Intelligent short message identity verification system

Abstract

The invention discloses an intelligent short message identity verification system, and relates to the technical field of identity verification, and the system comprises a data collection module which collects user historical behavior data in real time and encrypts and stores the user historical behavior data; the behavior mode prediction module preprocesses the data to construct a model prediction behavior, and identifies risks; the verification code generation and adjustment module adjusts the strength and duration of the verification code according to the prediction result; the verification execution module sends and verifies the verification code; the feedback learning module optimizes a prediction model and a verification code related algorithm according to an actual result; through multi-dimensional data acquisition and behavior mode prediction, dynamic risk analysis, accurate identification of high-risk operation, automatic adjustment of verification code complexity and duration, and balance of security and convenience, the system is continuously optimized by means of a feedback learning module, a model and an algorithm are dynamically updated according to user behaviors, and a verification code refresh mechanism is combined, so that the user experience is improved. Intelligent identity verification is realized, and user experience and system adaptability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of identity verification, in particular to an intelligent short message identity verification system. BACKGROUND

[0002] In today's digital age, with the rapid development of Internet technology and the widespread popularity of mobile devices, users' activities in the network space are increasingly frequent, covering socializing, shopping, financial transactions and other aspects. However, the openness and anonymity of the network environment also pose a great challenge to user information security. Identity verification, as the first line of defense to protect user account security, is self-evident in its importance. Traditional identity verification methods, such as static passwords and picture verification codes, have been difficult to meet the needs of the current complex and changing network environment. Static passwords are easy to steal or guess, while picture verification codes may be ineffective due to the development of automated recognition technology. Therefore, the development of a more secure, efficient and intelligent identity verification system has become an urgent need in the industry. The intelligent short message identity verification system emerged as the times required. It provides more reliable identity verification services for users by combining user behavior pattern analysis, risk assessment and dynamic verification code technology.

[0003] Traditional identity verification technologies, such as methods based on static passwords and picture verification codes, have many limitations. First, static passwords have low security. Users often choose simple and easy-to-guess passwords for convenience, or use the same password on different websites, which greatly increases the risk of account theft. Second, although picture verification codes improve the difficulty of automated recognition to some extent, their defense effect gradually weakens with the advancement of image recognition technology. In addition, traditional verification code systems lack in-depth analysis of user behavior and cannot dynamically adjust verification strategies based on users' historical behavior and current operating environment, resulting in verification processes that are either too simple and easy to bypass, or too complex and affect user experience. In particular, when facing advanced persistent threats and social engineering attacks, traditional identity verification technologies are not up to the task and are difficult to effectively resist these complex and varied attack methods. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art and provide an intelligent short message identity verification system. Through multi-dimensional data collection and behavior pattern prediction, combined with a two-dimensional risk identification algorithm, the system dynamically adjusts the complexity and timeliness of the verification code, realizes precise risk prevention and control, and covers five major modules: data collection, behavior prediction, verification code generation, verification execution and feedback learning. The system can identify abnormal operation scenarios in real time and adaptively optimize verification strategies.

[0005] To solve the above technical problems, the present application provides the following technical solution: an intelligent short message identity verification system, which comprises: Data collection module: interact with user's various application programs and systems, collect user's historical behavior data in real time, including login time, operation frequency, usage habits, transaction records, geographic location information, all collected data are stored in a secure database after encryption; Behavior pattern prediction module: preprocess the collected historical behavior data, build a user behavior pattern prediction model to predict the user's future possible behavior pattern, and use a two-dimensional risk identification algorithm to identify potential security risk scenarios; Verification code generation and adjustment module: according to the results obtained by the behavior pattern prediction module, adjust the strength of the verification code through the verification code complexity algorithm; adjust the effective time length of the verification code through the verification code time effectiveness algorithm; Verification execution module: when the user initiates an operation request, send a verification code to the user's registered mobile phone number through the SMS channel, display the verification code input box, related prompts and effective time length countdown on the operation interface, prompt the user to input within the specified time after receiving the SMS, and compare the SMS verification code with the system generated verification code after the user inputs, decide whether to allow or reject the user's operation request according to the verification result; Feedback learning module: according to the user's actual operation and the verification result of the verification execution module, optimize and update the prediction model of the behavior pattern prediction module and the algorithm of the verification code generation and adjustment module.

