Verification SMS Sending System Based on Big Data Analysis

Abstract

This invention relates to the field of verification SMS sending technology, and discloses a verification SMS sending system based on big data analysis. The system includes: a user behavior analysis module, which collects historical user operations and current business scenario characteristics to generate scenario-based tags; a dynamic SMS template generation module, which generates target verification SMS content based on scenario-based tags; a traffic prediction module, which predicts the peak SMS sending volume for a target time window based on historical communication load data and real-time concurrent request volume; a dynamic scheduling module, which generates a channel allocation strategy based on the peak SMS sending volume and prioritizes the target verification SMS content to generate a sending priority; and an SMS distribution execution module, which calls the communication interface according to the sending priority and channel allocation strategy to send the target verification SMS content. This invention can improve the efficiency of verification SMS sending, reduce latency and loss, and enhance user experience and business process smoothness.

Description

Technical Field

[0001] This invention relates to the field of verification SMS sending technology, and in particular to a verification SMS sending system based on big data analysis. Background Technology

[0002] In today's digital age, the demand for communication and identity verification is growing, leading to the emergence of verification SMS sending systems. While existing systems can basically deliver SMS messages, several problems remain to be solved. Firstly, SMS delivery efficiency is difficult to guarantee; during peak hours or when dealing with a large number of users, SMS delays and loss are frequent, affecting user experience and the smooth operation of business processes. Secondly, the accuracy of SMS content is lacking; it cannot accurately adapt verification codes and other information based on different user scenarios and behaviors, resulting in some users receiving SMS messages that do not meet their needs or are expired, reducing the effective utilization rate of SMS messages. These problems restrict the further development and application expansion of verification SMS sending systems.

[0003] Therefore, there is an urgent need to provide a technical solution to address the above problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a verification SMS sending system based on big data analysis.

[0005] In a first aspect, the present invention provides a verification SMS sending system based on big data analysis, the technical solution of which is as follows:

[0006] The user behavior analysis module is used to collect users' historical operation data and current business scenario characteristics in real time, and generate scenario-based tags.

[0007] The SMS template dynamic generation module is used to generate target verification SMS content carrying timeliness parameters by matching the scenario-based tags with a preset SMS content rule library.

[0008] The traffic prediction module is used to predict the peak SMS sending volume for a target time window based on historical communication load data and real-time concurrent request volume, using a time series model.

[0009] The dynamic scheduling module is used to generate the sending priority of the target verification SMS content based on the SMS sending peak generation channel allocation strategy and the timeliness parameter.

[0010] The SMS distribution execution module is used to call the corresponding communication interface according to the sending priority and the channel allocation strategy to send the target verification SMS content.

[0011] The beneficial effects of the verification SMS sending system based on big data analysis of the present invention are as follows:

[0012] The system of this invention achieves accurate SMS content adaptation by utilizing user behavior analysis. Combined with traffic prediction and dynamic scheduling, it rationally allocates resources and prioritizes sending messages, thereby improving the efficiency of sending verification SMS messages, reducing delays and loss during peak periods and with a large number of users, making the SMS content more accurately match user scenarios and timeliness requirements, improving user experience and business process smoothness, and enhancing its development and application potential in the digital age.

[0013] Based on the above solution, the verification SMS sending system based on big data analysis of the present invention can be further improved as follows.

[0014] In one alternative approach, the user behavior analysis module is specifically used for:

[0015] Collect the user's historical operation data, including operation frequency, operation time distribution, and verification request type.

[0016] Extract the page access path, device identifier, and real-time geographical location from the features of the current business scenario;

[0017] The operation frequency, operation time distribution, verification request type, page access path, device identifier, and real-time geographical location are input into a preset scene classification model, and the scene-based label containing scene classification identifier and timeliness level label is output.

[0018] In one alternative approach, the SMS template dynamic generation module is specifically used for:

[0019] Based on the scenario classification identifier in the scenario-based tags, query the preset SMS content rule library to match the corresponding basic SMS template;

[0020] The timeliness parameter is determined based on the timeliness level tag in the contextualized tag.

[0021] The basic SMS template is combined with the dynamic verification code variable and injected into the timeliness parameter to generate the target verification SMS content.

[0022] In one alternative approach, the traffic prediction module is specifically used for:

[0023] The historical communication load data is divided into historical communication load datasets according to preset time segments;

[0024] A real-time load sequence is generated by sampling the real-time concurrent request volume using a sliding window.

