A message sending method and message sending system based on message sending channel

Abstract

The present application relates to the field of communication technology, and specifically to a method and system for sending messages based on message sending channels. The method includes: sending a first message through at least one message sending channel; receiving receipt information corresponding to the first message from at least one message sending channel; determining a first weight and a second weight corresponding to each message sending channel in at least one message sending channel based on the receipt information received from at least one message sending channel; determining at least one first target message sending channel for a second message from at least one message sending channel based on the first weight and the second weight, wherein the first message and the second message are messages of the same type; sending the second message through at least one first target message sending channel, and dynamically selecting a message sending channel for message sending, thereby improving the accuracy of message sending and enhancing user experience.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a message sending method and a message sending system based on a message sending channel. Background Art

[0002] In today's information and digital age, the ways of interaction between enterprises and users are becoming increasingly diverse and intelligent, and messaging has become an important link between enterprises and users. As the core of this process, the messaging system can flexibly select appropriate messaging channels according to different scenarios. For example, in scenarios where user identity verification is required, verification codes are usually sent through SMS channels.

[0003] However, current messaging systems usually send messages through preset fixed messaging channels. For example, verification codes are usually sent through SMS channels by default. This fixed messaging method lacks flexibility and is difficult to adjust according to actual needs, thus affecting user experience. Summary of the invention

[0004] In order to solve the above problems, the embodiments of the present application propose a message sending method and a message sending system based on a message sending channel, which can dynamically select a message sending channel for message sending, thereby improving the accuracy of message sending and enhancing the user experience.

[0005] In order to achieve the above-mentioned objectives, in a first aspect, an embodiment of the present application provides a message sending method based on a message sending channel, including: sending a first message through at least one message sending channel; receiving receipt information corresponding to the first message from at least one message sending channel; based on the receipt information received from at least one message sending channel, determining a first weight and a second weight corresponding to each message sending channel in at least one message sending channel; the first weight is a weight related to the user's behavior after receiving the first message through the message sending channel, and the second weight is a weight related to the message sending environment of the message sending channel; based on the first weight and the second weight, determining at least one first target message sending channel for the second message from at least one message sending channel, the first message and the second message are messages of the same type; and sending the second message through at least one first target message sending channel.

[0006] The embodiment of the present application determines the first weight corresponding to the user behavior and the second weight corresponding to the sending environment in each message sending channel by monitoring the receipt information of the first message. Furthermore, based on the first weight corresponding to the user behavior and the second weight corresponding to the sending environment, one or more target message sending channels for sending the second message can be selected from at least one message sending channel. In this way, the weight of the message sending channel is dynamically adjusted according to the user behavior and the message sending environment, thereby realizing flexible selection of the message sending channel, thereby improving the accuracy of message sending and enhancing the user experience.

[0007] In an optional implementation, sending a first message through at least one message sending channel includes: obtaining the first message to be sent; obtaining all message sending channels corresponding to the first message from a database according to the first message; wherein the database includes at least one message sending channel and a weight corresponding to at least one of the message sending channels; determining at least one message sending channel among all message sending channels corresponding to the first message based on a weight threshold; and sending the first message through at least one message sending channel. Using the above method, when you want to send a message, dynamically determine at least one message sending channel from the database for sending. In this way, flexible selection of message sending channels can be achieved, thereby improving the accuracy of message sending and enhancing user experience.

[0008] In an optional implementation, based on the receipt information received from at least one message sending channel, the first weight corresponding to each message sending channel in at least one message sending channel is determined, including: extracting features from the receipt information to obtain user behavior features corresponding to at least one message sending channel; user behavior features include at least one of user click behavior, user sharing behavior, user reply behavior, user reading depth, user operation status, user viewing behavior and user stay time; using the user behavior features corresponding to at least one message sending channel as the input of a pre-trained neural network model, and using the neural network model to output the first weight of at least one message sending channel. Using the above method, the corresponding first weight is dynamically generated by the neural network model, so that the selection strategy of the message sending channel can be continuously optimized, and it can adapt to changes in user behavior, thereby improving the accuracy of message sending and enhancing user experience.

[0009] In an optional implementation, based on the receipt information received from at least one message sending channel, determining the first weight corresponding to each message sending channel in at least one message sending channel includes: extracting features from the receipt information to obtain user behavior features corresponding to at least one message sending channel; user behavior features include at least one of user click behavior, user sharing behavior, user reply behavior, user reading depth, user operation status, user viewing behavior, and user stay time; summing the product of user behavior features and preset weight parameters to obtain the first weight of at least one message sending channel. Using the above method, the first weight is dynamically generated by calculating the product of user behavior features and preset weight parameters and summing them. In this way, the selection strategy of the message sending channel can be continuously optimized to adapt to changes in user behavior, thereby improving the accuracy of message sending and enhancing user experience.

