A self-selection method for industry SMS channels based on naive Bayes algorithm
Through the industry SMS channel self-selecting method based on the Naive Bayes algorithm, the problem of inappropriate number segment ownership selection after number portability is solved, and efficient sending of industry SMS and logical simplification are achieved.
Patent Information
- Application Number
- CN202211729470.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-30
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2042-12-30
AI Technical Summary
In the scenario of number portability user, number segment ownership as the method of selecting SMS channels may lead to inappropriate channel selection, resulting in reduced industry SMS sending efficiency and business logic redundancy.
Using the industry SMS channel self-select method based on the Naive Bayes algorithm, by obtaining the characteristic factors after the target number portability is transferred to the network, the Naive Bayes classification model (based on the AdoBoost algorithm) determines the target operator and automatically selects the appropriate one. SMS gateways are used to send industry SMS messages.
It realizes the independent selection of industry SMS channels, reduces business logic redundancy, improves the efficiency of industry SMS sending, and ensures that the appropriate operator channel can be accurately selected after number portability.
Smart Images

Figure CN116095619B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of communication technology, and in particular to an industry short message channel self-selection method based on a naive Bayes algorithm. Background Art
[0002] With the popularization of intelligence, industry SMS is one of the essential capabilities in the scenario of IoT intelligence. Industry SMS refers to SMS without marketing nature, such as verification code, information notification, etc. The commonly used SMS gateways are the internal SMS gateways of the three major operators and other third-party SMS gateways. In the prior art, when sending industry SMS, it will choose to go through the internal channel of the operator or the external channel of a third party according to the current number segment of the number card. However, the number of users who port their numbers is growing exponentially. The transfer of users from one operator to another will inevitably lead to the weakening of the relationship between the operator and the number segment. Therefore, using the number segment as the channel for sending industry SMS may result in the selection of an inappropriate channel, which in turn makes the business logic redundant and greatly reduces the efficiency of sending industry SMS.
[0003] Therefore, a solution that can autonomously select matching channels for sending industry text messages has research significance. Summary of the invention
[0004] In view of this, an embodiment of the present invention provides an industry SMS channel self-selection method based on a naive Bayes algorithm, which can autonomously select a matching channel for sending industry SMS.
[0005] An aspect of an embodiment of the present invention provides an industry SMS channel self-selection method based on a naive Bayes algorithm, comprising:
[0006] Obtaining a first characteristic factor of the cardholder after the target number card is ported, where the first characteristic factor includes multiple characteristics of the cardholder;
[0007] Sending an industry SMS message carrying the first characteristic factor to the naive Bayes classification model to obtain the target operator to which the target number card currently belongs, the naive Bayes classification model is based on the AdoBoost algorithm and is trained with a second characteristic factor marked with an operator label as training data, the second characteristic factor includes multiple characteristics of multiple training number cards after number portability, and each characteristic in the second characteristic factor corresponds to the same as each characteristic in the first characteristic factor;
[0008] The industry SMS is sent to a target SMS gateway corresponding to the target operator, so that the target SMS gateway sends the industry SMS to the cardholder.
[0009] Preferably, the training process of the naive Bayes classification model includes:
[0010] Obtaining the second characteristic factors of multiple training number cards after number portability, wherein the second characteristic factors of each training number card are the first N characteristics with the largest weight fluctuation degree among the characteristic factors of each training number card when the portability changes, where N is a positive integer;
[0011] Dividing the second characteristic factor into a plurality of stock data sets based on years;
[0012] Based on the Naive Bayes algorithm, cluster the stock data set with the earliest year and output the class with the largest posterior probability;
[0013] The remaining stock data sets are discretized into years in chronological order, and the features of each remaining and discretized stock data set are extracted based on the AdoBoost algorithm to construct a naive Bayes transient classification model.
[0014] Constructing a naive Bayes original classification model based on the naive Bayes algorithm and the second characteristic factor;
[0015] According to the loss value of the classification result output by the naive Bayes transient classification model for the second feature factor compared with the naive Bayes original classification model, the naive Bayes transient classification model is iterated a set number of times to obtain the trained naive Bayes classification model.