[0006] Further, in the data collection module, the collected data includes login data covering time, location, device and browser type; operation data including clicking buttons, browsing pages and operation frequency interval; transaction data involving amount, object, time and payment method; and geographic location data obtained through device GPS or network positioning and recorded location change track.

[0007] Further, in the behavior pattern prediction module, the calculation formula of the user behavior pattern prediction model is: , wherein, is the user behavior credibility index, which is used to measure the matching degree of the current behavior and the historical normal behavior, is the matching times of the user's current behavior and the historical stable behavior, is the total sample size of the user's historical behavior, is the recent operation frequency stability coefficient, is the system preset baseline stability threshold, is the physical distance between the current operation location and the commonly used region, is the system set maximum regional deviation threshold, is the weight coefficient, which is manually set or determined by data training according to the scene security demand, and .

[0008] Further, in the behavior pattern prediction module, the calculation formula of the two-dimensional risk identification algorithm is: , is the identification result of the potential security risk scenario, represents that the potential security risk scenario is identified, represents that the potential security risk scenario is not identified, is the user behavior credibility index, is the physical distance between the current operation site and the commonly used region, is the average value of the physical distance between the user historical operation site and the commonly used region, is the standard deviation of the physical distance between the user historical operation site and the commonly used region, is the current operation frequency, is the average value of the user historical operation frequency, is the standard deviation of the user historical operation frequency, when and the deviation degree of the physical distance between the current operation site and the commonly used region is greater than 2, or and the deviation degree of the current operation frequency is greater than 2, it is determined that the potential security risk scenario is identified.

[0009] Further, in the behavior pattern prediction module, the identified potential security risk scenario includes: the same account is logged in on different network environments at the same time; continuous login attempts in a non-regular time period; operations initiated in multiple regions in a short time in a single session; large transaction requests with a difference in historical transaction amount exceeding a preset threshold; multiple changes of bound mobile phone numbers or bank card information in a short time; account information modification immediately after login, and the modification content involves core identity information; sensitive operation requests in a non-regular language environment.

[0010] Further, in the verification code generation and adjustment module, the calculation formula of the verification code complexity algorithm is: , wherein is the verification code complexity level, is the user behavior credibility index, is the identification result of the potential security risk scenario, 1=risk, 0=no risk, level 1: pure digital verification code, length 4-6 digits, no interference line / noise; level 2: mixed with numbers and lowercase letters, length 6-8 digits, with slight interference lines; level 3: mixed with numbers, uppercase and lowercase letters, and special symbols, length 8-10 digits, with interference lines, noise, and character skew.

[0011] Further, in the verification code generation and adjustment module, the calculation formula of the verification code time limit algorithm is: , wherein: is the final effective duration, The basic duration corresponding to the complexity level, The risk coefficient is set according to the risk scenario severity, and the value range , The risk weight is set according to the risk scenario severity, and the value range .

[0012] Further, the specific steps of the verification execution module for performing verification operations are as follows: after receiving a user operation request, an instruction is sent to the verification code generation and adjustment module to obtain the verification code and the valid duration information; the verification code is sent to the user's registered mobile phone through the SMS channel; the user operation interface displays the verification code input box, related prompts and the valid duration countdown; the user input SMS verification code is monitored, and the input content is captured after the user submits; the verification code is checked: check whether the SMS verification code source mobile number is consistent with the user's registered mobile number; check whether it is submitted within the valid duration, and if not, determine that the verification fails; compare the user input SMS verification code with the system generated verification code; according to the verification result, send an operation permission or rejection instruction to the business system; when the verification fails, record the failure reason and prompt the user with SMS verification code error or verification timeout information.

[0013] Further, in the verification execution module, when the verification code input box is displayed, the SMS verification code validity period countdown is displayed synchronously; when the remaining duration is less than 20% of the total valid duration, a SMS verification code expiration prompt is sent to the user.

