that the reverse keyword in the above steps is a keyword expressing a negative meaning, and the reverse keyword A and the reverse keyword B are different keywords expressing a negative meaning. If the user's original posting content contains "suspect me" or "suspect myself" and does not @ anyone, it means that the user suspects that his account has been stolen. If the user's original posting content contains "suspect me" or "suspect myself", @ a certain account and the original posting content contains "small account" or "new account", it means that the user suspects that his account has been stolen, and the account @ed by the user is another account of his own.
[0060] Optionally, the uplink operation information may include the number of uplink operations performed by the user in a unit period, and the step 1310 of obtaining the behavior characteristic information of each user may include obtaining the number of uplink operations of each user in each unit period in a specified period, and the specified period includes at least one unit period. Accordingly, the step 1320 of determining a suspected stolen account based on the behavior characteristic information may include the following steps:
[0061] Determine, based on the number of uplink operations of the user in each unit period within the specified period, a coefficient of variation of the number of uplink operations of each user within the specified period, wherein the coefficient of variation is a ratio between a standard deviation value of the number of uplink operations of the user within the specified period and an average value of the number of uplink operations of the user within the specified period;
[0062] The user's account whose coefficient of variation is above a preset ratio threshold is determined as a suspected stolen account.
[0063] The unit period may be a unit period such as half a day or a day.
[0064] The coefficient of variation is a normalized measure of the discreteness of a probability distribution. In an embodiment of the present application, the coefficient of variation may represent changes in the trajectory of a user's uplink operation behavior.
[0065] Understandably, stolen accounts are often used to post bad ads or engage in illegal activities. The most direct way to monetize stolen accounts is to sell them to organizations or individuals engaged in various illegal and criminal activities. Then, organizations or individuals engaged in various illegal and criminal activities use the purchased accounts to change vulgar profile pictures, post blogs and comments involving pornography, gambling, or gun trafficking, spread illegal and harmful information, and divert traffic to illegal websites for profit. Or they conduct malicious marketing, increase followers, repost, and like for marketing accounts, and use improper means to increase the exposure of blog posts and make illegal profits.
[0066] Each user has his or her own regular upward operation behavior track, and the trend of the track is relatively stable. Of course, there will be occasional small peaks of activity. Then, when an account is stolen and used to post bad advertisements or illegal behaviors, the behavior track will change abnormally. Therefore, monitoring the trend changes of upward operations within the user's specified period can timely discover suspected stolen accounts.
[0067] The above method for determining a suspected stolen account based on the number of uplink operations performed by the user in a unit period can obtain the coefficient of variation of the number of uplink operations performed by the user in a unit period according to the number of uplink operations performed by the user in a unit period, so as to use the coefficient of variation to see whether the user's behavior trajectory has undergone abnormal changes. If the user's behavior trajectory has undergone abnormal changes, it can be determined that the user's account is a suspected stolen account.
[0068] As mentioned above, the uplink operation may be one or more of a posting operation, a forwarding operation, a comment operation, and a like operation.
[0069] Optionally, the user information in the embodiment of the present application may include user types, and the user types may include star-chasing users, V users, and active users. The star-chasing users are social software users whose number of interactions with officially certified star users exceeds a preset first threshold value, the V users are officially certified social software users whose number of fans exceeds a preset second threshold value, and the active users are users whose posts on social software are continuous over multiple time periods;
[0070] Determining the user's account whose coefficient of variation is above a preset ratio threshold as a suspected stolen account may include:
[0071] Filter out target type users from the users whose coefficient of variation is above a preset ratio threshold, wherein the target type includes at least one of star-chasing users, V users, and active users;
[0072] The accounts of the remaining users whose coefficient of variation is above the preset ratio threshold are determined as suspected stolen accounts after filtering out the users.
[0073] For example, the time period in which the multiple time periods are all continuous may be any one of a day, a week, and a year. The time period may also be other time periods other than a day, a week, and a year, and the specific meaning of the time period is not limited in the embodiment of the present application.
[0074] It is understandable that the star-chasing users, V users and active users are usually more active on social software, that is, they have more upstream operations. Therefore, the accuracy of judging whether the star-chasing users, V users and active users are suspected of being stolen by the coefficient of variation of the upstream operations of the star-chasing users, V users and active users is low, so they are filtered out.
