Intelligent short message channel configuration system

Abstract

The invention relates to the technical field of short messages, and discloses a short message channel intelligent configuration system, which comprises a demand and attribute acquisition module used for acquiring a short message sending demand and a demand attribute in an expected time period; the load acquisition module is used for acquiring the limit load of each short message channel; the preliminary configuration module is used for acquiring a permission short message channel based on the service type and performing preliminary configuration according to the permission short message channel; the channel health degree analysis module is used for monitoring and acquiring performance data of each short message channel in real time and analyzing the channel health degree based on the performance data; and the intelligent channel selection module intelligently selects a channel based on an intelligent channel selection probability model, adjusts the short message channel of each service type based on the initial configuration, and sends a short message in the selected short message channel. The method can automatically detect the channel state, dynamically adjusts the short message sending channel based on the performance data of each short message channel, guarantees the timely, efficient and reliable delivery of the short message, and improves the customer experience.

Description

Technical Field

[0001] The present invention relates to the technical field of short message, and in particular to a short message channel intelligent configuration system. Background Art

[0002] The bank's post-loan management SMS notification function mainly relies on third-party information service providers. Currently, there are still deficiencies in the refined management of SMS sending channels. When a certain SMS channel encounters a failure, there is a lack of a mechanism to automatically switch to a backup channel. This not only significantly reduces the delivery rate of SMS, but also seriously affects the instant communication effect between the bank and its customers.

[0003] Although there are methods in the prior art that can switch SMS sending through multiple channels, the weight of each channel is configured according to the success rate of SMS sending through each channel, and the corresponding channel is allocated according to the weight to send SMS; or the weight is configured according to the sending cost of each channel, and the channel is allocated according to the weight to send SMS; however, these methods are highly limited in the business operation environment of banks. Summary of the invention

[0004] The present invention aims to provide an intelligent configuration system for SMS channels to automatically detect channel status and dynamically adjust SMS sending channels based on performance data of each SMS channel to ensure timely, efficient and reliable delivery of SMS and improve customer experience.

[0005] In order to achieve the above object, the present invention adopts the following technical scheme:

[0006] A short message channel intelligent configuration system, comprising:

[0007] A demand and attribute acquisition module, used to acquire the demand and demand attributes of SMS sending in the expected time period, wherein the demand attributes include the service type and quantity;

[0008] The load acquisition module is used to obtain the limit load of each SMS channel;

[0009] The preliminary configuration module obtains the authorized SMS channel based on the business type and performs preliminary configuration based on the authorized SMS channel;

[0010] The channel health analysis module is used to monitor and obtain the performance data of each SMS channel in real time, and analyze the channel health based on the performance data;

[0011] The intelligent channel selection module intelligently selects a channel based on an intelligent channel selection probability model, adjusts the SMS channel of each service type based on the preliminary configuration, and sends SMS messages on the selected SMS channel.

[0012] The principles and advantages of this solution are: in actual application, the demand and attribute acquisition module is used to obtain the demand for sending SMS in the expected time period and the demand attributes such as business type and quantity, and the load acquisition module is used to obtain the limit load of each SMS channel; the preliminary configuration module is used to obtain the authorized SMS channel based on the business type, and the preliminary configuration is performed according to the authorized SMS channel. The preliminary configuration is convenient for improving the configuration efficiency based on historical data performance, and the channel health analysis module is used to monitor and obtain the performance data of each SMS channel in real time, and the channel health is analyzed based on the performance data, so that each SMS is sent on the channel with the highest health; the intelligent channel selection module intelligently selects the channel based on the intelligent channel selection probability model, and adjusts the SMS channel of each business type based on the preliminary configuration, sends SMS on the selected SMS channel, and adjusts the SMS channel based on the preliminary configuration instead of allocating it randomly, which can improve the stability of the system, ensure the timely, efficient and reliable delivery of SMS, enhance customer experience, and bring a more stable and efficient business operation environment for the bank.

[0013] Preferably, as an improvement, the demand and attribute acquisition module further includes:

[0014] The feature association submodule is used to obtain the SMS template of the business type, obtain the feature information of the SMS template, and create an association table between the feature information and the business type;

[0015] The customized policy acquisition submodule is used to acquire the customized policy of the business type, wherein the customized policy includes the customized authority channel and the priority of the customized authority channel.