[0010] In an optional implementation, based on the receipt information received from at least one message sending channel, determining the second weight corresponding to each message sending channel in at least one message sending channel includes: obtaining the sending status of the first message within a preset cycle period, the sending status including successful sending or failed sending; when the sending status of the first message is successful sending, recording the first time point corresponding to the successful sending of the first message; when the sending status of the first message is failed sending, recording the second time point corresponding to the failed sending of the first message; based on the first time point and the cycle node corresponding to the preset cycle period, obtaining the second weight of at least one message sending channel; or, based on the second time point and the cycle node corresponding to the preset cycle period, obtaining the second weight of at least one message sending channel. In the above manner, the second weight is dynamically generated by determining the sending environment of the current message. In this way, the selection strategy of the message sending channel can be continuously optimized, and it can adapt to changes in user behavior, thereby improving the accuracy of message sending and enhancing user experience.

[0011] In an optional implementation, at least one first target message sending channel of the second message is determined from at least one message sending channel based on the first weight and the second weight, including: determining the comprehensive weight corresponding to at least one message sending channel based on the first weight and the second weight; sorting at least one comprehensive weight, and storing the sorted at least one comprehensive weight and the message sending channel corresponding to the comprehensive weight in a database; obtaining all message sending channels and the corresponding comprehensive weight corresponding to the corresponding second message from the database; and determining at least one first target message sending channel among all message sending channels based on the weight threshold and the comprehensive weight. Using the above method, the target sending channel of the second message is dynamically selected from the database through the generated comprehensive weight and weight threshold. In this way, the selection strategy of the message sending channel can be continuously optimized, and can adapt to changes in user behavior, thereby improving the accuracy of message sending and enhancing user experience.

[0012] In an optional implementation, based on the first weight and the second weight, determining the comprehensive weight corresponding to at least one message sending channel includes: multiplying the first weight and the first weight coefficient to obtain a first product; multiplying the second weight and the second weight coefficient to obtain a second product; summing the first product and the second product to determine the comprehensive weight of at least one message sending channel. In the above manner, the comprehensive weight is dynamically generated by calculating the product of the first weight and the second weight and their corresponding weight coefficients and summing them. In this way, the weight of each message sending channel can be evaluated more accurately, and the optimal selection of the message sending channel can be achieved.

[0013] In an optional implementation, after sending the second message through at least one first target message sending channel, it also includes: receiving the sending status of the second message from at least one message sending channel; when the sending status of the second message is a sending failure, obtaining at least one second target message sending channel from a database, the weight corresponding to the second target message sending channel being less than the weight corresponding to the first target message sending channel; and resending the second message through at least one second target message sending channel. With the above method, after monitoring the failure of the second message to be sent, the message sending channel can be reselected for sending, so that the message arrival rate can be guaranteed and the user experience can be improved.

[0014] In an optional implementation, before using the user behavior characteristics corresponding to at least one message sending channel as the input of a pre-trained neural network model and using the neural network model to output the first weight of at least one message sending channel, it also includes: obtaining a training set, the training set includes training receipt information corresponding to at least one message sending channel and a training weight value corresponding to the training receipt information; performing feature extraction on the training receipt information to obtain user behavior training characteristics corresponding to at least one message sending channel; using the user behavior training characteristics corresponding to at least one message sending channel as the input of the neural network model and using at least one training weight value as the output of the neural network model to train the neural network model. Using the above method, the trained neural network model is used to automatically process weight issues, reduce manual intervention, and improve the efficiency and accuracy of weight determination.

[0015] In an optional implementation, the receipt information includes data for monitoring the message sending status and the user's behavior after receiving the message. In this way, the user's behavior and the message sending environment can be analyzed through the receipt information, and the message sending channel selection strategy can be further optimized to improve the accuracy of message sending and user experience.

[0016] In an optional implementation, after determining the first weight and the second weight corresponding to each of the at least one message sending channel based on the receipt information received from the at least one message sending channel, it also includes: determining the third weight corresponding to each of the at least one message sending channel based on the receipt information, where the device weight is a weight related to the message sending device of the message sending channel; determining at least one third target message sending channel based on the first weight, the second weight and the third weight; and sending the second message through the at least one third target message sending channel. In this way, the weight of the message sending channel is dynamically adjusted according to the user behavior and the message sending environment, thereby realizing the flexible selection of the message sending channel, thereby improving the accuracy of message sending and enhancing the user experience.

[0017] In a second aspect, an embodiment of the present application provides a message sending system based on message sending channels, including: a multi-channel integration module, configured to: send a first message through at least one message sending channel; the multi-channel integration module, further configured to: receive receipt information corresponding to the first message from at least one message sending channel; a user preference analysis module, configured to: determine a first weight corresponding to each message sending channel in at least one message sending channel based on the receipt information received from at least one message sending channel; the first weight is a weight related to the user's behavior after receiving the first message through the message sending channel; a network environment monitoring module, configured to: determine a second weight corresponding to each message sending channel in at least one message sending channel based on the receipt information received from at least one message sending channel; the second weight is a weight related to the message sending environment of the message sending channel; an intelligent dynamic adjustment module, configured to: determine at least one first target message sending channel of the second message from at least one message sending channel based on the first weight and the second weight, the first message and the second message are messages of the same type; the multi-channel integration module, further configured to: send the second message through at least one first target message sending channel.

