Risk detection and processing method and device in user login process

Abstract

The invention provides a risk detection and processing method and device in a user login process, and the method comprises the steps: obtaining main authentication of login when a client opens a login page; if the main authentication is passed, judging whether man-machine verification is passed or not, and if the man-machine verification is not passed, judging whether man-machine verification is needed or not; if it is known that man-machine verification is needed, popping up a man-machine verification interface at the client, storing a man-machine verification result after the man-machine verification is passed, and turning to a processing step of judging whether the man-machine verification is passed; if it is known that the man-machine verification is allowed to be skipped or the response to the man-machine verification is passed, and the authentication mode of the main authentication is a first preset authentication mode, whether multi-factor authentication is needed or not is acquired; in response to the need of multi-factor authentication, further authentication is carried out according to a preset additional authentication mode, and login succeeds after authentication is passed; and in response to the fact that multi-factor authentication is not needed, login succeeds. According to the risk detection and processing method and device in the user login process provided by the invention, the risk detection and processing capability of the user login process is effectively improved, and the login security is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for detecting and processing risks in a user login process. Background Art

[0002] User login is a core feature in many online services and applications, which ensures that only authenticated users can access specific resources or services. The main purpose of this process is to ensure that only legitimate users can access sensitive data, perform specific operations or enjoy services.

[0003] However, there are a lot of risky behaviors in user login scenarios, such as machine account registration, non-personal login (such as stolen account login), etc. How to effectively detect and handle risks in the user login process has become an urgent problem to be solved. Summary of the invention

[0004] The present invention provides a method and device for detecting and processing risks in a user login process, so as to solve the defect in the prior art that there are a large number of risky behaviors in user login scenarios that cannot be effectively dealt with, and to effectively detect and process the risks in the user login process.

[0005] The present invention provides a risk detection and processing method in a user login process, comprising the following steps: in response to obtaining information that a client opens a login page, performing a primary authentication for login; in response to obtaining information that the primary authentication is passed, obtaining a result of whether the human-machine authentication has passed, and in response to the human-machine authentication not passing, obtaining a first risk control analysis result for judging whether the human-machine authentication is required; in response to knowing that the human-machine authentication is required according to the first risk control analysis result, popping up a human-machine authentication interface on the client, and saving the human-machine authentication result after the human-machine authentication passes, and going to the processing step of judging whether the human-machine authentication has passed; in response to knowing that the human-machine authentication is allowed to be skipped according to the first risk control analysis result or in response to the human-machine authentication having passed, and the authentication mode of the primary authentication is the first preset authentication mode, obtaining a second risk control analysis result for judging whether multi-factor authentication is required; in response to knowing that the multi-factor authentication is required according to the second risk control analysis result, further authentication is performed according to the preset additional authentication mode, and the login is successful after the authentication passes; in response to knowing that the multi-factor authentication is not required according to the second risk control analysis result, the login is successful.

[0006] According to a method for risk detection and processing in a user login process provided by the present invention, before determining whether human-machine verification has been passed, the method also includes: obtaining client environment information and user information; obtaining a first risk control analysis result for determining whether human-machine verification is required, including: inputting the client environment information and the user information into a pre-trained first risk control analysis model to obtain the first risk control analysis result; wherein the first risk control analysis model is trained with client environment information samples and user information samples as input, and with human-machine verification required, human-machine verification allowed to be skipped, and login prohibited as output labels.

[0007] According to a method for risk detection and processing in a user login process provided by the present invention, if the client performs user login through a browser, after the login is successful, the method also includes generating a trust token and a browser identifier and storing them in a cookie; the second risk control analysis result for determining whether multi-factor authentication is required includes: generating prior login reference data based on the trust token, the browser identifier and the client environment information, inputting the client environment information, the user information and the prior login reference data into a pre-trained second risk control analysis model to obtain the second risk control analysis result; wherein the second risk control analysis model is trained with client environment information samples, user information samples and online login reference data samples as input, and with multi-factor authentication required and multi-factor authentication not required as output labels.

[0008] According to a method for risk detection and processing in a user login process provided by the present invention, if the client performs user login through an application, the client environment information includes a device ID. After the login is successful, the method also includes binding the information passed by the multi-factor authentication with the device ID and adding it to the client environment information; the second risk control analysis result for determining whether multi-factor authentication is required includes: inputting the client environment information and the user information into a pre-trained third risk control analysis model to obtain the second risk control analysis result; wherein the third risk control analysis model is trained with client environment information samples and user information samples as input and with multi-factor authentication required and multi-factor authentication not required as output labels.

[0009] According to a method for risk detection and processing in a user login process provided by the present invention, in response to human-machine verification having passed, before obtaining the second risk control analysis result for determining whether multi-factor authentication is required, the method further includes: inputting the client environment information, the user information and the human-machine verification result into a pre-trained fourth risk control analysis model, obtaining a third risk control analysis result, and confirming that the human-machine verification result is reliable to execute subsequent steps, performing human-machine verification again or blocking login processing according to the third risk control result; wherein the fourth risk control analysis model is trained with client environment information samples, user information samples and human-machine verification result samples as input, and with confirming that the human-machine verification result is reliable, performing human-machine verification again or blocking login as output labels.

[0010] According to a method for risk detection and processing in a user login process provided by the present invention, the method also includes: in response to knowing that human-machine verification is allowed to be skipped according to the first risk control analysis result or in response to the human-machine verification having passed, and the authentication method of the main authentication is the second preset authentication method, the login is successful; wherein, the first preset authentication method includes an account and password authentication method, the second preset authentication method includes a text message verification code authentication method, a one-click login authentication method or a three-party login authentication method, and the preset additional authentication method includes a text message verification code authentication method.

[0011] According to a method for risk detection and processing in a user login process provided by the present invention, before the main authentication of the login is performed in response to obtaining information that the client opens the login page, the method also includes: in response to obtaining information that the client opens the registration page, obtaining client environment information, and obtaining a result of whether the human-machine verification has passed; in response to the human-machine verification failing, obtaining a fourth risk control analysis result for determining whether human-machine verification is required; in response to knowing that human-machine verification is required according to the fourth risk control analysis result, popping up a human-machine verification interface on the client, and saving the human-machine verification result after the human-machine verification passes, and going to the processing step of obtaining the client environment information; in response to knowing that human-machine verification is allowed to be skipped according to the fourth risk control analysis result or in response to the human-machine verification having passed, executing a user registration operation, and opening the login page on the client after successful registration; in response to knowing that registration is prohibited according to the fourth risk control analysis result, prohibiting user registration.