[0025] The historical communication load dataset and the real-time load sequence are input into the time series model for feature fusion, and the predicted load value of the target time window is calculated through the dynamic weight function in the time series model.

[0026] The peak SMS sending value is generated based on the predicted load value and the preset channel capacity threshold.

[0027] In one alternative approach, the traffic prediction module is specifically used for:

[0028] Extract the historical time period feature vector corresponding to the target time window from the historical communication load dataset, and perform standard deviation normalization on the real-time load sequence to generate a real-time fluctuation vector;

[0029] The historical period feature vector and the real-time fluctuation vector are input into the dynamic weighting function to perform linear weighted fusion, thereby obtaining the predicted load value.

[0030] In one alternative approach, the expression for calculating the predicted load value is: L_pred=[α(t)×(1+β)]×H+[1-α(t)]×R;

[0031] Wherein, L_pred represents the predicted load value, H represents the historical period feature vector, R represents the real-time fluctuation vector, α(t) represents the weight base value determined according to the period type of the target time window, and β represents the dynamic correction factor calculated based on the standard deviation of the real-time fluctuation vector.

[0032] In one alternative approach, the dynamic scheduling module is specifically used for:

[0033] The channel load rate is calculated based on the peak SMS sending rate and the preset channel capacity limit. When the channel load rate exceeds the first threshold, the backup communication channel is activated and the traffic allocation ratio of the backup communication channel is allocated to generate the channel allocation strategy.

[0034] Extract the timeliness parameter from the target verification SMS content and convert it into the remaining valid time value;

[0035] When the remaining valid time value is less than the second threshold, the target verification SMS content is marked as high time-sensitive content; when the remaining valid time value is greater than or equal to the second threshold, the target verification SMS content is marked as regular time-sensitive content.

[0036] When the target verification SMS content is marked as high timeliness content, the timeliness urgency score of the target verification SMS content is calculated, and the priority ranking is determined according to the timeliness urgency score; when the target verification SMS content is marked as regular timeliness content, the target verification SMS content is arranged in order according to the receiving time to determine the priority ranking.

[0037] The high-time-sensitivity content is sorted and placed before the regular-time-sensitivity content to generate the sending priority.

[0038] In one alternative approach, the SMS distribution execution module is specifically used for:

[0039] The channel allocation strategy is analyzed to obtain the traffic allocation ratio between the main communication channel and the backup communication channel;

[0040] When the real-time load of the main communication channel is lower than the capacity threshold, the main communication interface is called according to the sending priority to send the target verification SMS content;

[0041] When the real-time load of the main communication channel reaches the capacity threshold, the target verification SMS content is diverted to the backup communication channel according to the traffic allocation ratio. The target verification SMS content diverted to the backup communication channel is sent by calling the backup communication interface according to the sending priority.

[0042] In one alternative embodiment, the system further includes a real-time monitoring module; the real-time monitoring module is used for:

[0043] Real-time monitoring of the sending status of the target verification SMS content;

[0044] When the target verification SMS content fails to be sent and the target verification SMS content is high time-sensitive content, the sending priority is recalculated based on the timeliness urgency score and a resend mechanism is triggered.

[0045] When the target verification SMS content fails to be sent and the target verification SMS content is a regular time-sensitive content, the sending priority is regenerated according to the receiving time order and the resending mechanism is triggered.

[0046] Secondly, this invention provides a verification SMS sending method based on big data analysis, the technical solution of which is as follows:

[0047] Real-time collection of users' historical operation data and current business scenario characteristics to generate contextualized tags;

[0048] Based on the scenario-based tags, a preset SMS content rule library is matched to generate target verification SMS content carrying timeliness parameters;

[0049] Based on historical communication load data and real-time concurrent request volume, the peak SMS sending volume for the target time window is predicted using a time series model.

[0050] Based on the SMS sending peak generation channel allocation strategy, and based on the timeliness parameter, the sending priority of the target verification SMS content is sorted to generate the sending priority of the target verification SMS content.

[0051] The corresponding communication interface is invoked according to the sending priority and the channel allocation strategy to send the target verification SMS content.

[0052] The beneficial effects of the verification SMS sending method based on big data analysis of the present invention are as follows:

[0053] The method of this invention achieves accurate SMS content adaptation by utilizing user behavior analysis, and combines traffic prediction and dynamic scheduling to rationally allocate resources and prioritize sending messages. This can improve the efficiency of sending verification SMS messages, reduce delays and loss during peak periods and with a large number of users, and make SMS content more accurately match user scenarios and timeliness requirements. This enhances user experience and business process smoothness, and strengthens its development and application potential in the digital age.