[0018] The embodiment of the present application determines the first weight corresponding to the user behavior and the second weight corresponding to the sending environment in each message sending channel by monitoring the receipt information of the first message. Furthermore, based on the first weight corresponding to the user behavior and the second weight corresponding to the sending environment, one or more target message sending channels for sending the second message can be selected from at least one message sending channel. In this way, the weight of the message sending channel is dynamically adjusted according to the user behavior and the message sending environment, thereby realizing flexible selection of the message sending channel, thereby improving the accuracy of message sending and enhancing the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a first flow chart of a message sending method based on a message sending channel provided in an embodiment of the present application; Figure 2 is a second flow chart of a message sending method based on a message sending channel provided in an embodiment of the present application; Figure 3 It is a schematic diagram of a training method of a neural network model provided in an embodiment of the present application; Figure 4 This is a third flow chart of a message sending method based on a message sending channel provided in an embodiment of the present application; Figure 5 It is a schematic diagram of a message sending system based on a message sending channel provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments of the present application, other embodiments obtained by ordinary technicians in this field without making creative work all belong to the protection scope of the present application.

[0021] In the following, the terms "first", "second", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0022] In addition, in the present application, directional terms such as "upper", "lower", "inner" and "outer" are defined relative to the orientation of the components schematically placed in the drawings. It should be understood that these directional terms are relative concepts. They are used for relative description and clarification, and they can change accordingly according to the changes in the orientation of the components placed in the drawings.

[0023] To facilitate understanding of the solution, the following explains the relevant terms: Message sending channels refer to different ways of delivering messages, such as email channels, which send messages through email systems; SMS channels, which send messages through short message services provided by mobile phone operators; and instant messaging channels, which send real-time messages through some instant messaging programs (applications).

[0024] Database (DB) refers to a collection of systematic storage, management, and processing of data. It enables users and applications to efficiently manage data, including creating, querying, updating, and storing data.

[0025] The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0026] Messaging, also known as message sending, plays a vital role in modern information communication. With the rapid development of Internet technology, the interaction between enterprises and users has become increasingly complex. As a communication method, messaging can quickly, conveniently and effectively transmit information, improve work efficiency and improve user experience.

[0027] As the core of this process, the messaging system can flexibly select the appropriate messaging channel according to different scenarios. For example, in scenarios where user identity verification is required, the verification code is usually sent through a text message channel, while in scenarios where real-time reminders or updates are made within the application, the real-time reminders or updates are often sent through instant messaging channels. It can be seen that the current messaging system usually sends messages through preset fixed messaging channels. This fixed messaging channel selection method cannot be flexibly adjusted according to specific circumstances (such as user needs), so it is impossible to dynamically select the most appropriate messaging channel to meet user needs.

[0028] In order to solve the above problems, the embodiments of the present application provide a message sending method based on a message sending channel, which can dynamically select a message sending channel for message sending, thereby improving the accuracy of message sending and enhancing the user experience.

[0029] Figure 1 This is the first flowchart of a message sending method based on a message sending channel provided in an embodiment of the present application.

[0030] Figure 2 This is the second flow chart of a message sending method based on a message sending channel provided in an embodiment of the present application.

[0031] like Figure 1 and Figure 2 As shown, the message sending method based on the message sending channel provided in this embodiment includes the following steps: Step S101: Send a first message through at least one message sending channel.

[0032] In one implementation, step S101 includes steps S1011 to S1014.

[0033] Step S1011, obtaining the first message to be sent.

[0034] Messages are carriers of information and are transmitted in the form of language, text, images, data, etc. Messages can be classified according to their functions, such as authentication and security messages that authenticate identities and ensure user security, shopping messages used for sales promotion, etc.

[0035] Exemplarily, a first shopping message for promoting special offers, new product releases, etc. is obtained.

[0036] It should be understood that the specific type of the above-mentioned first message is not specifically limited.

[0037] Step S1012: According to the first message, all message sending channels corresponding to the first message are obtained from a database.

[0038] The database pre-stores multiple message sending channels corresponding to different message types and weights corresponding to the multiple message channels. These messages can be sent to users through multiple message sending channels, and the message sending channels can be set with different weights according to their effects and applicable scenarios.

[0039] Exemplarily, the database stores a first message for promoting special offers, new product launches, and other shopping-related items. The first message can be sent to the user via a text message channel, wherein the weight corresponding to the text message channel is 0.8; the first message can also be sent to the user via an email channel, wherein the weight corresponding to the email channel is 0.7; the first message can also be sent to the user via an instant messaging channel, wherein the weight corresponding to the instant messaging channel is 0.6.