[0016] Preferably, the process of determining the first characteristic factor includes:
[0017] Obtain the number card information generated when the target number card is used;
[0018] The time in the number card information is processed as the first discrete data in units of years, and the remaining information of the number card information is processed as the second discrete data with conditional independence assumption according to the requirements of the naive Bayes algorithm;
[0019] Removing outliers that do not meet preset requirements from the first and second discrete data;
[0020] Classify the first and second discrete data using equal interval classification and equal frequency classification to obtain basic classification results;
[0021] Index filtering is performed on each feature in the basic classification result to obtain a first feature factor.
[0022] Preferably, before sending the industry SMS message carrying the first characteristic factor to the naive Bayes classification model, the method further includes:
[0023] Sending the industry SMS message carrying the first characteristic factor to a pre-configured number card characteristic cache system;
[0024] The sending the industry text message carrying the first characteristic factor to the naive Bayes classification model includes:
[0025] The industry text message carrying the first characteristic factor is sent to the naive Bayes classification model through the number card feature cache system.
[0026] Preferably, after the target SMS gateway sends the industry SMS to the cardholder, it also includes:
[0027] The naive Bayes classification model is trained using the first characteristic factor labeled with the target operator label as training data.
[0028] Preferably, before sending the industry SMS to the target SMS gateway corresponding to the target operator, the method further includes:
[0029] If there is an abnormal feature in the first feature factor whose default degree is higher than a set threshold, the industry text message carrying the first feature factor is re-input into the naive Bayes classification model to obtain the operator to which the naive Bayes classification model classifies the first feature factor again.
[0030] Preferably, sending the industry SMS to a target SMS gateway corresponding to the target operator includes:
[0031] If the target operator is a self-owned operator, the industry SMS is sent to the self-owned internal SMS gateway, so that the self-owned internal SMS gateway sends the industry SMS to the cardholder;
[0032] If the target operator is not a self-owned operator, the industry SMS is sent to a third-party SMS gateway so that the third-party SMS gateway sends the industry SMS to the cardholder.
[0033] Preferably, it also includes:
[0034] If the self-owned internal SMS gateway is unable to send the industry SMS to the cardholder, the industry SMS will be sent to the third-party SMS gateway via the self-owned internal SMS gateway, so that the third-party SMS gateway can send the industry SMS to the cardholder.
[0035] Another aspect of the embodiment of the present invention further provides an industry short message channel self-selection device based on a naive Bayes algorithm, comprising:
[0036] A characteristic factor acquisition unit, used to acquire a first characteristic factor of the cardholder after the target number card is ported, wherein the first characteristic factor includes multiple characteristics of the cardholder;
[0037] An operator classification unit is used to send an industry SMS carrying the first characteristic factor to a naive Bayes classification model to obtain a target operator to which the target number card currently belongs, wherein the naive Bayes classification model is based on an AdoBoost algorithm and is trained using a second characteristic factor labeled with an operator label as training data, wherein the second characteristic factor includes multiple characteristics of multiple training number cards after number portability, and each characteristic in the second characteristic factor corresponds to the same as each characteristic in the first characteristic factor;
[0038] The industry SMS sending unit is used to send the industry SMS to the target SMS gateway corresponding to the target operator, so that the target SMS gateway sends the industry SMS to the cardholder.
[0039] Another aspect of an embodiment of the present invention further provides an electronic device, including a processor and a memory;
[0040] The memory is used to store programs;
[0041] The processor executes the program to implement the above-mentioned industry SMS channel self-selection method based on the naive Bayes algorithm.
[0042] Another aspect of an embodiment of the present invention further provides a computer-readable storage medium, which stores a program, and the program is executed by a processor to implement the above-mentioned industry SMS channel self-selection method based on the naive Bayes algorithm.
[0043] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the above-mentioned industry SMS channel self-selection method based on the naive Bayes algorithm.