[0054] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0056] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0057] Figure 1 This is a schematic diagram of an embodiment of a verification SMS sending system based on big data analysis according to the present invention.

[0058] Figure 2 This is a flowchart illustrating an embodiment of a big data analysis-based method for verifying SMS sending according to the present invention. Detailed Implementation

[0059] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0060] Figure 1 A schematic diagram of an embodiment of a verification SMS sending system based on big data analysis provided by the present invention is shown. Figure 1 As shown, the system includes:

[0061] The user behavior analysis module 110 is used to collect users' historical operation data and current business scenario characteristics in real time, and generate scenario-based tags.

[0062] Historical operation data refers to the collection of behavioral information recorded when a user performed verification operations in the past. Current business scenario characteristics refer to the real-time environmental parameters associated with when a user triggers a verification request. Contextualized tags refer to composite tags output by the scenario classification model, which include scenario classification identifiers and timeliness level tags.

[0063] The SMS template dynamic generation module 120 is used to generate target verification SMS content carrying timeliness parameters by matching the scenario-based tags with a preset SMS content rule library.

[0064] The pre-defined SMS content rule base refers to the set of SMS templates categorized by scenario and pre-stored in the system. The timeliness parameter refers to the valid duration of the verification SMS determined by the timeliness level label. The target verification SMS content refers to the final SMS entity to be sent, incorporating the timeliness parameter and dynamic verification code variables.

[0065] The traffic prediction module 130 is used to predict the peak SMS sending volume of a target time window based on historical communication load data and real-time concurrent request volume, and through a time series model.

[0066] Historical communication load data refers to a set of past SMS sending records divided into preset time segments. Real-time concurrent request volume refers to the number of verification SMS sending requests received per unit time. The time series model is a statistical analysis model that predicts load values ​​based on historical and real-time data. The target time window refers to the future time period for predicting the peak SMS sending volume. The peak SMS sending volume refers to the maximum amount of SMS that can be sent per unit time calculated based on the predicted load value and the channel capacity threshold.

[0067] The dynamic scheduling module 140 is used to generate the sending priority of the target verification SMS content by sorting the sending priority according to the SMS sending peak generation channel allocation strategy and the timeliness parameter.

[0068] Here, channel allocation strategy refers to the traffic allocation ratio scheme for primary and backup communication channels. Sending priority sorting refers to the operation of sequentially arranging the content of target verification SMS messages according to their urgency score or reception time. Sending priority refers to the sending order attribute of the target verification SMS content.

[0069] The SMS distribution execution module 150 is used to call the corresponding communication interface according to the sending priority and the channel allocation strategy to send the target verification SMS content.

[0070] The communication interface refers to the physical or logical transmission channel that connects to the SMS gateway.

[0071] The technical solution of this embodiment achieves accurate SMS content adaptation by utilizing user behavior analysis. Combined with traffic prediction and dynamic scheduling, it rationally allocates resources and prioritizes sending messages, which can improve the efficiency of sending verification SMS messages, reduce delays and loss during peak periods and with a large number of users, and make SMS content more accurately meet user scenarios and timeliness requirements. This enhances user experience and business process smoothness, and strengthens its development and application potential in the digital age.

[0072] In an alternative embodiment, the user behavior analysis module 110 is specifically used for:

[0073] Collect the user's historical operation data, including operation frequency, operation time distribution, and verification request type.

[0074] Here, operation frequency refers to the statistical value of the number of times a user triggers a verification operation per unit of time. Operation time distribution refers to the time period distribution characteristics of a user's historical verification operations. Verification request type refers to the verification scenario classification identifier corresponding to the operation.

[0075] Extract the page access path, device identifier, and real-time geographical location from the features of the current business scenario.

[0076] Here, the page access path refers to the sequence of pages browsed by the user before triggering the verification request. The device identifier is the unique identification code of the terminal device that initiated the verification request. The real-time geolocation refers to the latitude and longitude coordinates of the user when triggering the verification request.

[0077] The operation frequency, operation time distribution, verification request type, page access path, device identifier, and real-time geographical location are input into a preset scene classification model, and the scene-based label containing scene classification identifier and timeliness level label is output.