[0040] Step S1013: determining at least one message sending channel among all message sending channels corresponding to the first message based on the weight threshold.

[0041] The weight threshold is set based on a specific standard or algorithm, such as the weight threshold is set to 0.5.

[0042] In one example, if the weight in the database is greater than or equal to a weight threshold, at least one message sending channel among all message sending channels corresponding to the first message is determined.

[0043] Continuing with the above example, if all the message sending channels corresponding to the first message include the above three message sending channels, SMS channel, email channel, and instant messaging channel, and the weights corresponding to the SMS channel, email channel, and instant messaging channel are all greater than the weight threshold, the three message sending channels corresponding to the first message are determined.

[0044] Step S1014: Send the first message through at least one message sending channel.

[0045] Following the above example, a first shopping message for promoting special offers, new product releases, etc. is sent through a text message channel, an email channel, and an instant messaging channel respectively.

[0046] Step S102: receiving receipt information corresponding to the first message from at least one message sending channel.

[0047] Receipt information includes data used to monitor the delivery status of messages and the user's behavior after receiving the message. The delivery status includes delivery completion, delivery success, and delivery failure. Behavioral data includes click-through rate, interactive behavior, in-app actions, feedback submission, unsubscription rate, and time spent.

[0048] Among them, the click-through rate is used to record the frequency of how many users click on the link or button in the message, which is very important for evaluating the user's interest and appeal in the message content. Interactive behaviors include but are not limited to replying, forwarding, liking and other operations, which can show the user's attitude towards the message and the degree of user participation in the discussion. In-app actions are used to guide users to perform certain specific behaviors (such as purchasing products and filling out forms). The number and frequency of these behaviors are also important measurement indicators. Feedback submission is used to measure whether users have provided any form of feedback, such as ratings, comments, or expressing their opinions in other ways. The unsubscription rate refers to tracking how many users choose not to receive similar messages. Time spent is used to measure the length of time users stay on pages related to the message.

[0049] Continuing with the above example, receipt information corresponding to the first message is received from the SMS channel, the email channel, and the instant messaging channel respectively.

[0050] Step S103: determining a first weight corresponding to each message sending channel in the at least one message sending channel based on the receipt information received from the at least one message sending channel.

[0051] The first weight is a weight related to the behavior of the user after receiving the first message through the message sending channel.

[0052] In one implementation, step S103 includes step S1031 - step S1032.

[0053] Step S1031, extracting features from the receipt information to obtain user behavior features corresponding to at least one message sending channel.

[0054] Among them, the user behavior characteristics include at least one of the user's clicking behavior, the user's sharing behavior, the user's reply behavior, the user's reading depth, the user's operation status, the user's viewing behavior and the user's stay time.

[0055] In one example, the method of obtaining user behavior characteristics is as follows: First, the behavioral data in the receipt information is preprocessed.

[0056] For example, the behavior data in the receipt information is checked, and duplicate records in the behavior data are deleted. Furthermore, invalid records that lack key fields in the behavior data are identified and removed. For example, one of the multiple records that record the number of times the user clicks on the first message can be retained, and the other records can be deleted. Then, the record of "reply behavior" that lacks a field can be deleted.

[0057] Secondly, the preprocessed behavior data is encoded to obtain user behavior characteristics.

[0058] For example, the pre-processed behavior data is coded in binary to indicate whether the behavior occurs. If the user shares the message, it is recorded as "1" in the corresponding field; if not, it is recorded as "0". For example, when the interactive behavior is coded as "1", it means that the user clicked on the first message, and when the time spent is coded as "1", it means that the user viewed the first message for more than the preset time.

[0059] Step S1032: using the user behavior characteristics corresponding to at least one message sending channel as the input of the pre-trained neural network model, and using the neural network model to output a first weight of at least one message sending channel.

[0060] Among them, the pre-trained neural network model refers to a neural network model that has been trained on a large-scale data set before use. In order to ensure the generalization ability of the neural network model, it is necessary to train the neural network model based on a large amount of training data.

[0061] In one embodiment, in combination Figure 3 This section explains how to train a neural network model.

[0062] Figure 3 It is a schematic diagram of a training method for a neural network model provided in an embodiment of the present application.

[0063] like Figure 3 As shown, the training method of the neural network model includes the following steps: step S321-step S323.

[0064] Step S321, obtaining a training set.

[0065] The training set includes training receipt information corresponding to at least one message sending channel and training weight values ​​corresponding to the training receipt information.

[0066] Exemplarily, taking three message sending channels as an example, including SMS channel, email channel and instant messaging channel, the SMS channel corresponds to the first receipt information and the first training weight value, the email channel corresponds to the second receipt information and the second training weight value, and the instant messaging channel corresponds to the third receipt information and the third training weight value.

[0067] Step S322: extract features from the training receipt information to obtain user behavior training features corresponding to at least one message sending channel.

[0068] The specific content of step S322 can refer to the above step S1031 and will not be repeated here.