[0014] Further, in the verification code generation and adjustment module, the verification code refresh mechanism is provided, specifically: User-initiated refresh: when the user clicks the refresh verification code button on the operation interface, a new verification code is generated; the new verification code is sent to the user's registered mobile phone through the SMS channel, and the valid duration countdown is reset; Error-triggered refresh: when the system detects that the user has input the wrong verification code for two consecutive times, a new verification code is automatically generated; the new verification code is sent to the user's registered mobile phone through the SMS channel; the verification code complexity level is increased by one level, and the highest level does not exceed level 3; Reset the valid duration countdown; Single mobile phone sending frequency limit: the same registered mobile phone can receive a maximum of two verification code SMSs within 60 seconds.

[0015] Compared with the prior art, the intelligent SMS identity verification system has the following beneficial effects: One, the application improves the security and accuracy of identity verification through multi-dimensional data collection and behavior pattern prediction. The system collects historical behavior data of users in real time, covering login time, geographic location, and operation frequency information. The dual-dimensional risk identification algorithm is used to dynamically analyze potential security risk scenarios. Through the quantitative calculation of user behavior credibility index and risk deviation degree, the system can accurately identify high-risk operations and automatically adjust the complexity and effective duration of the verification code. This dynamic verification mechanism effectively reduces the risk of traditional static verification code being cracked or bypassed, while avoiding the decline of user experience caused by overly complex verification, achieving a balance between security and convenience.

[0016] Two, the application realizes the continuous optimization and adaptive adjustment of the system through the feedback learning module. The system dynamically updates the behavior pattern prediction model and the verification code generation algorithm according to the actual operation feedback of users. For users who frequently trigger high-risk scenarios, the complexity level of the verification code is gradually increased, while for users with stable behavior, low complexity verification is maintained. In addition, the verification code refresh mechanism and the single mobile phone sending frequency limit further enhance the fault tolerance and anti-attack ability of the system. When the user continuously inputs errors, the complexity is automatically increased and the effective duration is reset to avoid malicious probing. This closed-loop optimization based on user behavior and system feedback makes the identity verification process more intelligent, improving user experience and long-term adaptability of the system.

[0017] Other advantages, objects, and features of the application will be set forth in part in the following specification, and in part will become apparent to those skilled in the art from a study of the following specification, or can be learned from practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can be obtained from these drawings without creative labor.

[0019] Figure 1 Flowchart of intelligent short message identity verification system; Figure 2 Module interaction architecture diagram of intelligent short message identity verification system; Figure 3 Detail diagram of intelligent short message identity verification execution flow. DETAILED DESCRIPTION

[0020] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined inventive purpose, the specific embodiments, structures, features and effects thereof according to the present application are described in detail below in combination with the drawings and preferred embodiments.

[0021] Embodiment one: User daily login common shopping APP scene.

[0022] Data acquisition module: when the user opens the commonly used shopping APP and clicks to log in, the data acquisition module starts immediately, which establishes a secure interaction channel with the APP background system, and collects the key data of this login in real time, including login time (such as Wednesday afternoon 3 o'clock, which belongs to the user's daily shopping active period), login device (the user's commonly used smart phone, the device model is consistent with the historical record), browser type (the built-in browser of the APP, which is a commonly used type); At the same time, record the user's operation behavior, including the interval of clicking the login button (consistent with the historical operation habit); through the device GPS, the current geographic position (the user's commonly used home position) is obtained, and the historical appearance frequency of this position is automatically associated. These data cover the core dimensions of user behavior, which can provide a comprehensive basis for subsequent behavior analysis, and all data are encrypted before transmission and storage to ensure user privacy and data security. Finally, the encrypted data is stored in a secure database to prevent information leakage.