[0012] According to a risk detection and processing method in a user login process provided by the present invention, obtaining a fourth risk control analysis result for determining whether human-machine verification is required includes: inputting the client environment information into a pre-trained fifth risk control analysis model to obtain the fourth risk control analysis result; wherein the fifth risk control analysis model is trained with a client environment information sample as input and with human-machine verification required, human-machine verification allowed to be skipped, and registration prohibited as output labels.

[0013] The present invention also provides a risk detection and processing device in a user login process, comprising the following modules: a main authentication processing module, used to: in response to obtaining information that a client opens a login page, perform a main authentication for login; a first risk control analysis result acquisition module, used to: in response to obtaining information that the main authentication has passed, obtain a result of whether the human-machine authentication has passed, and in response to the human-machine authentication not passing, obtain a first risk control analysis result for determining whether the human-machine authentication is required; a human-machine authentication processing module, used to: in response to knowing that the human-machine authentication is required according to the first risk control analysis result, pop up a human-machine authentication interface on the client, and save the human-machine authentication result after the human-machine authentication passes. The result is transferred to the processing step of judging whether the human-machine verification has been passed; the second risk control analysis result acquisition module is used for: in response to learning that the human-machine verification is allowed to be skipped according to the first risk control analysis result or in response to the human-machine verification having passed, and the authentication method of the main authentication is the first preset authentication method, obtaining the second risk control analysis result for judging whether multi-factor authentication is required; the multi-factor authentication processing module is used for: in response to learning that multi-factor authentication is required according to the second risk control analysis result, further authentication is performed according to the preset additional authentication method, and the login is successful after the authentication is passed; in response to learning that multi-factor authentication is not required according to the second risk control analysis result, the login is successful.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the risk detection and processing method in the user login process as described in any one of the above-mentioned methods is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a risk detection and processing method in a user login process as described in any one of the above.

[0016] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the risk detection and processing methods in the user login process described above.

[0017] The risk detection and processing method and device in the user login process provided by the present invention perform the main authentication of login in response to obtaining the information that the client opens the login page, obtain the result of whether the human-machine authentication has passed in response to obtaining the information that the main authentication has passed, obtain the first risk control analysis result for judging whether the human-machine authentication is needed in response to the human-machine authentication not passing, pop up the human-machine authentication interface on the client in response to knowing that the human-machine authentication is needed according to the first risk control analysis result, save the human-machine authentication result after the human-machine authentication passes, and turn to the processing step of judging whether the human-machine authentication has passed, obtain the second risk control analysis result for judging whether the multi-factor authentication is needed in response to knowing that the human-machine authentication is allowed to be skipped according to the first risk control analysis result or in response to the human-machine authentication having passed and the authentication mode of the main authentication is the first preset authentication mode, and perform further authentication according to the preset additional authentication mode in response to knowing that the multi-factor authentication is needed according to the second risk control analysis result, and the login is successful after the authentication passes, and the login is successful in response to knowing that the multi-factor authentication is not needed according to the second risk control analysis result, so as to effectively improve the risk detection and processing capability of the user login process and improve the login security. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 This is one of the flow charts of the risk detection and processing method in the user login process provided by the present invention.

[0020] Figure 2 This is the second flow chart of the risk detection and processing method in the user login process provided by the present invention.

[0021] Figure 3 It is a schematic diagram of the registration process in the risk detection and processing method in the user login process provided by the present invention.

[0022] Figure 4 It is a structural schematic diagram of the risk detection and processing device in the user login process provided by the present invention.

[0023] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Figure 1 This is one of the flow charts of the risk detection and processing method in the user login process provided by the present invention, such as Figure 1 As shown, the method includes: Step S1: In response to obtaining information that the client opens a login page, a primary authentication for login is performed.

[0026] The risk detection and processing method in the user login process provided by the present invention is applied to the server. The server performs the main authentication of the login in response to obtaining the information that the client opens the login page. The relevant authentication methods of the main authentication can be directly displayed on the login page, and the main authentication of the login can be performed through the interaction between the server and the client. The authentication method of the main authentication can be, for example, an account password authentication method, a SMS verification code authentication method, a one-click login authentication method, or a three-party login authentication method.

[0027] After obtaining the information that the client has opened the login page, a unique identifier (loginTicketId) for the login operation can be generated and passed in the subsequent login process.

[0028] Step S2: in response to obtaining the information that the primary authentication has passed, obtaining the result of whether the human-machine verification has passed; in response to failing the human-machine verification, obtaining the first risk control analysis result for determining whether the human-machine verification is required.

[0029] In response to obtaining the information that the primary authentication has passed, the result of whether the human-machine verification has passed is further obtained. Among them, whether the human-machine verification has passed can be judged by judging whether there is a human-machine verification result and whether the data in the human-machine verification result indicates that the human-machine verification has passed. If there is a human-machine verification result and the data in the human-machine verification result indicates that the human-machine verification has passed, it is determined that the human-machine verification has passed; if there is no human-machine verification result or the data in the human-machine verification result indicates that the human-machine verification has failed, it is determined that the human-machine verification has not passed.

[0030] If the human-machine verification fails, a first risk control analysis result for determining whether human-machine verification is required is further obtained. The first risk control analysis result for determining whether human-machine verification is required can be obtained by calling a risk control strategy.

[0031] Step S3: In response to knowing that human-machine verification is required according to the first risk control analysis result, a human-machine verification interface pops up on the client, and the human-machine verification result is saved after the human-machine verification passes, and the process proceeds to the processing step of determining whether the human-machine verification has passed.

[0032] If it is known that human-machine verification is required according to the first risk control analysis result, the human-machine verification interface will pop up on the client, and the human-machine verification result will be saved after the human-machine verification is passed, and the process will be turned to the step of judging whether the human-machine verification has been passed. When judging whether the human-machine verification has been passed again, since the human-machine verification result exists and the human-machine verification result shows that the human-machine verification has been passed, it is confirmed that the human-machine verification has been passed.