[0016] Technical effect: It facilitates the compatibility of product customization strategies and facilitates rapid fault analysis when faults are sent, so as to make timely adjustments.

[0017] Preferably, as an improvement, the configuration rules of the preliminary configuration are: for business types with customized policies, they are configured according to the customized policies first; for business types without customized policies, they are configured according to the historically configured channels, and when the SMS channel exceeds the load, it is randomly configured in the channel that meets the load.

[0018] Technical effect: It is convenient to prioritize the SMS messages of key business types and avoid overloading of a single channel.

[0019] Preferably, as an improvement, the performance data includes the success rate of sending requests, average response time, error type and frequency; the analysis model for analyzing channel health based on performance data is:

[0020]

[0021] Among them, a 1 、a 2 、a 3is the health ratio adjustment constant factor, C i is the success rate of sending requests, x i is the average response time of channel i, q iu is the error frequency of error type u in channel i, ρ u is the weight value of error type u.

[0022] Technical effect: It is convenient to analyze the health of the channel from multiple angles.

[0023] Preferably, as an improvement, the health ratio adjustment constant factor is dynamically adjusted according to the time period.

[0024] Technical effect: It is easy to improve the flexibility of channel traffic configuration according to business needs and channel characteristics.

[0025] Preferably, as an improvement, the intelligent channel selection module includes:

[0026] The channel triggering and rejection submodule is used to determine whether the channel health meets the threshold. If it is, the channel is triggered; if not, the channel is rejected.

[0027] The intelligent selection submodule is used to create an intelligent channel selection probability model based on channel health. The intelligent selection probability model is:

[0028]

[0029] t i ∈t i_m

[0030] L i ≤L i0

[0031] Among them, J i is the health of channel i, T i_m The probability of selecting channel i for service type m, t i is the selected SMS channel, t i_m Indicates the permission channel of business type m; L i is the current flow rate of channel i, L i0 is the limit load of channel i.

[0032] Technical effect: It is easy to ensure that text messages are sent on the healthiest channel, thereby improving sending efficiency.

[0033] Preferably, as an improvement, the intelligent channel selection module gradually adjusts the flow of each channel by a gradient descent method based on the preliminary configuration.

[0034] Technical effect: It is easy to reduce the instability of the system caused by flow fluctuations in each channel.

[0035] Preferably, as an improvement, it further includes a fault analysis module, which is used to obtain the error type of the eliminated channel, and the error type includes network error, message queue error, authentication error, content violation, receiver error, and insufficient system resources. When the error type is content violation, an association table of feature information of the channel information and the service type is obtained, and feature clustering analysis is performed to obtain a clustering result;

[0036] The prompt module is used to prompt the error type and clustering results.

[0037] Technical effect: When the content violates the regulations, the SMS templates all contain the same features. Through analysis, it is easy to quickly obtain the violation features and make timely modifications. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The figure is a structural diagram of an intelligent configuration system for SMS channels. DETAILED DESCRIPTION

[0039] The following is further described in detail through specific implementation methods:

[0040] The embodiment is basically as shown in the attached Figure 1 As shown:

[0041] A short message channel intelligent configuration system, comprising:

[0042] The demand and attribute acquisition module is used to obtain the demand and demand attributes for sending SMS messages in the expected time period. In this embodiment, the expected time period is determined according to the product demand, such as the need to communicate with customers by SMS n days before the repayment date or m days after the overdue date; the demand attributes include the business type and quantity. The demand and attribute acquisition module also includes: a feature association submodule, which is used to obtain the SMS template of the business type and obtain the feature information of the SMS template. The feature information includes the string length, keywords, and SMS format, and creates an association table between the feature information and the business type, so that when an error occurs, the illegal features can be quickly locked, and modifications can be made in time to restore channel operation; a customized strategy acquisition submodule, which is used to obtain the customized strategy of the business type. The customized strategy includes the customized permission channel and the priority of the customized permission channel. Considering that different business types correspond to different products, the importance of the products themselves is different, and the products have customized SMS notification strategies, by obtaining customized strategies, it is convenient to improve the personalization of SMS sending.