[0069] Step S323: Train the neural network model by using the user behavior training features corresponding to at least one message sending channel as the input of the neural network model and by using at least one training weight value as the output of the neural network model.

[0070] In one example, combining Figure 3 As shown, the user behavior training features corresponding to at least one message sending channel are input into the initial neural network model, and the predicted weight value is output. Furthermore, the preset loss function is used to determine the loss value between the preset weight value and the training weight value, and the initial neural network model is continuously adjusted based on the loss value. When the loss value is less than the preset loss, it means that the training of the initial neural network model at this time is completed, and a trained neural network model is obtained.

[0071] After the neural network model is trained, the neural network model is deployed to a message sending system based on a message sending channel. Furthermore, during the application process, the user behavior training features corresponding to at least one message sending channel are input into the trained neural network model for processing, and a first weight of at least one message sending channel is output.

[0072] Continuing with the above example, the user behavior training features corresponding to the SMS channel are input into the trained neural network model for processing. Through the calculation and analysis of the neural network model, the weight value (first weight) of the SMS channel is obtained to be 0.3. The user behavior training features corresponding to the email channel are input into the trained neural network model for processing. Through the calculation and analysis of the neural network model, the weight value (first weight) of the email channel is obtained to be 0.18. The user behavior training features corresponding to the instant messaging channel are input into the trained neural network model for processing. Through the calculation and analysis of the neural network model, the weight value (first weight) of the instant messaging channel is obtained to be 0.45.

[0073] In another implementation, step S103 also includes steps S1033 and S1034.

[0074] Step S1033: extract features from the receipt information to obtain user behavior features corresponding to at least one message sending channel.

[0075] The specific content of step S1033 can refer to the above step S1031 and will not be repeated here.

[0076] Step S1034, summing the product of the user behavior feature and the preset weight parameter to obtain a first weight of at least one message sending channel.

[0077] In order to comprehensively evaluate user behavior data, not only a single user behavior feature is considered, but multiple behavior features are combined, and each user behavior feature corresponds to a preset weight parameter.

[0078] Exemplarily, the following formula (1) determines a first weight of at least one message sending channel.

[0079] Formula (1) in, is the first weight, , is the preset weight parameter. =0.6, =0.4.

[0080] It should be noted that formula (1) only exemplifies two user behavior characteristics and will not be listed one by one.

[0081] Continuing with the above example, the first behavior feature is user interaction (number of clicks by the user), and the second behavior feature is the user's time spent, that is, the length of time the user stays. The specific combination is shown in Table 1: Table 1 ; Continuing with Table 1, based on the number of user clicks and user stay time corresponding to each message sending channel, determine the first weight of the SMS channel =0.3, the first weight of the email channel =0.18, the first weight of instant messaging channel =0.45.

[0082] Step S104: determining a second weight corresponding to each message sending channel in the at least one message sending channel based on the receipt information received from the at least one message sending channel.

[0083] The second weight is a weight related to a message sending environment of a message sending channel.

[0084] In one implementation, step S104 includes steps S1041 to S1044.

[0085] Step S1041, obtaining the sending status of the first message within a preset cycle period.

[0086] The sending status includes sending success or sending failure. The preset cycle period can be set to 5 seconds.

[0087] Exemplarily, when the preset cycle period is 5 seconds, the sending status of the first message is obtained as successful sending.

[0088] Step S1042: When the sending status of the first message is successfully sent, record the first time point corresponding to the successful sending of the first message.

[0089] Continuing with the above example, if the first message is sent successfully at the 6th second, the first time point corresponding to the successful sending of the first message, that is, the 6th second, is recorded.

[0090] Step S1043, when the sending status of the first message is sending failure, record the second time point corresponding to the successful sending of the first message.

[0091] Continuing with the above example, if the first message fails to be sent at the 2nd second, the first time point corresponding to the failure to send the first message, ie, the 2nd second, is recorded.

[0092] Step S1044, based on the first time point, the second time point and the periodic nodes corresponding to the preset cycle period, obtain the second weight of at least one message sending channel; based on the second time point and the periodic nodes corresponding to the preset cycle period, obtain the second weight of at least one message sending channel.

[0093] In one example, the second weight is manually set based on the first time point, the second time point, and the period node corresponding to the preset cycle period.

[0094] The second weight can be To express.

[0095] Continuing with the above example, based on the first time point: 6 seconds, the second time point: 2 seconds, the period node corresponding to the preset cycle period: 5 seconds; or, the second time point: 2 seconds, the period node corresponding to the preset cycle period: 5 seconds, the second weight of at least one message sending channel is obtained to be 0.1, that is =0.1. Among them, is a weight related to the message sending environment of the message sending channel.

[0096] It should be understood that the order of step S103 and step S104 is not specifically limited.

[0097] Step S105: determining at least one first target message sending channel for the second message from at least one message sending channel based on the first weight and the second weight.