[0033] When the human-machine verification interface pops up on the client, a human-machine verification interface of a preset human-machine verification method may pop up. The preset human-machine verification method may be triggering and verification of a slider, icon selection, text selection, spatial reasoning, nine-square grid, etc.

[0034] Step S4: In response to learning from the first risk control analysis result that human-machine verification is allowed to be skipped or in response to the human-machine verification having passed, and the authentication method of the main authentication is the first preset authentication method, a second risk control analysis result is obtained for determining whether multi-factor authentication is required.

[0035] If it is known from the first risk control analysis result that human-machine verification is allowed to be skipped or that human-machine verification has been confirmed to have passed, then when the authentication method of the primary authentication is the first preset authentication method, a second risk control analysis result for determining whether multi-factor authentication is required is obtained. The need for multi-factor authentication indicates that other authentication methods are required for further authentication. Among them, the first preset authentication method is a preset primary authentication method with a lower security level. The second risk control analysis result for determining whether multi-factor authentication is required can be obtained by calling the risk control policy.

[0036] Step S5: In response to learning that multi-factor authentication is required according to the second risk control analysis result, further authentication is performed according to a preset additional authentication method, and login is successful after authentication is passed; in response to learning that multi-factor authentication is not required according to the second risk control analysis result, login is successful.

[0037] If it is known from the second risk control analysis result that multi-factor authentication is required, further authentication is performed according to the preset additional authentication method, and the login is successful after the authentication is passed. If it is known from the second risk control analysis result that multi-factor authentication is not required, the login is successful.

[0038] The risk detection and processing method in the user login process provided by the present invention performs the main authentication of the login in response to obtaining the information that the client opens the login page, obtains the result of whether the human-machine authentication has passed in response to obtaining the information that the main authentication has passed, obtains the first risk control analysis result for judging whether the human-machine authentication is needed in response to the human-machine authentication not passing, pops up the human-machine authentication interface on the client in response to knowing that the human-machine authentication is needed according to the first risk control analysis result, saves the human-machine authentication result after the human-machine authentication passes, and turns to the processing step of judging whether the human-machine authentication has passed, obtains the second risk control analysis result for judging whether the multi-factor authentication is needed in response to knowing that the human-machine authentication is allowed to be skipped according to the first risk control analysis result or in response to the human-machine authentication having passed and the authentication mode of the main authentication is the first preset authentication mode, and obtains the second risk control analysis result for judging whether the multi-factor authentication is needed in response to knowing that the multi-factor authentication is needed according to the second risk control analysis result, performs further authentication according to the preset additional authentication mode, and succeeds in the login after the authentication passes, and succeeds in the login in response to knowing that the multi-factor authentication is not needed according to the second risk control analysis result, thereby effectively improving the risk detection and processing capability of the user login process and improving the login security.

[0039] According to a method for risk detection and processing in a user login process provided by the present invention, before determining whether human-machine verification has been passed, the method also includes: obtaining client environment information and user information; obtaining a first risk control analysis result for determining whether human-machine verification is required, including: inputting the client environment information and the user information into a pre-trained first risk control analysis model to obtain the first risk control analysis result; wherein the first risk control analysis model is trained with client environment information samples and user information samples as input, and with human-machine verification required, human-machine verification allowed to be skipped, and login prohibited as output labels.

[0040] Client environment information may include user agent (UA), device ID, client type, latitude and longitude, MAC address, device model, brand, battery level, HTTP Referer, etc. Client environment information can be transmitted to the server after being obtained by the client. Client environment information can be obtained before or after primary authentication. The server can obtain the corresponding user information through login credentials (such as user name, mobile phone number, WeChat authorization code, etc.). User information is obtained after primary authentication.

[0041] When obtaining the first risk control analysis result for determining whether human-machine verification is required, the client environment information and user information are input into the pre-trained first risk control analysis model to obtain the first risk control analysis result. The first risk control analysis model is trained with client environment information samples and user information samples as inputs and human-machine verification required, human-machine verification allowed to be skipped, and login prohibited as output labels.

[0042] Therefore, the first risk control analysis result may also indicate that login is prohibited. If the first risk control analysis result indicates that login is prohibited, the user is prohibited from logging in, and the processing of the login process ends.

[0043] The risk detection and processing method in the user login process provided by the present invention obtains client environment information and user information, inputs the client environment information and user information into a pre-trained first risk control analysis model, and obtains a first risk control analysis result, thereby improving the accuracy of the first risk control analysis result used to determine whether human-machine verification is required.

[0044] According to a method for risk detection and processing in a user login process provided by the present invention, if the client performs user login through a browser, after the login is successful, the method also includes generating a trust token and a browser identifier and storing them in a cookie; the second risk control analysis result for determining whether multi-factor authentication is required includes: generating prior login reference data based on the trust token, the browser identifier and the client environment information, inputting the client environment information, the user information and the prior login reference data into a pre-trained second risk control analysis model to obtain the second risk control analysis result; wherein the second risk control analysis model is trained with client environment information samples, user information samples and online login reference data samples as input, and with multi-factor authentication required and multi-factor authentication not required as output labels.

[0045] Compared with applications (APP), browsers lack device identification capabilities. APP can use device ID to uniquely identify a device, but browsers do not have a stable device ID solution and can use cookies to identify devices.

[0046] If the client logs in the user through the browser, after the login is successful, the method also includes generating a trust token and a browser identifier and storing them in a cookie. The trust token indicates the user's identification through multi-factor authentication, and the browser identifier is an ID generated based on various parameters of the browser. The client environment information can also include browser-related information, such as browser canvas properties. Comparing and verifying the trust token, browser identifier, and client environment information can prevent users from forging or absorbing the trust token and browser identifier of other devices into the current browser to skip multi-factor authentication.

[0047] Accordingly, if the client logs in through a browser, when obtaining the second risk control analysis result for determining whether multi-factor authentication is required, the previous login reference data is generated based on the trust token, browser identifier and client environment information, and the client environment information, user information and previous login reference data are input into the pre-trained second risk control analysis model to obtain the second risk control analysis result. Among them, the third risk control analysis model is trained with client environment information samples and user information samples as inputs and with multi-factor authentication required or not required as output labels.