[0044] The method for self-selecting an industry SMS channel based on a naive Bayes algorithm of the present invention first obtains a first characteristic factor including multiple characteristics of a cardholder after the target card number is ported to another network; before sending the industry SMS, the industry SMS carrying the first characteristic factor is first sent to a naive Bayes classification model, and a classification result of the naive Bayes classification model on the operator to which the target card number currently belongs can be obtained, and the corresponding target operator can be determined according to the classification result, wherein the naive Bayes classification model is based on an AdoBoost algorithm and is trained using a second characteristic factor labeled with an operator label as training data; on the basis of knowing the target operator to which the target card number belongs before sending the industry SMS, a target SMS gateway corresponding to the target operator can be automatically selected, and then the channel of the target SMS gateway is used to send the industry SMS to the cardholder, and even if the cardholder ports the number to another network, the channel corresponding to the operator to which the target card number currently belongs can be automatically selected to send the industry SMS, thereby achieving the technical effect of self-selecting the industry SMS channel, reducing the logical redundancy of the business of sending industry SMS, and improving efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0046] Figure 1 A flowchart of a method for self-selecting an industry SMS channel based on a naive Bayesian algorithm provided by an embodiment of the present invention;
[0047] Figure 2 A flowchart of a specific example of industry SMS channel self-selection based on a naive Bayes algorithm provided in an embodiment of the present invention;
[0048] Figure 3 A flowchart of a training example of a naive Bayesian classification model based on the AdoBoost algorithm provided in an embodiment of the present invention;
[0049] Figure 4 A structural block diagram of an industry SMS channel self-selection device based on a naive Bayes algorithm provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0050] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0051] Reference Figure 1 The embodiment of the present invention provides an industry SMS channel self-selection method based on a naive Bayesian algorithm, which specifically includes the following steps S100 to S120. In addition, in order to more clearly introduce the industry SMS channel self-selection method based on a naive Bayesian algorithm of the present invention, refer to Figure 2 The embodiment of the present invention can also provide a flowchart of a specific example of industry SMS channel self-selection based on the Naive Bayes algorithm.
[0052] Step S100: Obtain a first characteristic factor of the cardholder after the target number card is ported, where the first characteristic factor includes multiple characteristics of the cardholder.
[0053] Specifically, the information generated when the cardholder uses the target card number may include: age, place of origin, household registration, initiating network, number segment network, outgoing operator or incoming operator, etc., as well as other optional information. The above information can be used as the characteristics of the cardholder, and the first characteristic factor can include multiple characteristics of the cardholder.
[0054] Step S110: sending the industry text message carrying the first characteristic factor to the naive Bayes classification model to obtain the target operator to which the target number card currently belongs.
[0055] Specifically, the naive Bayes classification model can be based on the AdoBoost algorithm and trained with the second characteristic factor labeled with the operator label as training data. The industry SMS carrying the first characteristic factor is input into the naive Bayes classification model to obtain the classification result output by the naive Bayes classification model for the operator to which the target card number belongs, and then the target operator to which the target card number currently belongs can be determined based on the classification result.
[0056] The construction process of the above naive Bayes classification model is as follows:
[0057] The Naive Bayes method makes a conditional independence assumption on the characteristic factors of the target card holder. Specifically, the conditional independence assumption may include:
[0058] P(X=x|Y=c k )=P(X (1) =x (1) , …, X (n) =X (n) |Y=c k )=ПP(X (j) =x (j) |Y=c k )
[0059] When using the naive Bayes method for classification, the posterior probability distribution is calculated by the model for the given multiple feature inputs x:
[0060] PY=c k |X=x
[0061] The class with the largest posterior probability is taken as the class output of the feature input x.
[0062] Based on the above theory, for the characteristic factors of the target card holder, the basic formula of naive Bayes classification is determined:
[0063]
[0064] The naive Bayes classification model formula corresponding to the above basic formula is:
[0065]
[0066] In addition, the second characteristic factor may include multiple characteristics of multiple training number cards after number portability, and each characteristic in the second characteristic factor may correspond to the same as each characteristic in the first characteristic factor.
[0067] Step S120: Send the industry SMS to the target SMS gateway corresponding to the target operator, so that the target SMS gateway sends the industry SMS to the cardholder.
[0068] Specifically, according to different target operators determined by the naive Bayes classification model, there may be different processes for autonomously selecting different channels to send industry SMS to the corresponding target SMS gateway, which may specifically include the following:
[0069] Case 1: if the target operator is a self-owned operator, the industry SMS is sent to the self-owned internal SMS gateway, so that the self-owned internal SMS gateway sends the industry SMS to the cardholder.
[0070] Case 2: if the target operator is not a self-owned operator, the industry SMS is sent to a third-party SMS gateway so that the third-party SMS gateway sends the industry SMS to the cardholder.