[0043] The load acquisition module is used to obtain the limit load of each SMS channel. The limit load is the maximum number of messages that the SMS channel can process per unit time. The limit load is affected by multiple factors, including the hardware performance of the SMS gateway, network bandwidth, operator restrictions, and system architecture design. The limit load in this embodiment is obtained through testing. The test process includes: determining the test environment, ensuring that the test environment is consistent with the production environment, including hardware configuration and network environment, preparing test tools and monitoring tools, formulating a test plan, designing test scenarios for different levels of concurrent users and message numbers, determining the time period and duration of the test, performing stress testing according to the test plan, recording system performance, monitoring various system indicators in real time, recording performance data, analyzing test results, adjusting system configuration and optimizing performance according to the test results, and determining the limit load of the system under different conditions according to the test results.

[0044] The preliminary configuration module obtains the authorized SMS channel based on the business type and performs preliminary configuration according to the authorized SMS channel; the configuration rules of the preliminary configuration are: for business types with customized policies, the customized policies are configured first; for business types without customized policies, the channels are configured according to the historical configuration. When the SMS channel exceeds the load, it is randomly configured in the channel that meets the load. The preliminary configuration facilitates the priority configuration of SMS for key business types and avoids the overload of a single channel.

[0045] The channel health analysis module is used to monitor and obtain the performance data of each SMS channel in real time, and analyze the channel health based on the performance data; the performance data includes the success rate of sending requests, average response time, error type and frequency;

[0046] The success rate of sending requests refers to the ratio of the number of SMS messages that are successfully sent and confirmed by the other party (such as SMS arrival receipt) to the total number of sending requests. The calculation model is:

[0047]

[0048] Among them, n i is the number of SMS messages successfully sent by channel i, n i0 is the total number of requests sent to channel i.

[0049] The average response time refers to the average time interval from the system sending a text message request to receiving a notification of success or failure. The average response time reflects the speed and efficiency of the system in processing requests. Considering the average response time will help improve the response speed and efficiency of the SMS channel.

[0050] Error type refers to the classification of problems that may be encountered during SMS sending. Error frequency refers to the ratio of the number of times a specific error type occurs within a period of time to the total number of sending requests. It is convenient for identifying the most common problem areas in the system. The error frequency calculation model is:

[0051]

[0052] Among them, m iu is the number of errors of error type u in channel i.

[0053] The analysis model for analyzing channel health based on performance data is:

[0054]

[0055] Among them, a 1 、a 2 、a 3 is the health ratio adjustment constant factor, C i is the success rate of sending requests, x i is the average response time of channel i, q iu is the error frequency of error type u in channel i, ρ u is the weight value of error type u.

[0056] The health ratio adjustment constant factor affects the weight of each performance data. The health ratio adjustment constant factor is dynamically adjusted according to the time period, so as to improve the flexibility of channel traffic configuration according to business needs and channel characteristics; for example, during peak hours, response time is more important than usual. By dynamically adjusting the health ratio adjustment constant factor to improve response speed, a better user experience is achieved.

[0057] The intelligent channel selection module intelligently selects channels based on the intelligent channel selection probability model, and adjusts the SMS channels of each service type based on the preliminary configuration, and sends SMS messages on the selected SMS channels. Considering that the fluctuation of the flow of each channel will cause system instability, this embodiment gradually adjusts the flow of each channel by gradient descent based on the preliminary configuration, with the goal of minimizing the delay time. The total delay time is minimized by adjusting the flow of each channel. Specifically, the objective function is defined as:

[0058] G(y 1 ,y 2 ,……,y R )=Σd i (y i )

[0059] Among them, i∈[1,2……,R],d i (y i ) is the flow rate y of the i-th channel iDelay time under; Set initial flow distribution y i (0) and learning rate γ; calculate the objective function gradient, the gradient calculation model is:

[0060]

[0061] Update the flow of each channel according to the gradient, and update the model as follows:

[0062]

[0063] Where t is the number of iterations.

[0064] The process of calculating gradients and updating flows is repeated until the objective function converges or the maximum number of iterations is reached.