[0048] The prior login reference data may include, for example, whether the trust token and the browser identifier match, and if so, whether the edit distance between the browser identifier and the UA and the browser canvas is small enough. If the trust token and the browser identifier match, and the edit distance between the browser identifier and the UA and the browser canvas is less than a preset distance threshold, it indicates that the risk is low.

[0049] The risk detection and processing method in the user login process provided by the present invention generates a trust token and a browser identifier and stores them in a cookie after the user logs in through a browser on the client side after the login is successful, generates previous login reference data based on the trust token, the browser identifier and the client side environment information, inputs the client side environment information, the user information and the previous login reference data into a pre-trained second risk control analysis model, obtains a second risk control analysis result, and improves the accuracy of the second risk control analysis result used to determine whether to perform multi-factor authentication when logging in through the browser.

[0050] According to a method for risk detection and processing in a user login process provided by the present invention, if the client performs user login through an application, the client environment information includes a device ID. After the login is successful, the method also includes binding the information passed by the multi-factor authentication with the device ID and adding it to the client environment information; the second risk control analysis result for determining whether multi-factor authentication is required includes: inputting the client environment information and user information into a pre-trained third risk control analysis model to obtain the second risk control analysis result; wherein the third risk control analysis model is trained with client environment information samples and user information samples as input and with multi-factor authentication required and multi-factor authentication not required as output labels.

[0051] The app can use the device ID to uniquely identify a device, and the device ID belongs to the client environment information. If the client logs in through the application, after the login is successful, the method also includes binding the information that the multi-factor authentication has passed with the device ID, and adding it to the client environment information. Then, at the next login, the information that the device corresponding to the device ID has previously passed the multi-factor authentication will be obtained to provide a reference for determining whether to perform multi-factor authentication in this login. When obtaining the second risk control analysis result for determining whether multi-factor authentication is required, the client environment information and user information are input into the pre-trained third risk control analysis model to obtain the second risk control analysis result; wherein, the third risk control analysis model is trained with client environment information samples and user information samples as input, and with multi-factor authentication required and multi-factor authentication not required as output labels.

[0052] The risk detection and processing method in the user login process provided by the present invention obtains a second risk control analysis result by inputting client environment information and user information into a pre-trained third risk control analysis model when the user logs in through an application on the client, thereby improving the accuracy of the second risk control analysis result used to determine whether to perform multi-factor authentication when logging in through the application.

[0053] According to a method for risk detection and processing in a user login process provided by the present invention, in response to human-machine verification having passed, before obtaining the second risk control analysis result for determining whether multi-factor authentication is required, the method further includes: inputting the client environment information, the user information and the human-machine verification result into a pre-trained fourth risk control analysis model, obtaining a third risk control analysis result, and confirming that the human-machine verification result is reliable to execute subsequent steps, performing human-machine verification again or blocking login processing according to the third risk control result; wherein the fourth risk control analysis model is trained with client environment information samples, user information samples and human-machine verification result samples as input, and with confirming that the human-machine verification result is reliable, performing human-machine verification again or blocking login as output labels.

[0054] In order to further improve risk detection capabilities and login security, after obtaining the result that the human-machine verification has passed, before obtaining the second risk control analysis result used to determine whether multi-factor authentication is required, the reliability of the human-machine verification result is further judged by calling the fourth risk control analysis model to identify forged human-machine verification results, and perform human-machine verification again or block login.

[0055] Among them, the client environment information, user information and human-machine verification results are input into the pre-trained fourth risk control analysis model to obtain the third risk control analysis result, and the human-machine verification result is confirmed to be reliable to execute the subsequent steps, perform human-machine verification again, or block login according to the third risk control result. If the third risk control result indicates that the human-machine verification result is reliable, the subsequent steps can be executed; if the third risk control result indicates that human-machine verification is to be performed again, the human-machine verification interface will be popped up again, and the human-machine verification method can be the same as or different from the last human-machine verification; if the third risk control result indicates that login is blocked, the user login process is exited.

[0056] The risk detection and processing method in the user login process provided by the present invention, in response to the human-machine verification having passed, before obtaining the second risk control analysis result for determining whether multi-factor authentication is required, inputs the client environment information, user information and the human-machine verification result into a pre-trained fourth risk control analysis model, obtains the third risk control analysis result, confirms that the human-machine verification result can reliably execute subsequent steps according to the third risk control result, performs human-machine verification again or blocks login, thereby further improving the risk detection capability and improving login security.

[0057] According to a method for risk detection and processing in a user login process provided by the present invention, the method also includes: in response to knowing that human-machine verification is allowed to be skipped according to the first risk control analysis result or in response to the human-machine verification having passed, and the authentication method of the main authentication is the second preset authentication method, the login is successful; wherein, the first preset authentication method includes an account and password authentication method, the second preset authentication method includes a text message verification code authentication method, a one-click login authentication method or a three-party login authentication method, and the preset additional authentication method includes a text message verification code authentication method.

[0058] If the security level of the primary authentication method is low, such as account and password authentication, it is further determined whether multi-factor authentication is required. If the security level of the primary authentication method is high, such as SMS verification code authentication, one-click login authentication, or three-party login authentication, you can log in directly without determining whether multi-factor authentication is required.

[0059] Therefore, in response to learning according to the first risk control analysis result that human-machine verification is allowed to be skipped or in response to the human-machine verification having passed, and the authentication method of the main authentication is the first preset authentication method, such as the user password authentication method, a second risk control analysis result is obtained for determining whether multi-factor authentication is required; in response to learning according to the second risk control analysis result that multi-factor authentication is required, further authentication is performed according to the preset additional authentication method, and login is successful after the authentication is passed, the preset additional authentication method includes the SMS verification code authentication method; in response to learning according to the second risk control analysis result that multi-factor authentication is not required, the login is successful.

[0060] In response to knowing that the human-machine verification is allowed to be skipped according to the first risk control analysis result or in response to the human-machine verification having passed, and the authentication method of the primary authentication is the second preset authentication method, the login is successful. Among them, the second preset authentication method includes a text message verification code authentication method, a one-click login authentication method, or a three-party login authentication method.