[0071] In addition, if after the industry SMS is sent to its own internal SMS gateway, the internal SMS gateway determines that the operator of the industry SMS is not its own operator, and thus cannot send the industry SMS to the cardholder, then the industry SMS can be sent to a third-party SMS gateway through its own internal SMS gateway, so that the third-party SMS gateway can send the industry SMS to the cardholder.
[0072] In some embodiments of the present invention, the above step S100 is introduced, which is a process of obtaining the first characteristic factor of the cardholder after the target number card is ported. Next, the process of determining the first characteristic factor will be further explained.
[0073] S1. Obtain the number card information generated when the target number card is used.
[0074] Specifically, the card number information may include the information generated when the cardholder uses the target card number as introduced in the above step S100: age, place of origin, household registration, initiating network, number segment network, outgoing operator or incoming operator, etc., as well as other optional information.
[0075] S2. The time in the number card information is processed as the first discrete data in units of years, and the remaining information of the number card information is processed as the second discrete data with conditional independence assumption according to the requirements of the naive Bayes algorithm.
[0076] Specifically, selecting the data type may include processing the time in the number card information as discrete data in units of years, and the remaining information of the number card information may be processed as discrete data with conditional independence assumption according to the requirements of the naive Bayes algorithm. The process of conditional independence assumption may specifically refer to the construction process of the above-mentioned naive Bayes classification model.
[0077] S3. Remove abnormal values that do not meet preset requirements from the first and second discrete data.
[0078] Specifically, features with missing attribute values may be removed, and effective outliers may also be removed, where effective outliers may include but are not limited to invalid outliers and valid outliers.
[0079] S4. Classify the first and second discrete data using equal interval classification and equal frequency classification to obtain basic classification results.
[0080] Specifically, since the number card information is processed as discrete data, if there are too many discrete types, the data of the same nature or magnitude can be clustered and roughly processed. For example, the data can be replaced by province category according to the place of origin.
[0081] Then, in order to reduce the number of samples, equal interval classification and equal frequency classification can be used for coarse classification to obtain the basic classification results of coarse classification.
[0082] S5. Perform index filtering on each feature in the basic classification result to obtain a first feature factor.
[0083] Specifically, all features in the number card information can be filtered according to the basic classification. Since the input is a discrete variable, the output can also be a discrete target variable. Then, indexes such as information difference or Krem value can be used to filter the features in the basic classification results to obtain the first characteristic factor.
[0084] In some embodiments of the present invention, the above-mentioned step S110 is introduced, which is a process of sending an industry SMS carrying the first characteristic factor to the naive Bayes classification model to obtain the target operator to which the target number card currently belongs. The training process of the naive Bayes classification model will be further explained below.
[0085] Specifically, it may include:
[0086] S1. Obtain the second characteristic factors of multiple training number cards after number portability. The second characteristic factors of each training number card are the top N characteristics with the largest weight fluctuation among the characteristic factors of each training number card when the state changes, and N is a positive integer.
[0087] Specifically, each training number card may include multiple features, which can be sorted according to the severity of the fluctuation of the weight of the feature factors when the state changes, and the top N feature factors with the largest fluctuation degree are selected as the second feature factors. Wherein, N is a positive integer, and the embodiment of the present invention can take N as 10 as an example, that is, the second feature factor of the embodiment of the present invention may include the top 10 features with the largest weight fluctuation degree among the feature factors.
[0088] S2. Divide the second characteristic factor into multiple stock data sets based on years.
[0089] Specifically, the time of number portability of each training number card may be quite long ago. In order to facilitate the division of the second characteristic factor, under an optional implementation method, the embodiment of the present invention can divide the second characteristic factor into multiple data sets based on years, and each data set is recorded as a stock data set.
[0090] Specifically, assuming that the time of number portability for each training number card is 2003, and the cutoff statistical time for using the characteristics of the training number cards is 2021, then the second characteristic factors of the training number cards from 2003 to 2021 can be divided into years, and the second characteristic factors for each year from 2003 to 2021 can be obtained and stored as a stock data set.
[0091] S3. Based on the Naive Bayes algorithm, cluster the stock data set with the earliest year and output the class with the largest posterior probability.