[0075] Therefore, the above-mentioned filtering out users of the target type from the users whose coefficient of variation is above the preset ratio threshold and determining the accounts of the remaining users after the filtering out of the users whose coefficient of variation is above the preset ratio threshold as suspected stolen accounts can improve the accuracy of determining suspected stolen accounts.
[0076] Figure 4 FIG. 1 is a flowchart of another method for determining a suspected stolen account provided by an embodiment of the present application. In an optional embodiment, Figure 4 As shown, in step 1320, based on the behavior feature information, determining the suspected stolen account can be specifically implemented by the following steps:
[0077] Step 1321, counting the number of uplink operations performed by each user per day.
[0078] Step 1322, filter out users whose total number of uplink operations within seven days is above the threshold and who have performed uplink operations for seven consecutive days;
[0079] Step 1323, calculating the average value of the number of uplink operations of the above-screened users within seven days and the standard deviation of the number of uplink operations within seven days;
[0080] Step 1324, calculating the coefficient of variation of the number of uplink operations of the user within seven days according to the average value and the standard deviation value;
[0081] Step 1325, filtering out a user set whose coefficient of variation is above a preset ratio threshold;
[0082] Step 1326, filtering out at least one of star-chasing users, V users, and active users from the users whose coefficient of variation is above a preset ratio threshold;
[0083] Step 1327: The accounts of the remaining users after filtering out the users whose coefficient of variation is above the preset ratio threshold are determined as suspected stolen accounts.
[0084] It is understandable that after an account is stolen by criminals, the criminals will usually perform a large number of operations on the stolen account, such as publishing a large number of advertisements. Therefore, the above steps screen out users whose total number of upstream operations within seven days is above the threshold and who have performed upstream operations for seven consecutive days. That is, the accounts of the users screened out above may be suspected stolen accounts.
[0085] In addition, the uplink operation information may also include the posting information of the user on the social software; obtaining the behavior characteristic information of each user in step 1310 may include obtaining a preset number of posting contents of each user on the social software. Accordingly, determining the suspected stolen account based on the behavior characteristic information in step 1320 may include:
[0086] For any user:
[0087] Determine a special topic based on the preset number of posts, wherein the special topic includes at least one of promotional advertisements and illegal content, and the illegal content includes at least one of pornographic content, violent content, and gambling content;
[0088] Based on the special topic, a suspected stolen account is determined.
[0089] The determination of the suspected stolen account based on the special topic may be to match the topic in the user's post content with the special topic, and if the user's post content successfully matches the special topic, the user's account is determined as a suspected stolen account.
[0090] Through data analysis of stolen accounts, it is found that after an account is stolen, the hacker will use the stolen account, which is currently in normal status, to maintain the account, that is, to keep the account active. One of the ways to maintain the account is to publish content with the same or similar topics, and the topics are changed regularly. Although the topics are constantly changing, the daily topics are fixed. After the hacker has maintained the account for a period of time, he will use the stolen account to publish bad advertisements or for illegal activities. Once the stolen account is used to publish bad advertisements or for illegal activities, the frequency of topic changes in the content will increase. By discovering new topics in a timely manner and capturing data, suspected stolen accounts can be identified.
[0091] The above-mentioned determination of a suspected stolen account based on a preset number of posts by the user on the social software can capture special topics in the user's posts, and then confirm the suspected stolen account based on the special topics.
[0092] Optionally, based on the special topic, determining the suspected stolen account may further include the following steps:
[0093] Preliminarily screening the user's posts to find the posts with multiple special topics from the user's most recent preset number of posts;
[0094] Summarize the special topics with multiple special topic posts filtered out within a preset time period;
[0095] Manually confirming in the summarized special topics the stolen special topics indicating that the user has a high probability of being stolen;
[0096] Matching the topic in the user's posting content with the stolen special topic;
[0097] If the topic in the user's post successfully matches the stolen special topic, the user's account is determined to be a suspected stolen account.
[0098] Optionally, the user's post content in the above step of matching the topic in the user's post content with the stolen special topic can be the user's real-time post content, or the user's post content within a preset number of days, or the user's real-time post content and the user's post content within a preset number of days.
[0099] It is understandable that in order to avoid omissions in the process of determining suspected stolen accounts based on special topics, the user's real-time posting content and the user's posting content within a preset number of days can be matched with the stolen special topic. Matching the user's posting content within a preset number of days with the stolen special topic is a method to remedy the above-mentioned omissions.