[0061] The risk detection and processing method in the user login process provided by the present invention responds to the fact that it is allowed to skip the human-machine verification according to the first risk control analysis result or in response to the human-machine verification having passed and the authentication method of the main authentication is the second preset authentication method, the login is successful, and the login efficiency is improved while ensuring the login security.

[0062] According to a method for risk detection and processing in a user login process provided by the present invention, the method for risk detection and processing in the user login process, before the main authentication of the login is performed in response to obtaining information that the client opens the login page, the method also includes: in response to obtaining information that the client opens the registration page, obtaining client environment information, and obtaining a result of whether the human-machine verification has passed; in response to the human-machine verification failing, obtaining a fourth risk control analysis result for determining whether human-machine verification is required; in response to knowing that human-machine verification is required according to the fourth risk control analysis result, popping up a human-machine verification interface on the client, and saving the human-machine verification result after the human-machine verification passes, and going to the processing step of obtaining the client environment information; in response to knowing that human-machine verification is allowed to be skipped according to the fourth risk control analysis result or in response to the human-machine verification having passed, executing a user registration operation, and opening the login page on the client after successful registration; in response to knowing that registration is prohibited according to the fourth risk control analysis result, prohibiting user registration.

[0063] Before executing user login, user registration needs to be performed first. In response to obtaining information that the client opens the registration page, the client environment information is obtained, and the result of whether the human-machine verification has passed is obtained; in response to the human-machine verification not passing, the fourth risk control analysis result for determining whether human-machine verification is required is obtained; in response to knowing that human-machine verification is required according to the fourth risk control analysis result, the human-machine verification interface is popped up on the client, and the human-machine verification result is saved after the human-machine verification passes, and the processing step of obtaining the client environment information is performed; in response to knowing that human-machine verification is allowed to be skipped according to the fourth risk control analysis result or in response to the human-machine verification having passed, the user registration operation is performed, and the login page is opened on the client after the registration is successful; in response to knowing that registration is prohibited according to the fourth risk control analysis result, the user registration is prohibited.

[0064] Among them, in the registration process, if SMS verification code registration is used, the hash value of the user's mobile phone number + the first verification code can be used as the unique identifier of the registration process and passed in the processing of the registration process.

[0065] The risk detection and processing method in the user login process provided by the present invention obtains the client environment information and obtains the result of whether the human-machine verification has passed in response to obtaining the information that the client opens the registration page. In response to the human-machine verification failing, a fourth risk control analysis result for judging whether the human-machine verification is needed is obtained. In response to knowing that the human-machine verification is needed according to the fourth risk control analysis result, a human-machine verification interface is popped up on the client, and the human-machine verification result is saved after the human-machine verification passes, and the processing step of obtaining the client environment information is turned to. In response to knowing that the human-machine verification is allowed to be skipped according to the fourth risk control analysis result or in response to the human-machine verification having passed, the user registration operation is executed, and the login page is opened on the client after the registration is successful. In response to knowing that the registration is prohibited according to the fourth risk control analysis result, the user registration is prohibited, thereby improving the security of the user registration and further ensuring the security of the user login.

[0066] According to a risk detection and processing method in a user login process provided by the present invention, obtaining a fourth risk control analysis result for determining whether human-machine verification is required includes: inputting the client environment information into a pre-trained fifth risk control analysis model to obtain the fourth risk control analysis result; wherein the fifth risk control analysis model is trained with a client environment information sample as input and with human-machine verification required, human-machine verification allowed to be skipped, and registration prohibited as output labels.

[0067] When obtaining the fourth risk control analysis result for determining whether human-machine verification is required, the client environment information is input into the pre-trained fifth risk control analysis model to obtain the fourth risk control analysis result; wherein the fifth risk control analysis model is trained with the client environment information sample as input and human-machine verification required, human-machine verification allowed to be skipped, and registration prohibited as output labels. The client environment information obtained during registration and the client environment information obtained during login may be the same or different.

[0068] The risk detection and processing method in the user login process provided by the present invention obtains the fourth risk control analysis result by inputting the client environment information into the pre-trained fifth risk control analysis model, thereby improving the accuracy of the fourth risk control analysis result used to determine whether to perform human-machine verification during registration.

[0069] According to a method for risk detection and processing in a user login process provided by the present invention, the method further includes: storing a first user behavior event and an authentication event in the user login process, and iteratively optimizing the first risk control analysis model, the second risk control analysis model, the third risk control analysis model, and the fourth risk control analysis model by analyzing the first user behavior event, the authentication event, and the human-machine verification result; The second user behavior event in the user registration process is stored, and the fifth risk control analysis model is iteratively optimized by analyzing the second user behavior event and the human-machine verification result.

[0070] Among them, authentication events include events generated during the main authentication and preset additional authentication processes, such as loginTicketId, authentication time, authentication method (including main authentication method and multi-factor authentication method), login terminal type, login system, environmental context information, etc. The first user behavior event includes loginTicketId, type of user behavior (such as human-machine verification, verification SMS verification code, verification account password), result of user behavior (verification success or verification failure), verification time, unique identifier (user identifier), and other fields.

[0071] The risk detection and processing method in the user login process provided by the present invention stores the first user behavior event and authentication event in the user login process, iteratively optimizes the first risk control analysis model, the second risk control analysis model, the third risk control analysis model and the fourth risk control analysis model by analyzing the first user behavior event, the authentication event and the human-machine verification result, stores the second user behavior event in the user registration process, and iteratively optimizes the fifth risk control analysis model by analyzing the second user behavior event and the human-machine verification result, thereby further improving the risk analysis capability and improving login security.

[0072] Figure 2 This is the second flow chart of the risk detection and processing method in the user login process provided by the present invention. Figure 2 As shown, the method includes: User login: indicates that the user opens the login page (including app page and browser page). When the login page is opened, the login system will generate a loginTicketId, which is used to string together a complete login process (including human-machine verification, primary authentication, and multi-factor authentication).

[0073] Obtain client environment information: Client environment information is obtained through the client or front-end. The environment information may include UA, device ID, client type, latitude and longitude, MAC address, device model, brand, battery power, refer, and other information.