[0078] Among them, the preset scenario classification model refers to the machine learning model that outputs scenario-based labels. The scenario classification label refers to the classification label that identifies the verification scenario type. The timeliness level label refers to the level mark that identifies the timeliness requirement of the SMS message.

[0079] Among the above-mentioned optional methods, it is even possible to comprehensively collect user operation data and business scenario characteristics, input them into the scenario classification model to generate scenario-based tags with timeliness levels, and provide accurate basis for subsequent SMS personalization.

[0080] In one alternative embodiment, the SMS template dynamic generation module 120 is specifically used for:

[0081] Based on the scenario classification identifier in the scenario-based tags, the corresponding basic SMS template is matched by querying the preset SMS content rule library.

[0082] Among them, the timeliness level label refers to the level mark that identifies the timeliness requirement of the SMS message. The basic SMS template refers to the original SMS text in the SMS content rule base without injected variables.

[0083] The timeliness parameter is determined based on the timeliness level tag in the contextualized tag.

[0084] The basic SMS template is combined with the dynamic verification code variable and injected into the timeliness parameter to generate the target verification SMS content.

[0085] Among them, dynamic verification code variables refer to the sequence of verification code numbers that need to be generated in real time.

[0086] Specifically: ① Locate the predefined verification code placeholder and time-limited parameter placeholder in the basic SMS template; ② Generate a dynamic verification code variable corresponding to the current verification request; ③ Insert the dynamic verification code variable into the verification code placeholder position; ④ Convert the time-limited parameter into time-limited text with units, where the unit conversion rule is: when the parameter value is ≥60 seconds, convert to minutes; when the parameter value is <60 seconds, keep the seconds; ⑤ Insert the time-limited text into the time-limited parameter placeholder position; ⑥ Perform syntax compliance checks on the template after variable injection, including checking the consistency of number format and text length limits; ⑦ Output the result that passes the check as the target verification SMS content.

[0087] Among the above optional methods, the basic template is further matched based on the scenario-based tags, and the target SMS is generated in combination with the timeliness parameters, so that the SMS content is more in line with the user scenario and timeliness requirements.

[0088] In an alternative embodiment, the traffic prediction module 130 is specifically used for:

[0089] The historical communication load data is divided into historical communication load datasets according to preset time segments.

[0090] The preset time segment refers to a fixed time unit for dividing historical data. The historical communication load dataset refers to a collection of SMS sending volume data organized according to the preset time segment.

[0091] A real-time load sequence is generated by sampling the real-time concurrent request volume using a sliding window.

[0092] The real-time load sequence refers to the real-time request volume sequence generated through sliding window sampling.

[0093] Specifically: ① Set a sliding window of fixed duration, the duration of which is one-twentieth of the preset time segment; ② Extract the time interval covered by the sliding window based on the current time; ③ Calculate the sum of real-time concurrent requests within the time interval as the current window sample value; ④ Slide the sliding window forward by one time step, the time step being one-tenth of the sliding window duration; ⑤ Repeat the operations of extracting the time interval and calculating the sum of requests until the target prediction period is covered; ⑥ Arrange all window sample values ​​in chronological order to generate a continuous, equally spaced real-time load sequence.

[0094] The historical communication load dataset and the real-time load sequence are input into the time series model for feature fusion, and the predicted load value of the target time window is calculated through the dynamic weight function in the time series model.

[0095] The dynamic weighting function refers to the function that performs a linear weighted fusion of historical feature vectors and real-time fluctuation vectors. The predicted load value refers to the predicted value for the target time window calculated using a time series model.

[0096] The peak SMS sending value is generated based on the predicted load value and the preset channel capacity threshold.

[0097] The preset channel capacity threshold refers to the maximum capacity setting for a single communication channel.

[0098] Specifically: ① Multiply the predicted load value by a preset redundancy coefficient δ to obtain a corrected predicted value, where the value of δ ranges from 1.05 to 1.20; ② Calculate the upper limit of the available capacity of a single communication channel. The upper limit of the available capacity is equal to the preset channel capacity threshold multiplied by the channel health coefficient η, where the value of η is linearly mapped from the channel's real-time packet loss rate to the range of 0.85 to 1.00; ③ Calculate the total available capacity of the system based on the number of currently active communication channels. The total available capacity of the system is the sum of the upper limits of the available capacity of all channels; ④ Compare the corrected predicted value with the total available capacity of the system; ⑤ When the corrected predicted value is less than or equal to the total available capacity of the system, the corrected predicted value is used as the peak value for SMS sending; when the corrected predicted value is greater than the total available capacity of the system, the total available capacity of the system is used as the peak value for SMS sending.