[0100] Considering that more suspected stolen users are mined based on multiple user behavior characteristics, the uplink operation information may also include user interaction information on social software; the acquisition of each user's behavior characteristic information in step 1310 may include acquiring each user's interaction information on social software, and the interaction information may include at least one of forwarding information, comment information, and like information. Accordingly, the determination of suspected stolen accounts based on the behavior characteristic information in step 1320 may include:
[0101] For any user;
[0102] Based on the interaction information, determine the account of the interaction object;
[0103] Determining whether the interactive object account has fraud characteristics;
[0104] In the case where the interactive object account has fraud characteristics, the user's account is determined to be a suspected stolen account.
[0105] It is understandable that the fraud characteristics refer to characteristics such as bad advertisements or illegal behaviors. Normal users will not have any interactive behavior with accounts with fraud characteristics. Therefore, there are only two possibilities for accounts that interact with accounts with fraud characteristics: one is the hacker's own account, and the other is the account stolen by the hacker. Therefore, if an account interacts with an account with fraud characteristics, and the account is not the hacker's own account, then the account is a suspected stolen account.
[0106] The above method for determining a suspected stolen account based on the user's interactive information on social software can determine whether the user interacts with the hacker's account based on the user's interactive information on the social software. If the user interacts with the hacker on the social software, the user's account can be determined as a suspected stolen account.
[0107] Optionally, when the interactive object account has fraud characteristics, determining that the user account is a suspected stolen account may include the following steps:
[0108] Determine whether there is a time gap between a user’s interactions with an account that has fraudulent characteristics;
[0109] If there is a time interval between the user's interaction with the account with fraudulent characteristics, determine whether the user's historical postings or the time interval between posts have original characteristics;
[0110] If the user's historical postings or the time interval between posts have original characteristics, the user's account will be identified as a suspected stolen account;
[0111] If the user's historical posting content or posting interval does not have original characteristics, determine whether the user's account registration time can be obtained.
[0112] If the user's account registration time cannot be obtained, the user's account is determined to be a suspected stolen account;
[0113] If the user's account registration time is less than the predetermined year and there is no original posting content, the user's account will be identified as a suspected stolen account;
[0114] If the user's account registration time is less than the preset year and there is original posting content, and the posting time of the latest original posting content is less than another preset year, the user's account is determined to be a suspected stolen account.
[0115] It is understandable that the time interval is the time interval between the time when the account with fraudulent characteristics posts and the time when the user interacts with it. If there is no time interval between the interaction between the user and the account with fraudulent characteristics, it means that the user is the hacker himself, and the user's account is the hacker's own small account, so he can interact with the main account through his small account immediately after his main account posts. In addition, the posts of hackers are usually advertisements and other content, so they are all unified or similar, and do not have the characteristics of original posting content. The time interval between posts by hackers is usually fixed, and does not have the characteristics of original posting time interval. The determination of whether the user's account registration time can be obtained is to determine whether the user's account is frozen. If the user's account registration time cannot be obtained, it means that the user's account has been frozen. Account freezing generally means that the official management account freezes the account after the user reports to the official management account that his account has been stolen. If the user's account registration time is less than the predetermined year, it can be said that the account is not a newly created small account. If the posting time of the latest original posting content is less than another preset year, it can be said that the account may have been stolen after the preset year.
[0116] Optionally, the user information may include the user's geographic location information, the user's authentication information, and the user's posting frequency; the acquisition of each user's behavioral characteristic information in step 1310 may also include acquiring each user's uplink operation information, as well as at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency.
[0117] Determining the suspected stolen account based on the behavioral characteristic information in step 1320 may also include: determining the suspected stolen account based on the user's uplink operation information, and at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency.
[0118] It is understandable that each user has his or her own unique behavioral characteristics. When the user's behavioral characteristics change, the user's account may be stolen. For example, each user has his or her own commonly used geographic location information. When the user's account login location changes, the user's account may be stolen; each user has his or her own commonly used login device. When the user's account login device changes, the user's account may be stolen. Of course, there are also cases where the same user logs in with another device at another location. Therefore, in order to ensure the accuracy of identifying suspected stolen accounts, when multiple information in the user's uplink operation information, the user's geographic location information, the user's authentication information, and the user's posting frequency meet the preset conditions, the account of the user who meets the conditions can be determined as a suspected stolen user.