[0074] Main authentication: login authentication process including one-click login, three-party login, short verification login, account and password login, etc.

[0075] Get user information: Get the corresponding user information through login credentials (such as user name, mobile phone number, WeChat authorization code, etc.) on the server.

[0076] Whether the human-machine verification is passed: The risk control strategy uses the human-machine verification credentials in the login credentials (which belong to the human-machine verification results) to determine whether the human-machine verification is passed. If there is no human-machine verification credentials or the human-machine verification fails, it is considered a failure; if the human-machine verification credentials are successfully verified, it is considered a pass.

[0077] Calling risk control strategy: After the user interacts on the page, the server calls the risk control interface to determine whether human-computer verification is required (passing loginTicketId, client environment information, and user information to the risk control strategy), and generates a record of human-computer interaction events.

[0078] Risk control result: The result returned by the risk control interface may be direct release, blocking login, or requiring human-machine verification (risk control will specify the required human-machine verification type, such as slider and click). If released, continue with the subsequent steps; if blocked, the login fails and a prompt is displayed; if human-machine verification is required, the specified type of human-machine verification will pop up.

[0079] Pop up the specified type of human-machine verification: including the server returning the specified human-machine verification type, the front end obtaining the human-machine verification question according to the type, and displaying it to the user.

[0080] User human-computer verification interaction: After the user completes the interaction with the human-computer verification question (completing the slider or clicking, etc.), the front end will carry the human-computer verification credentials and call the login interface again.

[0081] Risk control data recovery (retain processing capability): After completing human-machine verification, carry environmental information and user information, call the risk control interface again to recover the human-machine verification results, and retain the ability to handle them again, further determine whether the human-machine credentials are forged (transmit to the risk control interface loginTicketId, environmental information, user information, human-machine verification results), and generate human-machine interaction event records. If it is identified that the user uses forged human-machine credentials, the user will be dealt with by popping up the human-machine verification interface again or blocking login, thereby preventing and controlling the behavior of machine simulation account registration (including silent registration) in all aspects.

[0082] Calling the risk control multi-factor authentication strategy: The primary authentication method with a low security level, such as account and password login, is the main way to log in by stealing accounts. In such cases, the risk control strategy will be called again to determine whether multi-factor authentication is required.

[0083] Risk control result: The result returned by the risk control interface may be the need for multi-factor authentication or release.

[0084] Supplement SMS verification code: If the risk control interface returns that multi-factor authentication is required, the user will be asked to supplement the SMS verification code, including popping up the SMS verification code input box, sending a text message to the user, etc.

[0085] SMS verification result: triggered when the user supplements the SMS verification code and clicks login again. If the SMS verification code passes the verification, the login is successful; if it fails, you need to go through the multi-factor authentication process again, including obtaining the SMS verification code, sending SMS, and verifying the SMS verification code.

[0086] Compared with APP, browser lacks device identification capability. APP can use device ID to uniquely identify a device, but browser does not have a stable device ID solution, so it uses cookies to identify devices. On PC browser or mobile browser, after the user completes the main authentication, the enhanced authentication judgment, interaction, and verification process is completed, and a text message verification record is generated. After completing the overall login process (after successful login), an authentication event record is generated. On the APP side, after the user completes the main authentication, the enhanced authentication judgment, interaction, and verification process is completed, and a text message verification record is generated. After completing the overall login process (after successful login), an authentication event record is generated. After logging in on the APP side, the result of passing the multi-factor authentication will be bound to the device ID.

[0087] Trust token and browser ID: These two are used to indicate that the user has passed multi-factor authentication on the current device. The trust token indicates that the user has passed multi-factor authentication, and the browser ID is an ID generated based on various parameters of the browser.

[0088] Store in cookies: Store the trust token and browser ID in the browser cookie. The next time you log in, these two values ​​will be passed into the risk control policy as environmental information, and will serve as one of the criteria for determining whether the user needs multi-factor authentication. The two values ​​will be compared and verified with other environmental information to prevent users from forging or sucking the trust token and browser ID of other devices into the current browser, thereby skipping multi-factor authentication.

[0089] Risk control recovery results include real-time recovery (such as human-machine verification results) and big data offline recovery (such as authentication events and user behavior events). All data can be associated and compared through loginTicketId to provide data support for risk control strategy iteration.

[0090] Figure 3 Schematic diagram of the registration process in the risk detection and processing method in the user login process provided by the present invention. Figure 3 As shown, the method includes: User registration: indicates that the user opens the registration page (including app page and browser page) Obtaining environmental information: Environmental information is obtained through the client or front-end. Environmental information includes UA, device ID, client type, latitude and longitude, MAC address, device model, brand, battery power, refer, and other information.

[0091] Whether the human-machine verification is passed: By obtaining the human-machine verification credentials, it is determined whether the human-machine verification is passed. If there is no human-machine verification credential or the human-machine verification fails, it is considered as failed; if the human-machine verification credential verification is successful, it is considered as passed.

[0092] Calling risk control strategy: passing the client environment information to the risk control interface, and the risk control interface returns the result.

[0093] Risk control result: The result returned by risk control may be direct release, registration blocking, or human-machine verification required (risk control will specify the required human-machine verification type, such as slider and click). If released, the registration operation will be carried out; if blocked, the registration will fail and a prompt will be displayed; if human-machine verification is required, the specified type of human-machine verification will pop up.

[0094] Pop up the specified type of human-machine verification: including the server returning the specified human-machine verification type, the front end obtaining the human-machine verification question according to the type, and displaying it to the user.

[0095] User human-computer verification interaction: After the user completes the interaction with the human-computer verification question (completing the slider or clicking, etc.), the front end will carry the human-computer verification credentials and call the registration interface again.

[0096] Because there is no login state after registration, and existing accounts are not allowed to register, there is no need to prevent situations such as account theft and other situations where someone else is not the real person. Only human-machine verification is needed to prevent scenarios where non-human registration is performed by machines or scripts.

[0097] The risk detection and processing device in the user login process provided by the present invention is described below. The risk detection and processing device in the user login process described below and the risk detection and processing method in the user login process described above can be referenced to each other.