[0099] In the above-mentioned optional methods, the historical load data is further divided and the real-time concurrent request volume is sampled. The predicted load value is obtained by processing it through a time series model, thereby accurately predicting the peak value of SMS sending.

[0100] In an alternative embodiment, the traffic prediction module 130 is specifically used for:

[0101] Extract the historical time period feature vector corresponding to the target time window from the historical communication load dataset, and perform standard deviation normalization on the real-time load sequence to generate a real-time fluctuation vector.

[0102] The historical time period feature vector refers to the historical data feature vector associated with the target time window. The real-time fluctuation vector refers to the real-time load sequence vector after standard deviation normalization.

[0103] The historical period feature vector and the real-time fluctuation vector are input into the dynamic weighting function to perform linear weighted fusion, thereby obtaining the predicted load value.

[0104] In the above-mentioned optional methods, the historical period feature vector and real-time fluctuation vector are further extracted and fused by a dynamic weighting function to obtain the predicted load value, thereby improving the prediction accuracy.

[0105] In one alternative approach, the expression for calculating the predicted load value is: L_pred=[α(t)×(1+β)]×H+[1-α(t)]×R;

[0106] Wherein, L_pred represents the predicted load value (unit: records / second), H represents the historical period feature vector (unit: records / second), R represents the real-time fluctuation vector (normalized standard value), α(t) represents the weight base value determined according to the period type of the target time window (dimensionless), and β represents the dynamic correction factor calculated based on the standard deviation of the real-time fluctuation vector (dimensionless).

[0107] Here, "time period type" refers to the time attribute identifier of the target time window, such as weekdays or holidays. "Weight base value" refers to the initial weight value of the dynamic weight function determined based on the time period type. "Dynamic correction factor" refers to the weight adjustment coefficient calculated based on the standard deviation of the real-time fluctuation vector.

[0108] It should be noted that the calculation expression for the predicted load value dynamically adapts to the differences in traffic patterns between weekdays and holidays by using time period weight base values. It utilizes dynamic correction factors to respond to real-time fluctuation intensity to solve the problem of distortion in sudden traffic predictions. The historical time period feature vector reflects long-term traffic trends, while the real-time fluctuation vector captures short-term change characteristics. This expression ensures the reliability of the predicted value through dimensional unification and linear compensation mechanisms, significantly improves the prediction accuracy during peak periods, reduces the risk of channel allocation overload, optimizes system resource utilization, and maintains computational efficiency to meet real-time processing requirements.

[0109] Among the above optional methods, a formula for calculating the predicted load value is further provided, which integrates historical and real-time data to make traffic prediction more accurate and provide a reliable basis for the allocation of SMS sending resources.

[0110] In an alternative embodiment, the dynamic scheduling module 140 is specifically used for:

[0111] The channel load rate is calculated based on the peak SMS sending rate and the preset channel capacity limit. When the channel load rate exceeds the first threshold, the backup communication channel is activated and the traffic allocation ratio of the backup communication channel is allocated to generate the channel allocation strategy.

[0112] Here, channel load rate refers to the ratio of peak SMS sending volume to the channel capacity limit. The first threshold is the critical channel load rate value that triggers the activation of the backup communication channel. The backup communication channel refers to an emergency SMS transmission channel configured in addition to the main channel. The timeliness parameter is extracted from the target verification SMS content and converted into a remaining valid time value. The traffic allocation ratio refers to the proportion of service volume allocated to the backup communication channel.

[0113] Specifically: ① Calculate the channel load rate, which equals the peak SMS sending rate divided by the preset channel capacity limit; ② Compare the channel load rate with a first threshold of 70%; ③ When the channel load rate is less than or equal to 70%, generate a channel allocation strategy that uses only the primary communication channel; ④ When the channel load rate is greater than 70%, execute the backup communication channel activation command and establish a connection with the backup SMS gateway; ⑤ Query the real-time available bandwidth percentage of the backup communication channel; ⑥ Calculate the backup channel traffic allocation ratio based on the channel load rate, satisfying the formula: Backup channel percentage = min((channel load rate - 70%) / 30%, backup channel real-time available bandwidth percentage); ⑦ Generate a channel allocation strategy that includes the primary and backup channel traffic allocation ratios, where the primary communication channel percentage is 1 minus the backup channel percentage.