[0092] Specifically, taking the second characteristic factor of each year from 2003 to 2021 as an example, the stock data set of the 2003 training number cards can be preprocessed. The preprocessing process can refer to the determination process of the first characteristic factor mentioned above, and perform a clustering operation based on the naive Bayes algorithm based on all the features in the stock data set of 2003 to output the class with the largest posterior probability.
[0093] S4. Discretize the remaining stock data sets into years according to the chronological order, perform feature extraction on each remaining and discretized stock data set based on the AdoBoost algorithm, and construct a naive Bayes transient classification model.
[0094] Specifically, still taking the second characteristic factor of each year from 2003 to 2021 as an example, the stock data set of each year from 2004 to 2021 can be discretized into years, and the data can be preprocessed year by year. Feature extraction and model iteration are performed based on the principle of AdoBoost algorithm to construct a transient naive Bayes transient classification model.
[0095] S5. Constructing a Naive Bayes original classification model based on the Naive Bayes algorithm and the second characteristic factor.
[0096] Specifically, we can build a corresponding naive Bayes original classification model based on the idea of naive Bayes algorithm and the second characteristic factors of each year.
[0097] S6. According to the loss value of the classification result output by the naive Bayes transient classification model for the second feature factor compared with the naive Bayes original classification model, the naive Bayes transient classification model is iterated a set number of times to obtain the trained naive Bayes classification model.
[0098] Reference Figure 3 , Figure 3 A flowchart of a training example of a naive Bayesian classification model based on the AdoBoost algorithm provided in an embodiment of the present invention.
[0099] Specifically, the stock data set of the next year can be used to perform regression tests on the naive Bayes original classification model and the naive Bayes transient classification model, and the accuracy and loss values can be calculated to compare the performance of the two models.
[0100] Then, according to the results of the performance comparison, the learning rate and other parameters of AdoBoost are dynamically adjusted to make the naive Bayes transient classification model more accurate.
[0101] Taking the data used as the second characteristic factor of each year from 2003 to 2021 as an example, after 17 iterations, the final naive Bayes transient classification model is output as the naive Bayes classification model.
[0102] Furthermore, the above-mentioned step S120, after the target SMS gateway sends the industry SMS to the cardholder, may also include:
[0103] The naive Bayes classification model is trained using the first characteristic factor labeled with the target operator label as training data.
[0104] Specifically, after each industry SMS is sent, the first feature information carried by the industry SMS can be used as a training sample, and the operator to which the target number card belongs can be used as a training label to continue training the naive Bayes classification model. Therefore, once deployed, the naive Bayes classification model has the ability to learn autonomously, does not require subsequent manual maintenance, and can complete iterations by itself. In addition, if any emergency occurs, it can also be adjusted through the bias model of AdoBoost to adapt to any extreme situations.
[0105] Considering that some features of the first feature factor may be defaulted, before sending the industry SMS to the target SMS gateway corresponding to the target operator in the above step S120, the following may also be included:
[0106] If there is an abnormal feature in the first feature factor whose default degree is higher than a set threshold, the industry text message carrying the first feature factor is re-input into the naive Bayes classification model to obtain the operator to which the naive Bayes classification model classifies the first feature factor again.
[0107] In order to improve the prediction efficiency of the naive Bayes classification model, the embodiment of the present invention may further include, before the above step S110, sending the industry SMS carrying the first feature factor to the naive Bayes classification model:
[0108] The industry SMS carrying the first characteristic factor is sent to a pre-configured number card characteristic cache system.
[0109] Specifically, for the same batch of requests, multiple industry text messages carrying the first characteristic factor can be cached into the number card feature cache system first, and then each first characteristic factor in the number card feature cache system can be quickly predicted to improve the prediction efficiency of the naive Bayes classification model.
[0110] On this basis, the above step S110, sending the industry text message carrying the first characteristic factor to the naive Bayes classification model, may include: sending the industry text message carrying the first characteristic factor to the naive Bayes classification model through the number card feature cache system.
[0111] Next, the practical application scenarios of the present invention will be described with specific examples.