[0098] Figure 4 Schematic diagram of the structure of the risk detection and processing device in the user login process provided by the present invention. Figure 4As shown, the device includes a main authentication processing module 10, a first risk control analysis result acquisition module 20, a human-machine verification processing module 30, a second risk control analysis result acquisition module 40 and a multi-factor authentication processing module 50, wherein: the main authentication processing module 10 is used to: in response to obtaining information that the client opens the login page, perform the main authentication of the login; the first risk control analysis result acquisition module 20 is used to: in response to obtaining information that the main authentication is passed, obtain the result of whether the human-machine verification has passed, and in response to the human-machine verification not passing, obtain the first risk control analysis result for determining whether the human-machine verification is required; the human-machine verification processing module 30 is used to: in response to knowing that the human-machine verification is required according to the first risk control analysis result, pop up the human-machine verification prompt on the client verification interface, and saves the human-machine verification result after the human-machine verification is passed, and goes to the processing step of judging whether the human-machine verification has been passed; the second risk control analysis result acquisition module 40 is used for: in response to knowing that the human-machine verification is allowed to be skipped according to the first risk control analysis result or in response to the human-machine verification having passed, and the authentication method of the main authentication is the first preset authentication method, obtaining the second risk control analysis result for judging whether multi-factor authentication is required; the multi-factor authentication processing module 50 is used for: in response to knowing that multi-factor authentication is required according to the second risk control analysis result, further authentication is performed according to the preset additional authentication method, and the login is successful after the authentication is passed; in response to knowing that multi-factor authentication is not required according to the second risk control analysis result, the login is successful.

[0099] The risk detection and processing device in the user login process provided by the present invention performs the main authentication of the login in response to obtaining the information that the client opens the login page, obtains the result of whether the human-machine authentication has passed in response to obtaining the information that the main authentication has passed, obtains the first risk control analysis result for judging whether the human-machine authentication is needed in response to the human-machine authentication not passing, pops up the human-machine authentication interface on the client in response to knowing that the human-machine authentication is needed according to the first risk control analysis result, saves the human-machine authentication result after the human-machine authentication passes, and turns to the processing step of judging whether the human-machine authentication has passed, obtains the second risk control analysis result for judging whether the multi-factor authentication is needed in response to knowing that the human-machine authentication is allowed to be skipped according to the first risk control analysis result or in response to the human-machine authentication having passed and the authentication mode of the main authentication is the first preset authentication mode, and performs further authentication according to the preset additional authentication mode in response to knowing that the multi-factor authentication is needed according to the second risk control analysis result, and the login is successful after the authentication passes, and the login is successful in response to knowing that the multi-factor authentication is not needed according to the second risk control analysis result, thereby effectively improving the risk detection and processing capability of the user login process and improving the login security.

[0100] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5As shown, the electronic device may include: a processor (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530 and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call the logic instructions in the memory 530 to execute the risk detection and processing method in the user login process, which method includes: in response to obtaining information that the client opens the login page, performing the main authentication of the login; in response to obtaining the information that the main authentication is passed, obtaining the result of whether the human-machine authentication has passed, and in response to the human-machine authentication not passing, obtaining the first risk control analysis result for judging whether the human-machine authentication is needed; in response to knowing that the human-machine authentication is needed according to the first risk control analysis result, popping up the human-machine authentication interface on the client, and saving the human-machine authentication result after the human-machine authentication passes, and going to the processing step of judging whether the human-machine authentication has passed; in response to knowing that the human-machine authentication is allowed to be skipped according to the first risk control analysis result or in response to the human-machine authentication having passed, and the authentication method of the main authentication is the first preset authentication method, obtaining the second risk control analysis result for judging whether multi-factor authentication is needed; in response to knowing that multi-factor authentication is needed according to the second risk control analysis result, further authentication is performed according to the preset additional authentication method, and the login is successful after the authentication passes; in response to knowing that multi-factor authentication is not needed according to the second risk control analysis result, the login is successful.

[0101] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0102] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the risk detection and processing methods in the user login process provided by the above methods, the method comprising: in response to obtaining information that the client opens a login page, performing a primary authentication for the login; in response to obtaining information that the primary authentication has passed, obtaining a result of whether the human-machine authentication has passed, and in response to the human-machine authentication not passing, obtaining a first risk control analysis result for determining whether the human-machine authentication is required; in response to knowing that human-machine authentication is required according to the first risk control analysis result, In response to the first risk control analysis result that it is known that skipping human-machine verification is allowed or in response to the human-machine verification having passed, and the authentication method of the main authentication is the first preset authentication method, a second risk control analysis result for determining whether multi-factor authentication is required is obtained; in response to the second risk control analysis result that multi-factor authentication is required, further authentication is performed according to the preset additional authentication method, and the login is successful after the authentication is passed; in response to the second risk control analysis result that multi-factor authentication is not required, the login is successful.

[0103] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the risk detection and processing methods in the user login process provided by the above methods, the method comprising: in response to obtaining information that a client opens a login page, performing a primary authentication for login; in response to obtaining information that the primary authentication has passed, obtaining a result of whether the human-machine authentication has passed, and in response to the human-machine authentication not passing, obtaining a first risk control analysis result for determining whether human-machine authentication is required; in response to knowing that human-machine authentication is required according to the first risk control analysis result, popping up a human-machine authentication prompt on the client; The method of the first risk control analysis is to determine whether multi-factor authentication is required; in response to the first risk control analysis result, it is known that multi-factor authentication is required, and the second risk control analysis result is used to determine whether multi-factor authentication is required; in response to the second risk control analysis result, it is known that multi-factor authentication is required, and further authentication is performed according to the preset additional authentication method, and the login is successful after the authentication is passed; in response to the second risk control analysis result, it is known that multi-factor authentication is not required, and the login is successful.