[0098] The first message and the second message are messages of the same type. For example, the first message and the second message are both shopping messages for promoting special offers, new product releases, etc.

[0099] Similarly, in order to comprehensively evaluate the comprehensive weight corresponding to the message sending channel, we cannot only consider the weight corresponding to a single influencing factor, but need to combine multiple influencing factors, each of which corresponds to a preset weight adjustment parameter.

[0100] In one implementation, step S105 includes steps S1051 to S1054.

[0101] Step S1051: Determine a comprehensive weight corresponding to at least one message sending channel based on the first weight and the second weight.

[0102] In one example, step S1051 includes steps S511 to S513.

[0103] Step S511, multiplying the first weight and the first weight coefficient to obtain a first product.

[0104] Step S512: multiply the second weight and the second weight coefficient to obtain a second product.

[0105] Step S513: sum the first product and the second product to determine a comprehensive weight of at least one message sending channel.

[0106] Exemplarily, step S1051 may determine the comprehensive weight of at least one message sending channel by formula (2).

[0107] Formula (2) in, is the comprehensive weight, , is the weight adjustment coefficient. For example, the weight adjustment coefficient It can be set to 0.3. Can be set to 0.4.

[0108] Continuing with the above example, based on formula (2), the comprehensive weight of the SMS channel is determined =0.33, the comprehensive weight of the email channel =0.54, the comprehensive weight of instant messaging channels =0.87.

[0109] Step S1052: sort at least one comprehensive weight, and store the sorted at least one comprehensive weight and the message sending channel corresponding to the comprehensive weight in a database.

[0110] For example, it is assumed that the database originally stores message sending channels and their corresponding comprehensive weights corresponding to shopping messages such as special offers and new product releases, such as a website pop-up window channel and its corresponding weight of 0.2.

[0111] As shown in Table 2, the database stores at least one sorted comprehensive weight and the message sending channel corresponding to the comprehensive weight.

[0112] Table 2 ; Step S1053, obtaining from the database all message sending channels and corresponding comprehensive weights corresponding to the corresponding second message.

[0113] Exemplarily, in combination with what is shown in Table 2 above, all message sending channels corresponding to the second message are obtained, which are instant messaging channel: 0.87, email channel: 0.54, SMS channel: 0.33, website pop-up window channel: 0.2.

[0114] Step S1054: determining at least one first target message sending channel among all message sending channels based on the weight threshold.

[0115] For example, if the weight threshold is set to 0.5.

[0116] In one example, if the threshold in the database is greater than or equal to the weight threshold, at least one first target message sending channel among all message sending channels is determined.

[0117] Continuing with the above example, if all the message sending channels corresponding to the second message include the above four first target message sending channels, namely, the email channel, the instant messaging channel, the SMS channel and the website pop-up window channel, the thresholds corresponding to the email channel and the instant messaging channel are both greater than the weight threshold, and the two first target message sending channels corresponding to the first message are determined, and the two first target message sending channels are the email channel and the instant messaging channel respectively.

[0118] Step S106: Send the second message through at least one first target message sending channel.

[0119] Following the above example, a second shopping message for promoting special offers, new product releases, etc. is sent through an email channel or an instant messaging channel.

[0120] In order to improve the arrival rate of message sending, after the second message is sent, it is also necessary to monitor the sending status of the second message. If the second message is not sent successfully, the second message is resent.

[0121] Step S107: receiving a sending status of a second message from at least one message sending channel.

[0122] The specific content of step S107 can refer to the specific content of the above step S101 and will not be repeated here.

[0123] Step S108, when the sending status of the second message is sending failure, obtaining at least one second target message sending channel from the database.

[0124] Among them, the weight corresponding to the second target message sending channel is less than the weight corresponding to the first target message sending channel.

[0125] Continuing with the example of step S1054 above, the weights corresponding to the first target message sending channel include the email channel and its corresponding weight of 0.54, and the instant messaging channel and its corresponding weight of 0.87. When trying to send the second message through the email channel and the instant messaging channel, if both fail to be successfully delivered, the second target message sending channel, i.e., the SMS channel, which is lower than the first target message sending channel, is obtained from the database.

[0126] Step S109: resend the second message through at least one second target message sending channel.

[0127] Exemplarily, the second shopping-related message for promoting special offers, new product releases, etc. is resent through the SMS channel.

[0128] In summary, based on the first weight and the second weight, a message sending channel can be dynamically selected for message sending, thereby improving the accuracy of message sending and enhancing user experience.

[0129] Since some devices do not support all message channels, you also need to consider the compatibility of the device when choosing a sending channel. Figure 4 As shown, the present application also provides a second embodiment corresponding to the first embodiment.

[0130] Figure 4 This is the third flow chart of a message sending method based on a message sending channel provided in an embodiment of the present application.

[0131] like Figure 2 and Figure 4 As shown, the message sending method based on the message sending channel provided in this embodiment includes the following steps: Step S401: Send a first message through at least one message sending channel.