[0065] The intelligent channel selection module includes: a channel triggering and elimination submodule, which is used to determine whether the channel health meets the threshold. If it is determined to be yes, the channel is triggered; if not, the channel is eliminated; an intelligent selection submodule, which is used to create an intelligent channel selection probability model based on the channel health. The intelligent selection probability model is:

[0066]

[0067] t i ∈t i_m

[0068] L i ≤L i0

[0069] Among them, J i is the health of channel i, T i_m The probability of selecting channel i for service type m, t i is the selected SMS channel, t i_m Indicates the permission channel of business type m; L i is the current flow rate of channel i, L i0 is the limit load of channel i.

[0070] It also includes a fault analysis module, which is used to obtain the error type of the rejection channel, including network error, message queue error, authentication error, content violation, receiver error, and insufficient system resources. When the error type is content violation, the characteristic information of the channel information and the association table of the business type are obtained, and characteristic clustering analysis is performed to obtain the clustering result; the prompt module is used to prompt the error type and the clustering result. When the content violates the rules, the SMS template contains the same features. Through analysis, it is easy to quickly obtain the violation features and make modifications in time.

[0071] The above is only an embodiment of the present invention, and the common knowledge such as the known specific technical solutions and / or characteristics in the solution is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several modifications and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. An intelligent configuration system for SMS channels, characterized in that: include: A demand and attribute acquisition module, used to acquire the demand and demand attributes of SMS sending in the expected time period, wherein the demand attributes include the service type and quantity; The load acquisition module is used to obtain the limit load of each SMS channel; The preliminary configuration module obtains the authorized SMS channel based on the business type and performs preliminary configuration based on the authorized SMS channel; The channel health analysis module is used to monitor and obtain the performance data of each SMS channel in real time, and analyze the channel health based on the performance data; The intelligent channel selection module intelligently selects a channel based on an intelligent channel selection probability model, adjusts the SMS channel of each service type based on the preliminary configuration, and sends SMS messages on the selected SMS channel.

2. The SMS channel intelligent configuration system according to claim 1, characterized in that: The demand and attribute acquisition module also includes: The feature association submodule is used to obtain the SMS template of the business type, obtain the feature information of the SMS template, and create an association table between the feature information and the business type; The customized policy acquisition submodule is used to acquire the customized policy of the business type, wherein the customized policy includes the customized authority channel and the priority of the customized authority channel.

3. The SMS channel intelligent configuration system according to claim 2, characterized in that: The configuration rules of the preliminary configuration are: for business types with customized policies, the customized policies are configured first; for business types without customized policies, they are configured according to historical configuration channels. When the SMS channel exceeds the load, it is randomly configured in the channel that meets the load.

4. The SMS channel intelligent configuration system according to claim 1, characterized in that: The performance data includes the success rate of sending requests, average response time, error type and frequency; the analysis model for analyzing channel health based on performance data is: Among them, a1, a1, a3 are health ratio adjustment constant factors, C i is the success rate of sending requests, x i is the average response time of channel i, q iu is the error frequency of error type u in channel i, ρ u is the weight value of error type u.

5. The SMS channel intelligent configuration system according to claim 4, characterized in that: The health ratio adjustment constant factor is dynamically adjusted according to the time period.

6. A SMS channel intelligent configuration system according to claim 5, characterized in that: The intelligent channel selection module comprises: The channel triggering and rejection submodule is used to determine whether the channel health meets the threshold. If it is, the channel is triggered; if not, the channel is rejected. The intelligent selection submodule is used to create an intelligent channel selection probability model based on channel health. The intelligent selection probability model is: t i ∈t i_m L i ≤L i0 Among them, J i is the health of channel i, T i_m The probability of selecting channel i for service type m, t i is the selected SMS channel, t i_m Indicates the permission channel of business type m; L i is the current flow rate of channel i, L i0 is the limit load of channel i.

7. The SMS channel intelligent configuration system according to claim 1, characterized in that: The intelligent channel selection module gradually adjusts the flow of each channel by using a gradient descent method based on the preliminary configuration.

8. The SMS channel intelligent configuration system according to claim 7, characterized in that: It also includes a fault analysis module, which is used to obtain the error type of the rejection channel, and the error type includes network error, message queue error, authentication error, content violation, receiver error, and insufficient system resources. When the error type is content violation, the association table of the feature information of the channel information and the business type is obtained, and feature clustering analysis is performed to obtain a clustering result; The prompt module is used to prompt the error type and clustering results.

Images