[0112] Specifically, to achieve the technical effect of the present invention, it is necessary to have the ability to send industry SMS service API, model capability integrated management console and SMS gateway capabilities. The industry SMS sending capability is provided in the form of API, which can be attached to the external API management and marketing platform of the platform itself for unified management and control. The model capability integrated management console is designed as a Web terminal, with a developer view (a unified service entrance provided to developers, providing them with user management, application management, capability directory, statistical query, help center and other functions), an operator view (the operator can manage applications, capabilities, APIs, and management system configuration, monitoring, resources, security, etc. through the view to ensure the normal operation of the API capability opening platform), and an operator view (a unified entrance provided to the operation and management personnel of the capability opening platform, providing them with management functions such as developer management, application management, capability management, statistical query, system security management, agency review management, announcement management, service management, system management, log management, and monitoring management, and providing a one-stop process-based operation and maintenance control service entrance view). In this way, the naive Bayes classification model capabilities can be sustainably developed, reused, and monitored.
[0113] Reference Figure 4 The embodiment of the present invention provides an industry short message channel self-selection device based on a naive Bayes algorithm, comprising:
[0114] A characteristic factor acquisition unit, used to acquire a first characteristic factor of the cardholder after the target number card is ported, wherein the first characteristic factor includes multiple characteristics of the cardholder;
[0115] An operator classification unit is used to send an industry SMS carrying the first characteristic factor to a naive Bayes classification model to obtain a target operator to which the target number card currently belongs, wherein the naive Bayes classification model is based on an AdoBoost algorithm and is trained using a second characteristic factor labeled with an operator label as training data, wherein the second characteristic factor includes multiple characteristics of multiple training number cards after number portability, and each characteristic in the second characteristic factor corresponds to the same as each characteristic in the first characteristic factor;
[0116] The industry SMS sending unit is used to send the industry SMS to the target SMS gateway corresponding to the target operator, so that the target SMS gateway sends the industry SMS to the cardholder.
[0117] The embodiment of the present invention also discloses a computer program product or a computer program, wherein the computer program product or the computer program includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes Figure 1 The method shown.
[0118] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.
[0119] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions, and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.
[0120] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.
[0121] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.
[0122] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.
[0123] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0124] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0125] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.
[0126] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.
Claims
1. A method for self-selection of industry SMS channels based on naive Bayes algorithm, characterized in that: include: Obtaining a first characteristic factor of the cardholder after the target number card is ported, where the first characteristic factor includes multiple characteristics of the cardholder; Sending an industry SMS message carrying the first characteristic factor to the naive Bayes classification model to obtain the target operator to which the target number card currently belongs, the naive Bayes classification model is based on the AdoBoost algorithm and is trained with a second characteristic factor marked with an operator label as training data, the second characteristic factor includes multiple characteristics of multiple training number cards after number portability, and each characteristic in the second characteristic factor corresponds to the same as each characteristic in the first characteristic factor; Sending the industry SMS to a target SMS gateway corresponding to the target operator, so that the target SMS gateway sends the industry SMS to the cardholder; The training process of the naive Bayes classification model includes: Obtaining the second characteristic factors of multiple training number cards after number portability, wherein the second characteristic factors of each training number card are the first N characteristics with the largest weight fluctuation degree among the characteristic factors of each training number card when the portability changes, where N is a positive integer; Dividing the second characteristic factor into a plurality of stock data sets based on years; Based on the Naive Bayes algorithm, cluster the stock data set with the earliest year and output the class with the largest posterior probability; The remaining stock data sets are discretized into years in chronological order, and the features of each remaining and discretized stock data set are extracted based on the AdoBoost algorithm to construct a naive Bayes transient classification model. Constructing a naive Bayes original classification model based on the naive Bayes algorithm and the second characteristic factor; According to the loss value of the classification result output by the naive Bayes transient classification model for the second feature factor compared with the naive Bayes original classification model, the naive Bayes transient classification model is iterated a set number of times to obtain the trained naive Bayes classification model.
2. The industry SMS channel self-selection method based on the naive Bayes algorithm according to claim 1 is characterized in that: The process of determining the first characteristic factor includes: Obtain the number card information generated when the target number card is used; The time in the number card information is processed as the first discrete data in units of years, and the remaining information of the number card information is processed as the second discrete data with conditional independence assumption according to the requirements of the naive Bayes algorithm; Removing outliers that do not meet preset requirements from the first and second discrete data; Classify the first and second discrete data using equal interval classification and equal frequency classification to obtain basic classification results; Index filtering is performed on each feature in the basic classification result to obtain a first feature factor.