[0119] The above method for determining suspected stolen accounts based on the user's uplink operation information, as well as at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency further enriches the method for determining suspected stolen accounts. Suspected stolen users are determined based on the user's various behavioral characteristic information, and more suspected stolen accounts can be discovered.
[0120] Optionally, determining a suspected stolen account based on the uplink operation information of the user, and at least one of the geographical location information of the user, the authentication information of the user, and the posting frequency of the user may include the following steps:
[0121] In the case where the profile content or original posting content of the user contains the stolen data reported by the user, obtaining at least one of the geographical location information of each user, the authentication information of the user, and the posting frequency of the user;
[0122] Determining whether at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency satisfies a preset condition, where the preset condition corresponds to the above information one-to-one (e.g., determining whether the user's current geographic location information is the same as the user's commonly used geographic location information);
[0123] If at least one of the geographical location information of the user, the authentication information of the user, and the posting frequency of the user meets a corresponding preset condition, the account of the user is determined to be a suspected stolen account.
[0124] Optionally, the frequently used geographical location information may be geographical location information of a user who logs in to a frequently used account within a preset number of days, and the authentication information of the user may be identity information of the user or level information of the user.
[0125] Optionally, determining a suspected stolen account based on the uplink operation information of the user, and at least one of the geographical location information of the user, the authentication information of the user, and the posting frequency of the user may include the following steps:
[0126] In the case where the coefficient of variation of the number of uplink operations of the user within the specified period is above a preset ratio threshold, obtaining at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency; determining whether at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency meets a preset condition; if at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency meets the corresponding preset condition, determining the user's account as a suspected stolen account.
[0127] Optionally, determining a suspected stolen account based on the uplink operation information of the user, and at least one of the geographical location information of the user, the authentication information of the user, and the posting frequency of the user may include the following steps:
[0128] In the case that the content of the user's post successfully matches the special topic, the user's geographic location information, the user's authentication information and at least one of the user's posting frequency are obtained; it is determined whether at least one of the user's geographic location information, the user's authentication information and the user's posting frequency meets a preset condition; if at least one of the user's geographic location information, the user's authentication information and the user's posting frequency meets the corresponding preset condition, the user's account is determined to be a suspected stolen account.
[0129] Optionally, determining a suspected stolen account based on the uplink operation information of the user, and at least one of the geographical location information of the user, the authentication information of the user, and the posting frequency of the user may include the following steps:
[0130] In the case where the interactive object account of the user has the fraud feature, at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency is obtained. It is determined whether at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency meets a preset condition; if at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency meets the corresponding preset condition, the user's account is determined to be a suspected stolen account.
[0131] Optionally, before obtaining at least one of the geographical location information of the user, the authentication information of the user, and the posting frequency of the user, frozen users or users with too low a level may be filtered out.
[0132] According to the above description of the embodiment of the present application, it can be seen that after determining the suspected stolen account, the embodiment of the present application can put the determined suspected stolen account into the anti-account theft preparation library, and match the target account with the account in the anti-account theft preparation library. When there is an account matching the target account in the anti-account theft preparation library, the target account is determined to be a stolen account. In addition, because the method for identifying stolen accounts provided by the embodiment of the present application can determine the suspected stolen account through the user's multiple behavioral feature information when determining the suspected stolen account, that is, the above method for determining the suspected stolen account covers a wide range and can obtain a large number of suspected stolen accounts. Therefore, in the process of matching the target account with the suspected stolen account in the anti-account theft preparation library, the hit rate of the target account can be improved, thereby determining more stolen accounts, solving the problem that the number of suspected stolen accounts mined by the related technology is too low, and then causing some suspected stolen accounts to still not be mined.
[0133] Optionally, after a suspected stolen account is identified through the above method, the method for identifying a stolen account in the embodiment of the present application may further include: real-time monitoring of the identified suspected stolen account.
[0134] It is understandable that the suspected stolen accounts identified are monitored in real time, the characteristics of the suspected stolen accounts can be statistically analyzed through the data obtained from the monitoring, and the results of the method for identifying stolen accounts in the embodiment of the present application can be displayed and summarized based on the data obtained from the monitoring to facilitate subsequent improvements.