[0114] When the remaining valid time value is less than the second threshold, the target verification SMS content is marked as high time-sensitive content; when the remaining valid time value is greater than or equal to the second threshold, the target verification SMS content is marked as regular time-sensitive content.

[0115] The remaining valid time value refers to the number of valid seconds remaining after the timeliness parameter is converted to the current time. The second threshold is the time threshold that distinguishes between highly time-sensitive content and regular time-sensitive content. Highly time-sensitive content refers to target verification SMS content with a remaining valid time value less than the second threshold. Regularly time-sensitive content refers to target verification SMS content with a remaining valid time value greater than or equal to the second threshold.

[0116] Specifically: ① Extract the remaining valid time value from the metadata of the target verification SMS content;

[0117] ② Call the preset second threshold parameter, which is 60 seconds; ③ Perform a numerical comparison operation to determine whether the remaining valid time value is less than 60 seconds; ④ If the judgment result is yes, write the high timeliness flag value 1 into the attribute identifier field of the target verification SMS content; ⑤ If the judgment result is no, write the regular timeliness flag value 0 into the attribute identifier field of the target verification SMS content.

[0118] When the target verification SMS content is marked as high timeliness content, the timeliness urgency score of the target verification SMS content is calculated, and the priority ranking is determined according to the timeliness urgency score; when the target verification SMS content is marked as regular timeliness content, the target verification SMS content is arranged in order according to the receiving time to determine the priority ranking.

[0119] Among them, the timeliness score refers to the priority quantification value that is negatively correlated with the remaining effective time value.

[0120] Specifically: ① Detect the attribute identifier field value of the target verification SMS content; ② When a high timeliness flag value of 1 is detected, perform the following operations: read the remaining valid time value of the target verification SMS content; calculate the timeliness urgency score, which is equal to the weight coefficient K divided by the remaining valid time value, where K is a preset constant and K≥1000; insert the target verification SMS content into the high timeliness sorting queue in descending order of timeliness urgency score; When a regular timeliness flag value of 0 is detected, perform the following operations: extract the receiving timestamp from the metadata of the target verification SMS content; insert the target verification SMS content into the regular timeliness sorting queue in ascending order of receiving timestamp; ③ Merge the two queues to generate the final priority sorting, where all content in the high timeliness sorting queue is arranged before the regular timeliness sorting queue.

[0121] The high-time-sensitivity content is sorted and placed before the regular-time-sensitivity content to generate the sending priority.

[0122] Specifically: ① Read the high-timeliness sorting queue that has been internally sorted, keeping its timeliness urgency score in descending order; ② Read the regular timeliness sorting queue that has been internally sorted, keeping its receiving timestamp in ascending order; ③ Copy all elements of the high-timeliness sorting queue sequentially to the head of the final sending queue; ④ Copy all elements of the regular timeliness sorting queue sequentially to the tail of the final sending queue; ⑤ Assign each element in the final sending queue a continuously increasing sequence number as a sending priority value to determine the sending priority of the target verification SMS content.

[0123] Among the above optional methods, the load rate is further calculated based on the peak SMS sending volume and channel capacity, and backup channels are activated and traffic is allocated when necessary; SMS timeliness is divided and sorted according to timeliness parameters to optimize SMS sending priority.

[0124] In an alternative embodiment, the SMS distribution execution module 150 is specifically used for:

[0125] The channel allocation strategy is analyzed to obtain the traffic allocation ratio between the main communication channel and the backup communication channel;

[0126] When the real-time load of the main communication channel is lower than the capacity threshold, the main communication interface is called according to the sending priority to send the target verification SMS content;

[0127] When the real-time load of the main communication channel reaches the capacity threshold, the target verification SMS content is diverted to the backup communication channel according to the traffic allocation ratio. The target verification SMS content diverted to the backup communication channel is sent by calling the backup communication interface according to the sending priority.

[0128] Among the above optional methods, the channel strategy is further analyzed, and the interface is called to send SMS messages according to the load of the main channel. When overloaded, the traffic is proportionally diverted to the backup channel to ensure efficient SMS sending.

[0129] In one alternative embodiment, the system further includes a real-time monitoring module; the real-time monitoring module is used for:

[0130] Real-time monitoring of the sending status of the target verification SMS content;

[0131] When the target verification SMS content fails to be sent and the target verification SMS content is high time-sensitive content, the sending priority is recalculated based on the timeliness urgency score and a resend mechanism is triggered.