[0023] Behavior pattern prediction module: after receiving the encrypted data transmitted by the data acquisition module, the data is first decrypted and preprocessed (such as eliminating invalid data and unifying data format), and then the effective data is analyzed in depth. This module calls the user behavior pattern prediction model to compare the data of the current login time, device, location, and operation frequency with the historical normal behavior data to calculate the user behavior credibility index (UBI), and the calculation formula is: , wherein, UBI is the user behavior credibility index, which is used to measure the matching degree of the current behavior and the historical normal behavior, is the matching number of the user's current behavior and the historical stable behavior, is the total sample size of the user's historical behavior, is the recent operation frequency stability coefficient, is the system preset baseline stability threshold, is the physical distance between the current operation location and the commonly used region, is the maximum regional deviation threshold set by the system, is the weight coefficient, which is manually set or determined by data training according to the scene safety requirements, and Since all the data of this login are highly matched with the historical normal behavior, the model determines that the current behavior is highly credible. Subsequently, the module calls the two-dimensional risk identification algorithm, and the calculation formula of the two-dimensional risk identification algorithm is: , potential security risk scenario, potential security risk scenario is identified, potential security risk scenario is not identified, user behavior credibility index, physical distance between current operation location and frequently used region, average value of physical distance between user historical operation location and frequently used region, standard deviation of physical distance between user historical operation location and frequently used region, current operation frequency, average value of user historical operation frequency, standard deviation of user historical operation frequency, when and the deviation degree of the physical distance between the current operation location and the frequently used region is greater than 2, or and the deviation degree of the current operation frequency is greater than 2, it is determined that the potential security risk scenario is identified, and the deviation degree of the operation location is further investigated from the operation frequency stability dimension. Since no abnormal features are found, the potential security risk scenario recognition result (SRS) is determined to be 0. This process is verified by double checking, which can accurately identify normal user behavior and detect potential risks in advance, providing a scientific basis for subsequent verification strength adjustment.

[0024] Verification code generation and adjustment module: after receiving the result (UBI≥0.7 and SRS=0) of the behavior pattern prediction module, the verification code complexity algorithm is called to adjust the strength of the verification code. The calculation formula of the verification code complexity algorithm is: wherein is the verification code complexity level, is the user behavior credibility index, is the potential security risk scenario recognition result, 1=risk, 0=no risk, level 1: pure digital verification code, length 4-6 digits, no interference line / noise; level 2: mixed with numbers and lowercase letters, length 6-8 digits, with slight interference lines; level 3: mixed with numbers, uppercase and lowercase letters, and special symbols, length 8-10 digits, with interference lines, noise and character skew According to the algorithm rule, there is no need for complex verification in low-risk scenarios, so the verification code complexity level is determined to be 1, and a pure digital verification code (length 4-6 digits, no interference line and noise) is generated. This simple verification code can not only meet the basic security needs, but also reduce the user's operation burden and improve the user experience. Then, the module calls the verification code validity algorithm to adjust the effective duration of the verification code. The calculation formula of the verification code validity algorithm is: wherein: is the final effective duration, is the basic duration corresponding to the complexity level, Risk coefficient, set according to the risk scenario severity, value range , Risk weight, set according to the risk scenario severity, value range , based on the basic duration of level 1, combined with the current risk-free scenario, determine the valid duration of the verification code (such as 60 seconds), reserve sufficient input time for the user.

[0025] Verification execution module: after obtaining the verification code and valid duration information from the verification code generation and adjustment module, the following is displayed on the APP login interface: pure digital verification code input box, "please input the verification code within 60 seconds" text prompt, dynamic countdown progress bar, the countdown function can intuitively remind the user of the time limit, avoid operation failure due to timeout, the user inputs the correct verification code within 30 seconds, the system immediately performs verification (first confirm within the valid duration, then compare the input content with the generated verification code), after verification, send "allow login" instruction to APP background, the user successfully enters the APP homepage, the whole process is efficient and smooth, while ensuring security, reduce user waiting time, as shown in Figure 3 .

[0026] Feedback learning module: real-time capture of user successful login operation record and verification result (input correct, not timeout), mark these data as "normal behavior sample", feedback to user behavior pattern prediction model, the model learns from the sample, further strengthens the memory of "Wednesday afternoon, home location, common device login" behavior pattern, improves the recognition speed and accuracy of similar normal behaviors in the future; at the same time, feedback "low complexity verification code is effective in this scenario and user experience is good" to the algorithm of the verification code generation and adjustment module, optimize the verification code generation strategy of the algorithm for similar scenarios, realize self-iteration and upgrading of the system.

[0027] In summary, in the user's daily login to the commonly used shopping APP scenario, the intelligent SMS identity verification system modules work together, the data collection module encrypts the user login related data, laying the foundation for subsequent analysis; the behavior pattern prediction module accurately identifies normal behaviors through related models and algorithms; the verification code generation and adjustment module generates low complexity verification codes according to the results, taking into account security and experience; the verification execution module efficiently completes the verification process; the feedback learning module optimizes the model algorithm, as shown in Figure 1 , the whole process not only ensures login security, but also improves user convenience.