[0132] Step S402: receiving receipt information corresponding to the first message from at least one message sending channel.

[0133] The receipt information includes information for monitoring the delivery status of the message, the user's behavior data after receiving the message, and the type of device that sent the message. Exemplarily, the device type may include smartphones, feature phones, tablets, desktop computers, and laptops.

[0134] Step S403: determining a first weight corresponding to each message sending channel in the at least one message sending channel based on the receipt information received from the at least one message sending channel.

[0135] Step S404: determining a second weight corresponding to each message sending channel in the at least one message sending channel based on the receipt information received from the at least one message sending channel.

[0136] The specific contents of step S401 to step S404 may refer to the above-mentioned step S101 to step S104 and will not be described in detail.

[0137] Step S405: determining a third weight corresponding to each message sending channel in the at least one message sending channel based on the receipt information received from the at least one message sending channel.

[0138] The third weight is a weight related to a message sending device of the message sending channel.

[0139] Step S406: Determine at least one first target message sending channel for the second message from at least one message sending channel based on the first weight, the second weight, and the third weight.

[0140] The specific content of step S406 can refer to the above-mentioned similar step S106 and will not be repeated here.

[0141] Step S407: Send the second message through at least one first target message sending channel.

[0142] Step S408: receiving a sending status of a second message from at least one message sending channel.

[0143] Step S409: when the sending status of the second message is sending failure, obtaining at least one second target message sending channel from the database.

[0144] Step S410: resend the second message through at least one second target message sending channel.

[0145] The specific contents of step S407 to step S410 may refer to the above-mentioned step S106 to step S109 and will not be described in detail.

[0146] In summary, based on the first weight, the second weight and the third weight, a message sending channel can be dynamically selected for message sending, thereby improving the accuracy of message sending and enhancing user experience.

[0147] Corresponding to the above-mentioned embodiment of the method for sending a message based on a message sending channel, Figure 5 As shown, the present application also provides an embodiment of a message sending system based on a message sending channel.

[0148] Figure 5 It is a schematic diagram of a message sending system based on a message sending channel provided in an embodiment of the present application.

[0149] Combination Figure 5 As shown, the message sending system based on the message sending channel includes: The multi-channel integration module is configured to: send the first message through at least one message sending channel; the multi-channel integration module is further configured to: receive receipt information corresponding to the first message from at least one message sending channel; The user preference analysis module is configured to: determine a first weight corresponding to each of the at least one message sending channel based on receipt information received from the at least one message sending channel; the first weight is a weight related to the behavior of the user after receiving the first message through the message sending channel; The network environment monitoring module is configured to: determine a second weight corresponding to each message sending channel in the at least one message sending channel based on receipt information received from the at least one message sending channel; the second weight is a weight related to the message sending environment of the message sending channel; The intelligent dynamic adjustment module is configured to: determine at least one first target message sending channel of the second message from at least one message sending channel based on the first weight and the second weight, the first message and the second message being messages of the same type; The multi-channel integration module is further configured to: send the second message through at least one first target message sending channel.

[0150] It should be noted that those skilled in the art will easily think of other embodiments of the present application after considering the specification and practicing the application disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application. The specification and examples are only regarded as exemplary, and the true scope of the present application is indicated by the claims.

[0151] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A message sending method based on a message sending channel, characterized in that: include: Sending a first message through at least one message sending channel; receiving receipt information corresponding to the first message from the at least one message sending channel; Determine, based on the receipt information received from the at least one message sending channel, a first weight and a second weight corresponding to each of the at least one message sending channel; The first weight is a weight related to the behavior of the user after receiving the first message through the message sending channel, and the second weight is a weight related to the message sending environment of the message sending channel; Based on the first weight and the second weight, determining at least one first target message sending channel for a second message from the at least one message sending channel, the first message and the second message being messages of the same type; The second message is sent through the at least one first target message sending channel.

2. The message sending method based on the message sending channel according to claim 1, characterized in that: The sending of the first message through at least one message sending channel includes: Obtaining the first message to be sent; According to the first message, all message sending channels corresponding to the first message are obtained from a database; wherein the database includes at least one of the message sending channels and a weight corresponding to the at least one of the message sending channels; Based on a weight threshold, determining at least one message sending channel among all the message sending channels corresponding to the first message; The first message is sent through the at least one message sending channel.

3. The message sending method based on the message sending channel according to claim 1, characterized in that: The determining, based on the receipt information received from the at least one message sending channel, a first weight corresponding to each of the at least one message sending channel comprises: Extracting features from the receipt information to obtain user behavior features corresponding to the at least one message sending channel; the user behavior features include at least one of a user's click behavior, a user's sharing behavior, a user's reply behavior, a user's reading depth, a user's operation status, a user's viewing behavior, and a user's stay time; The user behavior feature corresponding to the at least one message sending channel is used as an input of a pre-trained neural network model, and the first weight of the at least one message sending channel is output using the neural network model.