[0132] When the target verification SMS content fails to be sent and the target verification SMS content is a regular time-sensitive content, the sending priority is regenerated according to the receiving time order and the resending mechanism is triggered.

[0133] The retransmission mechanism refers to the fault handling process of regenerating the priority and re-executing the transmission when transmission fails.

[0134] Specifically: ① When a failure to send the target verification SMS content is detected, the attribute identifier field value of the content is read; ② When the attribute identifier field value is a high timeliness flag value of 1, the following operations are performed: extract the original timeliness urgency score of the content; calculate the attenuated timeliness urgency score, the new score is equal to the original score multiplied by the attenuation coefficient γ, where γ=0.8; insert the attenuated score as the new priority into the high timeliness retransmission queue; send an instruction containing the content identifier to the retransmission controller to trigger retransmission; When the attribute identifier field value is a regular timeliness flag value of 0, the following operations are performed: extract the original receiving timestamp of the content from the metadata; insert the original receiving timestamp as the new priority into the regular timeliness retransmission queue; send an instruction containing the content identifier to the retransmission controller to trigger retransmission; ③ After receiving the instruction, the retransmission controller performs the following operations: read a copy of the target verification SMS content from persistent storage; re-execute the distribution process according to the retransmission queue priority.

[0135] Among the above-mentioned optional methods, the SMS sending status can be further monitored in real time, and different resend strategies can be adopted for different time-sensitive SMS sending failures to improve the reliability of SMS sending.

[0136] Figure 2 This diagram illustrates a flowchart of an embodiment of a big data analysis-based method for verifying SMS sending provided by the present invention. Figure 2 As shown, it includes the following steps:

[0137] S1. Collect users' historical operation data and current business scenario characteristics in real time to generate scenario-based tags;

[0138] S2. Based on the scenario-based tags, match the preset SMS content rule library to generate target verification SMS content carrying timeliness parameters;

[0139] S3. Based on historical communication load data and real-time concurrent request volume, predict the peak SMS sending volume of the target time window through a time series model;

[0140] S4. Based on the SMS sending peak generation channel allocation strategy, and based on the timeliness parameter, sort the sending priority of the target verification SMS content to generate the sending priority of the target verification SMS content.

[0141] S5. According to the sending priority and the channel allocation strategy, call the corresponding communication interface to send the target verification SMS content.

[0142] The technical solution of this embodiment achieves accurate SMS content adaptation by utilizing user behavior analysis. Combined with traffic prediction and dynamic scheduling, it rationally allocates resources and prioritizes sending messages, which can improve the efficiency of sending verification SMS messages, reduce delays and loss during peak periods and with a large number of users, and make SMS content more accurately meet user scenarios and timeliness requirements. This enhances user experience and business process smoothness, and strengthens its development and application potential in the digital age.

[0143] Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0144] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0145] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0146] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A verification SMS sending system based on big data analysis, characterized in that, The system includes: The user behavior analysis module is used to collect users' historical operation data and current business scenario characteristics in real time, and generate scenario-based tags. The SMS template dynamic generation module is used to generate target verification SMS content carrying timeliness parameters by matching the scenario-based tags with a preset SMS content rule library. The traffic prediction module is used to predict the peak SMS sending volume for a target time window based on historical communication load data and real-time concurrent request volume, using a time series model. The dynamic scheduling module is used to generate the sending priority of the target verification SMS content based on the SMS sending peak generation channel allocation strategy and the timeliness parameter. The SMS distribution execution module is used to call the corresponding communication interface according to the sending priority and the channel allocation strategy to send the target verification SMS content; The dynamic scheduling module is specifically used for: The channel load rate is calculated based on the peak SMS sending rate and the preset channel capacity limit. When the channel load rate exceeds the first threshold, the backup communication channel is activated and the traffic allocation ratio of the backup communication channel is allocated to generate the channel allocation strategy. Extract the timeliness parameter from the target verification SMS content and convert it into the remaining valid time value; When the remaining valid time value is less than the second threshold, the target verification SMS content is marked as high time-sensitive content; when the remaining valid time value is greater than or equal to the second threshold, the target verification SMS content is marked as regular time-sensitive content. When the target verification SMS content is marked as high timeliness content, the timeliness urgency score of the target verification SMS content is calculated, and the priority ranking is determined according to the timeliness urgency score; when the target verification SMS content is marked as regular timeliness content, the target verification SMS content is arranged in order according to the receiving time to determine the priority ranking. The high-time-sensitivity content is sorted and placed before the regular-time-sensitivity content to generate the sending priority.