[0028] Example two: User conducts large transaction in a different place.

[0029] Data collection module: when the user is traveling in a foreign place, opens a payment APP and initiates a large amount of transfer that is much larger than the historical transaction amount, the data collection module immediately starts and interacts with the APP background, and collects multi-dimensional data in real time: including transaction time (not the user's historical large transaction common time period), transaction amount (much higher than the historical average transaction amount and the preset threshold), geographic location obtained by GPS (foreign city, significantly different from the commonly used local location, and the location change trajectory is recorded), operation device (the user's commonly used mobile phone, the device model and system version information are consistent with the historical record), operation frequency (the interval between clicking the transfer button and inputting the amount has slight differences with the historical operation habits), these data comprehensively cover the key characteristics of the transaction behavior, and can provide sufficient basis for subsequent risk judgment; and all collected data is encrypted during transmission, and stored in a secure database in an encrypted form, effectively preventing user transaction information and location information from being leaked or tampered with, as shown in Figure 2 .

[0030] Behavior pattern prediction module: after receiving the encrypted data transmitted by the data collection module, the data is first decrypted and preprocessed (such as filtering duplicate data and calibrating location information format), and then the data is analyzed in depth, the module calls the user behavior pattern prediction model, compares and calculates the amount, time, geographic location, and operation frequency data of the current transaction with the historical normal transaction behavior data, and obtains the user behavior credibility index, the calculation formula of the user behavior pattern prediction model is: , since there are obvious differences between the multiple key data and the historical normal behavior, the model determines that UBI<0.3, then the module calls the two-dimensional risk identification algorithm to further verify from the operation location deviation (the current location is significantly deviated from the commonly used region) and the operation frequency deviation (the interval of this operation is significantly different from the historical average frequency), the calculation formula of the two-dimensional risk identification algorithm is: , combined with the result of UBI<0.3, the final potential security risk scene identification result (SRS) is determined as 1, and it is clear that this scene belongs to “large transaction request with a difference of more than a preset threshold from the historical transaction amount” and “operation initiated across multiple regions in a short time in a single session”, this process through model calculation and algorithm identification can accurately capture abnormal behavior characteristics and clearly determine the risk scene type, providing targeted basis for subsequent verification code adjustment.

[0031] Verification code generation and adjustment module: after receiving the result (UBI<0.3 and SRS=1) of the behavior pattern prediction module, the verification code complexity algorithm is called to adjust the strength of the verification code, the calculation formula of the verification code complexity algorithm is: , according to the algorithm rule, the verification strength needs to be improved in the high-risk scenario, so the verification code complexity level is determined as 3, and the verification code composed of numbers, uppercase and lowercase letters and special symbols is generated (length 8-10 bits, containing interference lines, noise and character tilt design), this high complexity verification code can effectively improve the difficulty of malicious cracking and strengthen the security protection capability, then the verification code time algorithm is called to adjust the effective time of the verification code, and the calculation formula of the verification code time algorithm is: , based on the basic time length of level 3, combined with the high-risk scenario of large amount of transaction and out-of-town operation, the corresponding risk coefficient and risk weight are set, and the effective time length of the verification code (such as 30 seconds) is finally determined, the shorter time length can reduce the risk of the verification code being intercepted for illegal operation, and the high complexity verification code forms a double security protection.

[0032] Verification execution module: after obtaining the verification code and effective time length information from the verification code generation and adjustment module, the mixed type verification code input box, the prompt text "please input the verification code within 30 seconds, containing numbers, uppercase and lowercase letters and special symbols", real-time countdown (displaying the remaining time in seconds), when the remaining time is less than 20% (i.e. within 6 seconds), the interface pops up a pop-up window reminding "the remaining time is insufficient, please input as soon as possible", to avoid the user from not paying attention to the time and exceeding the time, after the user inputs the verification code, the system immediately performs verification: first checks whether it is submitted within 30 seconds, and then compares whether the input content is completely consistent with the generated verification code, after the verification is passed, the "allow transaction" instruction is sent to the APP background, and the transaction process continues; when the verification fails (such as input error or timeout), it prompts "verification code error or expired, please get it again" and refuses the current transaction request, the whole process through the clear prompt and strict verification, while ensuring the safety of transaction, as far as possible to reduce the repeated operation caused by the user's operation mistake.