[0104] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0105] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for risk detection and processing in a user login process, characterized in that: include: In response to obtaining information that the client opens the login page, performing primary authentication of the login; In response to obtaining the information that the primary authentication has passed, obtaining a result of whether the human-machine authentication has passed, and in response to failing the human-machine authentication, obtaining a first risk control analysis result for determining whether the human-machine authentication is required; In response to knowing that human-machine verification is required according to the first risk control analysis result, a human-machine verification interface is popped up on the client, and after the human-machine verification is passed, the human-machine verification result is saved, and the process of determining whether the human-machine verification has been passed is performed; In response to knowing that skipping human-machine authentication is allowed according to the first risk control analysis result or in response to the human-machine authentication having passed, and the authentication method of the primary authentication is the first preset authentication method, obtaining a second risk control analysis result for determining whether multi-factor authentication is required; In response to learning that multi-factor authentication is required according to the second risk control analysis result, further authentication is performed according to a preset additional authentication method, and the login is successful after the authentication is passed; In response to learning, based on the second risk control analysis result, that multi-factor authentication is not required, the login is successful.

2. The risk detection and processing method in the user login process according to claim 1 is characterized in that: Before determining whether the human-machine verification has been passed, the method further includes: obtaining client environment information and user information; The obtaining of the first risk control analysis result for determining whether human-machine verification is required includes: Inputting the client environment information and the user information into a pre-trained first risk control analysis model to obtain the first risk control analysis result; Among them, the first risk control analysis model is trained with client environment information samples and user information samples as input, and with human-machine verification required, human-machine verification allowed to be skipped, and login prohibited as output labels.

3. The risk detection and processing method in the user login process according to claim 1 is characterized in that: If the client performs user login via a browser, after successful login, the method further includes generating a trust token and a browser identifier and storing them in a cookie; The obtaining of a second risk control analysis result for determining whether multi-factor authentication is required includes: generating prior login reference data based on the trust token, the browser identifier and the client environment information, inputting the client environment information, the user information and the prior login reference data into a pre-trained second risk control analysis model to obtain the second risk control analysis result; The second risk control analysis model is trained with client environment information samples, user information samples and online login reference data samples as inputs, and with multi-factor authentication required and multi-factor authentication not required as output labels.

4. The risk detection and processing method in the user login process according to claim 1 is characterized in that: If the client performs user login through an application, the client environment information includes a device id, and after successful login, the method further includes binding the information passed by the multi-factor authentication with the device id and adding it to the client environment information; The obtaining of a second risk control analysis result for determining whether multi-factor authentication is required includes: Inputting the client environment information and the user information into a pre-trained third risk control analysis model to obtain the second risk control analysis result; The third risk control analysis model is trained with client environment information samples and user information samples as inputs and with multi-factor authentication required or not required as output labels.

5. The risk detection and processing method in the user login process according to claim 1 is characterized in that: In response to the human-machine verification having passed, before obtaining the second risk control analysis result for determining whether multi-factor authentication is required, the method further includes: Input the client environment information, the user information and the human-machine verification result into a pre-trained fourth risk control analysis model to obtain a third risk control analysis result, and confirm that the human-machine verification result is reliable to execute subsequent steps, perform human-machine verification again or block login according to the third risk control result; Among them, the fourth risk control analysis model is trained with client environment information samples, user information samples and human-machine verification result samples as input, and with output labels such as confirming the reliability of the human-machine verification results, performing human-machine verification again or preventing login.

6. The risk detection and processing method in the user login process according to claim 1 is characterized in that: The method further comprises: In response to knowing that skipping human-machine verification is allowed according to the first risk control analysis result or in response to passing human-machine verification and the authentication method of the primary authentication is the second preset authentication method, the login is successful; Among them, the first preset authentication method includes an account and password authentication method, the second preset authentication method includes a text message verification code authentication method, a one-click login authentication method or a three-party login authentication method, and the preset additional authentication method includes a text message verification code authentication method.

7. The risk detection and processing method in the user login process according to claim 1 is characterized in that: Before performing the primary authentication of the login in response to obtaining the information that the client opens the login page, the method further includes: In response to obtaining information that the client opens a registration page, obtaining client environment information, and obtaining a result of whether human-machine verification has passed; In response to the human-machine verification failing, obtaining a fourth risk control analysis result for determining whether the human-machine verification is required; In response to knowing that human-machine verification is required according to the fourth risk control analysis result, a human-machine verification interface is popped up on the client, and after the human-machine verification passes, the human-machine verification result is saved, and the process proceeds to the step of obtaining the client environment information; In response to knowing that skipping human-machine verification is allowed according to the fourth risk control analysis result or in response to passing the human-machine verification, performing a user registration operation, and opening the login page on the client after successful registration; In response to learning that registration is prohibited according to the fourth risk control analysis result, user registration is prohibited.

8. The risk detection and processing method in the user login process according to claim 7 is characterized in that: The obtaining of the fourth risk control analysis result for determining whether human-machine verification is required includes: Inputting the client environment information into a pre-trained fifth risk control analysis model to obtain the fourth risk control analysis result; Among them, the fifth risk control analysis model is trained with client environment information samples as input and with human-machine verification required, human-machine verification allowed to be skipped, and registration prohibited as output labels.

9. A risk detection and processing device in a user login process, characterized in that: include: The main authentication processing module is used to: in response to obtaining the information that the client opens the login page, perform the main login authentication; A first risk control analysis result acquisition module is used to: in response to obtaining the information that the primary authentication has passed, obtain the result of whether the human-machine verification has passed, and in response to the human-machine verification not passing, obtain the first risk control analysis result used to determine whether the human-machine verification is required; A human-machine verification processing module, configured to: in response to learning that human-machine verification is required according to the first risk control analysis result, pop up a human-machine verification interface on the client, save the human-machine verification result after the human-machine verification passes, and proceed to the processing step of determining whether the human-machine verification has passed; A second risk control analysis result acquisition module is used to: in response to knowing that it is allowed to skip the human-machine verification according to the first risk control analysis result or in response to the human-machine verification having passed, and the authentication method of the primary authentication is the first preset authentication method, acquire a second risk control analysis result for determining whether multi-factor authentication is required; A multi-factor authentication processing module, configured to: in response to learning that multi-factor authentication is required according to the second risk control analysis result, perform further authentication according to a preset additional authentication method, and log in successfully after the authentication is passed; In response to learning, based on the second risk control analysis result, that multi-factor authentication is not required, the login is successful.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the risk detection and processing method in the user login process as described in any one of claims 1 to 8 is implemented.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the risk detection and processing method in the user login process as described in any one of claims 1 to 8 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the risk detection and processing method in the user login process as described in any one of claims 1 to 8 is implemented.

Images