2. The verification SMS sending system based on big data analysis according to claim 1, characterized in that, The user behavior analysis module is specifically used for: Collect the user's historical operation data, including operation frequency, operation time distribution, and verification request type. Extract the page access path, device identifier, and real-time geographical location from the features of the current business scenario; The operation frequency, operation time distribution, verification request type, page access path, device identifier, and real-time geographical location are input into a preset scene classification model, and the scene-based label containing scene classification identifier and timeliness level label is output.

3. The verification SMS sending system based on big data analysis according to claim 2, characterized in that, The SMS template dynamic generation module is specifically used for: Based on the scenario classification identifier in the scenario-based tags, query the preset SMS content rule library to match the corresponding basic SMS template; The timeliness parameter is determined based on the timeliness level tag in the contextualized tag. The basic SMS template is combined with the dynamic verification code variable and injected into the timeliness parameter to generate the target verification SMS content.

4. The verification SMS sending system based on big data analysis according to claim 3, characterized in that, The traffic prediction module is specifically used for: The historical communication load data is divided into historical communication load datasets according to preset time segments; A real-time load sequence is generated by sampling the real-time concurrent request volume using a sliding window. The historical communication load dataset and the real-time load sequence are input into the time series model for feature fusion, and the predicted load value of the target time window is calculated through the dynamic weight function in the time series model. The peak SMS sending value is generated based on the predicted load value and the preset channel capacity threshold.

5. The verification SMS sending system based on big data analysis according to claim 4, characterized in that, The traffic prediction module is specifically used for: Extract the historical time period feature vector corresponding to the target time window from the historical communication load dataset, and perform standard deviation normalization on the real-time load sequence to generate a real-time fluctuation vector; The historical period feature vector and the real-time fluctuation vector are input into the dynamic weighting function to perform linear weighted fusion, thereby obtaining the predicted load value.

6. The verification SMS sending system based on big data analysis according to claim 5, characterized in that, The formula for calculating the predicted load value is: ; in, This represents the predicted load value. This represents the feature vector of the historical time period. This represents the real-time fluctuation vector. This represents the weight base value determined based on the time period type of the target time window. This represents the dynamic correction factor calculated based on the standard deviation of the real-time fluctuation vector.

7. The verification SMS sending system based on big data analysis according to claim 6, characterized in that, The SMS distribution execution module is specifically used for: The channel allocation strategy is analyzed to obtain the traffic allocation ratio between the main communication channel and the backup communication channel; When the real-time load of the main communication channel is lower than the capacity threshold, the main communication interface is called according to the sending priority to send the target verification SMS content; When the real-time load of the main communication channel reaches the capacity threshold, the target verification SMS content is diverted to the backup communication channel according to the traffic allocation ratio. The target verification SMS content diverted to the backup communication channel is sent by calling the backup communication interface according to the sending priority.

8. The verification SMS sending system based on big data analysis according to claim 7, characterized in that, Also includes: Real-time monitoring module; the real-time monitoring module is used for: Real-time monitoring of the sending status of the target verification SMS content; When the target verification SMS content fails to be sent and the target verification SMS content is high time-sensitive content, the sending priority is recalculated based on the timeliness urgency score and a resend mechanism is triggered. When the target verification SMS content fails to be sent and the target verification SMS content is a regular time-sensitive content, the sending priority is regenerated according to the receiving time order and the resending mechanism is triggered.

9. A method for sending verification SMS messages based on big data analysis, employing the verification SMS sending system based on big data analysis as described in any one of claims 1 to 8, characterized in that, The method includes: Real-time collection of users' historical operation data and current business scenario characteristics to generate contextualized tags; Based on the scenario-based tags, a preset SMS content rule library is matched to generate target verification SMS content carrying timeliness parameters; Based on historical communication load data and real-time concurrent request volume, the peak SMS sending volume for the target time window is predicted using a time series model. Based on the SMS sending peak generation channel allocation strategy, and based on the timeliness parameter, the sending priority of the target verification SMS content is sorted to generate the sending priority of the target verification SMS content. The corresponding communication interface is invoked according to the sending priority and the channel allocation strategy to send the target verification SMS content.

Images