[0033] Feedback learning module: collect the complete operation record of the user's this time out-of-town large amount of transaction (including transaction initiation, verification code input, verification result) and the verification result of the verification execution module (such as verification passed, user successfully completed the transaction), mark these data as "normal operation sample in high-risk scenario", feedback to the user behavior pattern prediction model, the model learns the sample, updates the recognition logic of the user's "transaction behavior in out-of-town scenario" and "operation habit in specific travel period" characteristics, improves the discrimination ability of similar reasonable high-risk behavior in the future, at the same time, the algorithm of the verification code generation and adjustment module is fed back "level 3 verification code and corresponding time length can guarantee safety in this high-risk scenario without affecting the user operation too much", optimize the algorithm of the verification code complexity and time setting rules for the risk scenario such as "out-of-town large amount of transaction", so that the system realizes more accurate balance between safety and user experience.

[0034] In summary, in the scenario of large transaction conducted by users in different places, the system shows strong risk response capability. The data collection module comprehensively collects transaction and environment data and stores them in encrypted form. The behavior pattern prediction module accurately identifies high risk through models and algorithms. The verification code generation and adjustment module generates high complexity verification codes and shortens the time limit, thereby strengthening security protection. The verification execution module strictly checks and timely reminds. The feedback learning module optimizes the model and improves the identification of reasonable high-risk behaviors. The system balances user experience while ensuring transaction security.

[0035] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with a preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, without departing from the technical solution of the present application. Any modification, change, equivalent change and modification of the above embodiments made in accordance with the technical essence of the present application are still within the scope of the technical solution of the present application.

Claims

1. An intelligent SMS identity verification system, characterized in that, The system includes: Data acquisition module: Interacts with various user applications and systems to collect users' historical behavior data in real time, covering login time, operation frequency, usage habits, transaction records, and geographical location information. All collected data is encrypted and stored in a secure database. Behavior pattern prediction module: preprocesses the collected historical behavior data, builds a user behavior pattern prediction model to predict the user's possible future behavior patterns, and uses a two-dimensional risk identification algorithm to identify potential security risk scenarios. CAPTCHA generation and adjustment module: Based on the results from the behavior pattern prediction module, the strength of the CAPTCHA is adjusted using a CAPTCHA complexity algorithm; the validity period of the CAPTCHA is adjusted using a CAPTCHA expiration algorithm. Verification Execution Module: When a user initiates an operation request, a verification code is sent to the user's registered mobile phone number via SMS. The operation interface displays a verification code input box, relevant prompts, and a countdown timer for the validity period, prompting the user to enter the code within the specified time after receiving the SMS. After the user enters the code, the system compares whether the SMS verification code matches the system-generated verification code and decides whether to allow or deny the user's operation request based on the verification result. Feedback Learning Module: Based on the user's actual operations and the verification results of the verification execution module, optimize and update the prediction model of the behavior pattern prediction module and the algorithm of the verification code generation and adjustment module.

2. The intelligent SMS identity verification system according to claim 1, characterized in that, The data acquisition module collects the following data: login data includes time, location, device, and browser type; operation data includes button clicks, page browsing, and operation frequency intervals; transaction data involves amount, object, time, and payment method; and geolocation data is obtained and recorded through device GPS or network positioning to track location changes.

3. The intelligent SMS identity verification system according to claim 1, characterized in that, In the behavior pattern prediction module, the calculation formula for the user behavior pattern prediction model is as follows: ,in, It is a user behavior credibility index, used to measure the degree of match between current behavior and historical normal behavior. The number of times a user's current behavior is matched with their historical stable behavior. This represents the total sample size of users' historical behavior. This represents the stability coefficient based on recent operating frequency. The preset baseline stability threshold for the system, The physical distance between the current operating location and frequently used areas. The maximum geographical deviation threshold set for the system. These are weighting coefficients, manually set or determined through data training based on scenario security requirements. .