[0135] Figure 5 FIG. 1 is a flowchart of another method for identifying a stolen account provided by an embodiment of the present application. Figure 5 As shown, the method for identifying a stolen account provided by the embodiment of the present application may include the following steps:
[0136] Step 510: Acquire behavior characteristic information of each user, where the behavior characteristic information includes uplink operation information and user information.
[0137] Step 520: Determine a suspected stolen account based on the behavior characteristic information.
[0138] Step 530: put the discovered suspected stolen accounts into the anti-account theft preparation database.
[0139] Step 540: Obtain target accounts that meet the specified policy.
[0140] Step 550: Match the target account with an account in a pre-generated anti-account theft preparation database.
[0141] Step 560: If there is an account matching the target account in the anti-account theft preparation database, determine that the target account is a stolen account.
[0142] It is understandable that the method for determining a suspected stolen account based on the user's behavioral characteristic information in step 520 may include one or more of the methods for determining a suspected stolen account based on the user's multiple uplink operation information and / or multiple user information described above. Furthermore, the methods for determining a suspected stolen account based on the user's multiple uplink operation information and / or multiple user information described above are all in parallel.
[0143] The method for identifying stolen accounts provided in the embodiment of the present application can determine suspected stolen accounts based on various behavioral feature information of users, such as stolen data reported by users, the number of uplink operations of users, topics in the content of user posts, and interactive information between users and others on social software, and put the suspected stolen accounts into the anti-account theft preparation library. After obtaining a target account that meets the specified policy, the target account is matched with the account in the anti-account theft preparation library. If there is an account matching the target account in the anti-account theft preparation library, the target account is determined to be a stolen account.
[0144] Since the method for identifying stolen accounts provided in the embodiment of the present application can determine suspected stolen accounts based on the user's behavioral characteristic information, and there are many conditions for determining suspected stolen accounts, more suspected stolen accounts can be discovered. When the target account is matched with the suspected stolen accounts in the anti-account theft preparation library, the hit rate of the target account can be improved, thereby determining more stolen accounts.
[0145] Corresponding to the method for identifying a stolen account described above, an embodiment of the present application further provides a device for identifying a stolen account. The method for identifying a stolen account described above can be applied to the device for identifying a stolen account provided in an embodiment of the present application.
[0146] Figure 6 is a structural block diagram of a device for identifying a stolen account provided by an embodiment of the present application. Figure 6 As shown, an embodiment of the present application provides a device 600 for identifying a stolen account, including an acquisition module 601 , a matching module 602 , and a determination module 603 .
[0147] The acquisition module 601 is used to acquire a target account that meets a specified policy.
[0148] The matching module 602 is used to match the target account with an account in a pre-generated anti-account theft preparation database, where the anti-account theft preparation database includes suspected stolen account numbers mined based on user behavior characteristics.
[0149] The determination module 603 is used to determine that the target account is a stolen account when there is an account matching the target account in the anti-account theft preparation database.
[0150] Optionally, the acquisition module 601 may also be used to acquire behavior feature information of each user, the behavior feature information including uplink operation information and user information. The determination module 603 may also be used to determine a suspected stolen account based on the behavior feature information, and put the suspected stolen account into the anti-theft reserve library.
[0151] Optionally, the uplink operation information may include stolen data reported by users. In the process of obtaining the behavior characteristic information of each user, the acquisition module 601 is used to: obtain the stolen data reported by each user on the social software. Accordingly, in the process of determining a suspected stolen account based on the behavior characteristic information, the determination module 603 is used to: determine the suspected stolen account based on the stolen data reported by the user on the social software.
[0152] Optionally, in the process of determining a suspected stolen account based on the stolen data reported by the user on the social software, the determination module 603 is used to: for any user: determine whether the user's profile and / or original posting content contains the stolen data reported by the user; if the user's profile and / or original posting content contains the stolen data reported by the user, determine that the user's account is a suspected stolen account.
[0153] Optionally, the uplink operation information may include the number of uplink operations performed by the user within a unit period. In the process of acquiring the behavioral characteristic information of each user, the acquisition module 601 is used to acquire the number of uplink operations performed by each user in each unit period within a specified period, and the specified period includes at least one unit period. Accordingly, in the process of determining a suspected stolen account based on the behavioral characteristic information, the determination module 603 is used to: determine the coefficient of variation of the number of uplink operations of each user within the specified period based on the number of uplink operations performed by each user in each unit period within the specified period, wherein the coefficient of variation is the ratio between the standard deviation of the number of uplink operations performed by the user within the specified period and the average value of the number of uplink operations performed by the user within the specified period; and determine the account of the user whose coefficient of variation is above the preset ratio threshold as a suspected stolen account.