3. The industry SMS channel self-selection method based on the naive Bayes algorithm according to claim 1 is characterized in that: Before sending the industry text message carrying the first characteristic factor to the naive Bayes classification model, the method further includes: Sending the industry SMS message carrying the first characteristic factor to a pre-configured number card characteristic cache system; The sending the industry text message carrying the first characteristic factor to the naive Bayes classification model includes: The industry text message carrying the first characteristic factor is sent to the naive Bayes classification model through the number card feature cache system.
4. The industry SMS channel self-selection method based on the naive Bayes algorithm according to claim 1 is characterized in that: After the target SMS gateway sends the industry SMS to the cardholder, the method further includes: The naive Bayes classification model is trained using the first characteristic factor labeled with the target operator label as training data.
5. The industry SMS channel self-selection method based on the naive Bayes algorithm according to claim 1 is characterized in that: Before sending the industry SMS to the target SMS gateway corresponding to the target operator, the method further includes: If there is an abnormal feature in the first feature factor whose default degree is higher than a set threshold, the industry text message carrying the first feature factor is re-input into the naive Bayes classification model to obtain the operator to which the naive Bayes classification model classifies the first feature factor again.
6. The industry SMS channel self-selection method based on the naive Bayes algorithm according to claim 1 is characterized in that: The sending the industry SMS to the target SMS gateway corresponding to the target operator includes: If the target operator is a self-owned operator, the industry SMS is sent to the self-owned internal SMS gateway, so that the self-owned internal SMS gateway sends the industry SMS to the cardholder; If the target operator is not a self-owned operator, the industry SMS is sent to a third-party SMS gateway so that the third-party SMS gateway sends the industry SMS to the cardholder.
7. The industry SMS channel self-selection method based on the naive Bayes algorithm according to claim 6 is characterized in that: Also includes: If the self-owned internal SMS gateway is unable to send the industry SMS to the cardholder, the industry SMS will be sent to the third-party SMS gateway via the self-owned internal SMS gateway, so that the third-party SMS gateway can send the industry SMS to the cardholder.
8. An industry SMS channel self-selection device based on the naive Bayes algorithm, characterized in that: include: A characteristic factor acquisition unit, used to acquire a first characteristic factor of the cardholder after the target number card is ported, wherein the first characteristic factor includes multiple characteristics of the cardholder; An operator classification unit is used to send an industry SMS carrying the first characteristic factor to a naive Bayes classification model to obtain a target operator to which the target number card currently belongs, wherein the naive Bayes classification model is based on an AdoBoost algorithm and is trained using a second characteristic factor labeled with an operator label as training data, wherein the second characteristic factor includes multiple characteristics of multiple training number cards after number portability, and each characteristic in the second characteristic factor corresponds to the same as each characteristic in the first characteristic factor; An industry SMS sending unit, used for sending the industry SMS to a target SMS gateway corresponding to the target operator, so that the target SMS gateway sends the industry SMS to the cardholder; The training process of the naive Bayes classification model includes: Obtaining the second characteristic factors of multiple training number cards after number portability, wherein the second characteristic factors of each training number card are the first N characteristics with the largest weight fluctuation degree among the characteristic factors of each training number card when the portability changes, where N is a positive integer; Dividing the second characteristic factor into a plurality of stock data sets based on years; Based on the Naive Bayes algorithm, cluster the stock data set with the earliest year and output the class with the largest posterior probability; The remaining stock data sets are discretized into years in chronological order, and the features of each remaining and discretized stock data set are extracted based on the AdoBoost algorithm to construct a naive Bayes transient classification model. Constructing a naive Bayes original classification model based on the naive Bayes algorithm and the second characteristic factor; According to the loss value of the classification result output by the naive Bayes transient classification model for the second feature factor compared with the naive Bayes original classification model, the naive Bayes transient classification model is iterated a set number of times to obtain the trained naive Bayes classification model.
9. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the industry SMS channel self-selection method based on the naive Bayes algorithm as described in any one of claims 1 to 7.
Citation Information
Patent Citations
Inter-carrier short messaging service providing phone number only experience
AU2002336423A1
Method, system and device for realizing short message intercommunication among networks under number portable service
CN101651933A
Information auditing apparatus and method
CN107483420A
Number portability number management method and system
CN110248321A
Number portability method, number portability server and operator server
CN113676855A