4. The intelligent SMS identity verification system according to claim 3, characterized in that, In the behavior pattern prediction module, the calculation formula for the two-dimensional risk identification algorithm is as follows: , The results of identifying potential security risk scenarios This indicates that potential security risk scenarios have been identified. This indicates that no potential security risk scenarios were identified. This is a credibility index for user behavior. The physical distance between the current operating location and frequently used areas. This is the average physical distance between the user's historical operation locations and frequently used regions. The standard deviation of the physical distance between the user's historical operation locations and frequently used regions. This is the current operating frequency. This represents the average frequency of the user's historical operations. Let be the standard deviation of the user's historical operation frequency, when And the deviation between the current operating location and the commonly used geographical location is greater than 2, or Furthermore, if the deviation of the current operation frequency is greater than 2, it is determined that a potential security risk scenario has been identified.

5. The intelligent SMS identity verification system according to claim 1, characterized in that, The behavioral pattern prediction module identifies potential security risk scenarios including: simultaneous login of the same account on devices in different network environments; continuous login attempts during non-user-usual time periods; initiating operations across multiple regions in a short period of time in a single session; large transaction requests with a difference from historical transaction amounts exceeding a preset threshold; changing the bound mobile phone number or bank card information multiple times in a short period of time; modifying account information immediately after login, and the modified content involves core identity information; and sensitive operation requests in non-user-usual language environments.

6. The intelligent SMS identity verification system according to claim 1, characterized in that, In the verification code generation and adjustment module, the formula for calculating the verification code complexity algorithm is as follows: ,in For the complexity level of the CAPTCHA, This is a credibility index for user behavior. The results are for identifying potential security risk scenarios. 1 = risk exists, 0 = no risk. Level 1: Pure numeric verification code, 4-6 digits long, no interference lines / noise; Level 2: Mixed numbers and lowercase letters, 6-8 digits long, with slight interference lines; Level 3: Mixed numbers, uppercase and lowercase letters and special symbols, 8-10 digits long, with interference lines, noise and character tilt.

7. The intelligent SMS identity verification system according to claim 6, characterized in that, In the verification code generation and adjustment module, the calculation formula for the verification code validity period algorithm is as follows: ,in: For the final effective duration, The base time corresponding to the complexity level, This is a risk coefficient, set according to the severity of the risk scenario, with a range of values. , Risk weights are set according to the severity of the risk scenario, and their value range is [not specified]. .

8. The intelligent SMS identity verification system according to claim 1, characterized in that, The specific steps for performing the verification operation in the verification execution module are as follows: after receiving the user's operation request, send an instruction to the verification code generation and adjustment module to obtain the verification code and validity period information; send the verification code to the user's registered mobile phone number through the SMS channel; display the verification code input box, related prompts and validity period countdown on the user operation interface; monitor the SMS verification code entered by the user and capture the input content after the user submits it; Verify the verification code: Check whether the mobile phone number from which the SMS verification code originates matches the user's registered mobile phone number; Check if the submission is made within the valid time limit; if it expires, the verification is deemed to have failed. Compare whether the SMS verification code entered by the user matches the verification code generated by the system; Based on the verification result, send an operation permission or denial instruction to the business system; when verification fails, record the reason for failure and prompt the user with an SMS verification code error or verification timeout message.

9. The intelligent SMS identity verification system according to claim 8, characterized in that, In the verification execution module, when the verification code input box is displayed, the countdown timer for the validity period of the SMS verification code is displayed simultaneously; when the remaining time is less than 20% of the total valid time, an SMS verification code expiration reminder is sent to the user.

10. The intelligent SMS authentication system according to claim 1, characterized in that, The verification code generation and adjustment module includes a verification code refresh mechanism, specifically: User-initiated refresh: When a user clicks the refresh verification code button on the operation interface, a new verification code is generated; the new verification code is sent to the user's registered mobile phone number via SMS to reset the validity period countdown; Error-triggered refresh: When the system detects that a user has entered the wrong verification code twice in a row, it automatically generates a new verification code; sends the new verification code to the user's registered mobile phone number via SMS; and increases the complexity level of the verification code by one level, up to a maximum of level 3. Reset the countdown timer for the valid duration; Single mobile number sending frequency limit: The same registered mobile number can receive a maximum of 2 verification code SMS messages within 60 seconds.

Images