[0154] Optionally, the user information may include user types, which include star-chasing users, V users and active users. The star-chasing users are social software users whose number of interactions with officially certified celebrity users exceeds a preset first threshold value, the V users are social software users who are officially certified and have a number of fans exceeding a preset second threshold value, and the active users are users who post on the social software continuously on an annual average basis.
[0155] In the process of determining the accounts of users whose coefficient of variation is above the preset ratio threshold as suspected stolen accounts, the determination module 603 is used to: filter out users of a target type from the users whose coefficient of variation is above the preset ratio threshold, wherein the target type includes at least one of star-chasing users, V users and active users; and determine the accounts of users remaining after filtering out the users whose coefficient of variation is above the preset ratio threshold as suspected stolen accounts.
[0156] Optionally, the uplink operation information may include posting information of the user on the social software. In the process of obtaining the behavioral characteristic information of each user, the acquisition module 601 is used to: obtain a preset number of posting contents of each user on the social software. Correspondingly, in the process of determining suspected stolen accounts based on the behavioral characteristic information, the determination module 603 is used to: for any user: determine a special topic based on the preset number of posting contents, wherein the special topic includes at least one of promotional advertisements and illegal content, and the illegal content includes: at least one of pornographic content, violent content and gambling content; based on the special topic, determine the suspected stolen account.
[0157] Optionally, the uplink operation information may include the user's interactive information on the social software. In the process of obtaining the behavioral characteristic information of each user, the acquisition module 601 is used to: obtain the interactive information of each user on the social software, and the interactive information includes at least one of forwarding information, comment information and like information. Accordingly, in the process of determining a suspected stolen account based on the behavioral characteristic information, the determination module 603 is used to: for any user: determine the interactive object account based on the interactive information; determine whether the interactive object account has fraud characteristics; if the interactive object account has fraud characteristics, determine the user's account as a suspected stolen account.
[0158] Optionally, the user information may include at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency. In the process of obtaining the behavioral characteristic information of each user, the acquisition module 601 is used to: obtain the uplink operation information of each user, as well as at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency. It should be understood that the method for identifying a stolen account described above can be applied to the device for identifying a stolen account provided in an embodiment of the present application, so the content of the device for identifying a stolen account here can refer to the description of the method part above.
[0159] Through the description of the above embodiments, it will be understood by those skilled in the art that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0160] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0161] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0162] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0163] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0164] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0165] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0166] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0167] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.
Claims
1. A method for identifying a stolen account, characterized in that: The method comprises: Get the target account that meets the specified policy; Matching the target account with accounts in a pre-generated anti-account theft preparation database, the anti-account theft preparation database including suspected stolen accounts mined based on user behavior characteristics; If there is an account matching the target account in the anti-account theft preparation database, determining that the target account is a stolen account; The generation process of the anti-account theft preparation library includes: Acquire behavior characteristic information of each user, wherein the behavior characteristic information includes uplink operation information and user information; Based on the behavioral characteristic information, determine the suspected stolen account; Putting the discovered suspected stolen accounts into the anti-account theft reserve database; The uplink operation information includes the number of uplink operations performed by the user within a unit period; the user information includes user types, and the user types include star-chasing users, V users, and active users. The star-chasing users are social software users whose number of interactions with officially certified star users exceeds a preset first threshold value, the V users are officially certified social software users whose number of fans exceeds a preset second threshold value, and the active users are users whose posts on social software are continuous within multiple time periods; The obtaining of the behavior characteristic information of each user includes: obtaining the number of uplink operations of each user in each unit period within a specified period, wherein the specified period includes at least one unit period; The determining of a suspected stolen account based on the behavior characteristic information includes: Determine, based on the number of uplink operations of each user in each unit period within the specified period, a coefficient of variation of the number of uplink operations of each user within the specified period, wherein the coefficient of variation is a ratio between a standard deviation value of the number of uplink operations of the user within the specified period and an average value of the number of uplink operations of the user within the specified period; Filter out target type users from the users whose coefficient of variation is above a preset ratio threshold, wherein the target type includes at least one of star-chasing users, V users, and active users; The accounts of the remaining users whose coefficient of variation is above the preset ratio threshold are determined as suspected stolen accounts after filtering out the users.