4. The message sending method based on the message sending channel according to claim 1, characterized in that: The determining, based on the receipt information received from the at least one message sending channel, a first weight corresponding to each of the at least one message sending channel comprises: Extracting features from the receipt information to obtain user behavior features corresponding to the at least one message sending channel; the user behavior features include at least one of a user's click behavior, a user's sharing behavior, a user's reply behavior, a user's reading depth, a user's operation status, a user's viewing behavior, and a user's stay time; The product of the user behavior feature and the preset weight parameter is summed to obtain the first weight of the at least one message sending channel.

5. The message sending method based on a message sending channel according to any one of claims 1 to 4, characterized in that: The determining, based on the receipt information received from the at least one message sending channel, a second weight corresponding to each of the at least one message sending channel comprises: Within a preset cycle period, obtaining a sending status of the first message, wherein the sending status includes a sending success or a sending failure; When the sending status of the first message is that the first message is sent successfully, recording a first time point corresponding to the successful sending of the first message; When the sending status of the first message is sending failure, recording a second time point corresponding to the failure to send the first message; Based on the first time point and the periodic node corresponding to the preset cycle period, the second weight of the at least one message sending channel is obtained; or, based on the second time point and the periodic node corresponding to the preset cycle period, the second weight of the at least one message sending channel is obtained.

6. The message sending method based on the message sending channel according to claim 1, characterized in that: The determining, based on the first weight and the second weight, at least one first target message sending channel of the second message from the at least one message sending channel comprises: Determining a comprehensive weight corresponding to the at least one message sending channel based on the first weight and the second weight; sorting at least one of the comprehensive weights, and storing the sorted at least one comprehensive weight and the message sending channel corresponding to the comprehensive weight in a database; Acquire from the database all message sending channels and the corresponding comprehensive weights corresponding to the second message; Based on the weight threshold and the comprehensive weight, the at least one first target message sending channel among all the message sending channels is determined.

7. The message sending method based on the message sending channel according to claim 6, characterized in that: The determining, based on the first weight and the second weight, a comprehensive weight corresponding to the at least one message sending channel includes: multiplying the first weight and a first weight coefficient to obtain a first product; multiplying the second weight and the second weight coefficient to obtain a second product; The first product and the second product are summed to determine the comprehensive weight of the at least one message sending channel.

8. The message sending method based on the message sending channel according to claim 1, characterized in that: After sending the second message through the at least one first target message sending channel, the method further includes: receiving a sending status of the second message from the at least one message sending channel; When the sending status of the second message is sending failure, obtaining at least one second target message sending channel from the database, wherein the weight corresponding to the second target message sending channel is less than the weight corresponding to the first target message sending channel; The second message is resent through the at least one second target message sending channel.

9. The message sending method based on the message sending channel according to claim 3, characterized in that: Before using the user behavior feature corresponding to the at least one message sending channel as an input of a pre-trained neural network model and using the neural network model to output the first weight of the at least one message sending channel, the method further includes: Acquire a training set, wherein the training set includes at least one training receipt information corresponding to the message sending channel and a training weight value corresponding to the training receipt information; Extracting features from the training receipt information to obtain user behavior training features corresponding to the at least one message sending channel; The neural network model is trained by using the user behavior training feature corresponding to at least one of the message sending channels as input to the neural network model and using at least one of the training weight values ​​as output of the neural network model.

10. The message sending method based on the message sending channel according to claim 1, characterized in that: The receipt information includes data used to monitor the sending status of the message and the user's behavior after receiving the message.

11. The message sending method based on the message sending channel according to claim 1, characterized in that: After determining the first weight and the second weight corresponding to each of the at least one message sending channel based on the receipt information received from the at least one message sending channel, the method further includes: Based on the receipt information, determine a third weight corresponding to each of the at least one message sending channel, where the third weight is a weight associated with a message sending device of the message sending channel; Determining at least one third target message sending channel based on the first weight, the second weight, and the third weight; The second message is sent through the at least one third target message sending channel.

12. A message sending system based on a message sending channel, characterized in that: include: The multi-channel integration module is configured to: send a first message through at least one message sending channel; The multi-channel integration module is further configured to: receive receipt information corresponding to the first message from the at least one message sending channel; A user preference analysis module is configured to: determine a first weight corresponding to each of the at least one message sending channel based on the receipt information received from the at least one message sending channel; The first weight is a weight related to the behavior of the user after receiving the first message through the message sending channel; The network environment monitoring module is configured to: determine a second weight corresponding to each of the at least one message sending channel based on the receipt information received from the at least one message sending channel; The second weight is a weight related to a message sending environment of the message sending channel; an intelligent dynamic adjustment module, configured to: determine at least one first target message sending channel of a second message from the at least one message sending channel based on the first weight and the second weight, the first message and the second message being messages of the same type; The multi-channel integration module is further configured to: send the second message through the at least one first target message sending channel.

Images