2. The method according to claim 1, characterized in that: The uplink operation information includes stolen data fed back by the user; The obtaining of the behavior characteristic information of each user includes: obtaining the stolen data reported by each user on the social software; The determining of the suspected stolen account based on the behavior characteristic information includes: determining the suspected stolen account based on stolen data fed back by users on social software.
3. The method according to claim 2, characterized in that The stolen data reported by the user on the social software is used to determine the suspected stolen account, including: For any user: Determining whether the profile of the user contains stolen data reported by the user; If the profile of the user contains stolen data reported by the user, the account of the user is determined to be a suspected stolen account; If the user's profile does not contain the stolen data reported by the user, filter out the user's non-original posts and determine whether the user's posts contain the stolen data reported by the user; If the posting content of the user contains the stolen data reported by the user, the account of the user is determined to be a suspected stolen account.
4. The method according to claim 1, characterized in that: The uplink operation information includes the posting information of the user on the social software; The obtaining of the behavior characteristic information of each user includes: obtaining a preset number of posts of each user on the social software; The determining of a suspected stolen account based on the behavior characteristic information includes: For any user: Determine a special topic based on the preset number of posts, wherein the special topic includes at least one of promotional advertisements and illegal content, and the illegal content includes at least one of pornographic content, violent content, and gambling content; Based on the special topic, a suspected stolen account is determined.
5. The method according to claim 1, characterized in that The uplink operation information includes the user's interactive information on the social software. The obtaining of the behavior characteristic information of each user includes: obtaining the interaction information of each user on the social software, wherein the interaction information includes at least one of forwarding information, comment information and like information; The determining of a suspected stolen account based on the behavior characteristic information includes: For any user: Based on the interaction information, determine the interaction object account; Determining whether the interactive object account has fraud characteristics; In the case where the interactive object account has fraud characteristics, the user's account is determined to be a suspected stolen account.
6. The method according to claim 1, characterized in that The user information includes at least one of the user's geographic location information, the user's authentication information, and the user's posting frequency; Acquiring the behavior characteristic information of each user includes: acquiring uplink operation information of each user, and at least one of geographic location information, authentication information, and posting frequency of each user.
7. A device for identifying a stolen account, characterized in that: The device comprises: The acquisition module is used to obtain the target account that meets the specified policy; A matching module, used to match the target account with accounts in a pre-generated anti-account theft preparation library, wherein the anti-account theft preparation library includes suspected stolen accounts mined based on user behavior characteristics; A determination module, configured to determine that the target account is a stolen account if there is an account matching the target account in the anti-account theft preparation database; The acquisition module is further used to: acquire behavior characteristic information of each user, wherein the behavior characteristic information includes uplink operation information and user information; The determination module is further used to: determine a suspected stolen account based on the behavior feature information, and put the suspected stolen account into the anti-account theft preparation database; The uplink operation information includes the number of uplink operations performed by the user within a unit period; the user information includes user types, and the user types include star-chasing users, V users, and active users. The star-chasing users are social software users whose number of interactions with officially certified star users exceeds a preset first threshold value, the V users are officially certified social software users whose number of fans exceeds a preset second threshold value, and the active users are users whose posts on social software are continuous within multiple time periods; in the process of obtaining the behavioral characteristic information of each user, the acquisition module is used to: obtain the number of uplink operations of each user in each unit period within a specified period, and the specified period includes at least one unit period; accordingly, based on the behavior Characteristic information, in the process of determining a suspected stolen account, the determination module is used to: determine the coefficient of variation of the number of uplink operations of each user within the specified period based on the number of uplink operations of each user in each unit period within the specified period, wherein the coefficient of variation is the ratio of the standard deviation value of the number of uplink operations of the user within the specified period to the average value of the number of uplink operations of the user within the specified period; filter out users of the target type from users whose coefficient of variation is above a preset ratio threshold, wherein the target type includes at least one of star-chasing users, V users and active users; determine the accounts of users whose coefficient of variation is above the preset ratio threshold and the remaining users after filtering out the operations as suspected stolen accounts.
