Method, device, terminal equipment and storage medium for sending a short message

Abstract

The application is suitable for the field of artificial intelligence technology, and provides a method and device for sending short messages, terminal equipment and storage medium. First, test set data of a target commodity is acquired, which contains a first test set and a second test set; then, the first test set is input into a probability prediction model for processing to obtain a predicted probability value of a historical user purchasing the target commodity, and each predicted probability value obtained is divided into each probability distribution interval; next, the average conversion rate of each probability distribution interval is calculated according to the predicted probability value contained in each probability distribution interval; then, the second test set is also input into the probability prediction model for processing to obtain a predicted probability value of a target user purchasing the target commodity, and based on the predicted probability value obtained and the average conversion rate of each probability distribution interval, a predicted conversion number of the target commodity is calculated; finally, whether to send a short message of the target commodity to the target user is determined according to the predicted conversion number.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and provides a method, apparatus, terminal device, and storage medium for sending text messages. Background Technology

[0002] SMS marketing involves both audience selection and product selection. One key concern is choosing the right products from a large pool for SMS marketing campaigns. Currently, the common practice is to conduct a small-scale test before sending out large-scale SMS messages, then estimate the profit and reach of each product based on the test results. However, sending test SMS messages is expensive, and executing test campaigns requires cooperation from multiple parties and involves lengthy communication times, resulting in high costs. Summary of the Invention

[0003] In view of this, this application proposes a method, apparatus, terminal device and storage medium for sending text messages, which can reduce the cost of pushing marketing messages to products via text message.

[0004] In a first aspect, embodiments of this application provide a method for sending text messages, including:

[0005] Acquire test set data for the target product. The test set data includes a first test set and a second test set. The first test set includes user characteristic information of each historical user and product characteristic information of the target product. The second test set includes user characteristic information of each target user and product characteristic information of the target product.

[0006] The first test set is input into the trained probability prediction model for processing to obtain the predicted probability value of each historical user purchasing the target product.

[0007] The predicted probability value of each historical user purchasing the target product is divided into various probability distribution intervals;

[0008] The average conversion rate of each probability distribution interval is calculated based on the predicted probability value contained in each probability distribution interval.

[0009] The second test set is input into the probability prediction model for processing to obtain the predicted probability value of each target user purchasing the target product;

[0010] The predicted number of conversions for the target product is calculated based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval.

[0011] Based on the predicted number of conversions, determine whether to send an SMS message containing the target product to each target user.

[0012] This embodiment first acquires test set data for the target product, including a first test set corresponding to historical users and a second test set corresponding to the target user. Then, the first test set is input into a trained probability prediction model for processing to obtain the predicted probability value for each historical user to purchase the target product, and each predicted probability value is divided into a preset probability distribution interval. Next, based on the predicted probability values ​​contained in each probability distribution interval, the average conversion rate for each probability distribution interval is calculated. Based on the principles of calculus, the average conversion rate calculated within a smaller probability distribution interval can be considered stable. Then, the second test set is also input into the probability prediction model for processing to obtain the predicted probability value for each target user to purchase the target product. Based on the obtained predicted probability values ​​and the average conversion rate of each probability distribution interval, the predicted number of conversions for the target product is calculated, i.e., how many target users are likely to purchase the target product. Finally, based on the calculated predicted number of conversions, it is determined whether to send an SMS message about the target product to the target users. This process obtains the predicted number of conversions for the product without performing any SMS sending test activities, thus reducing the cost of SMS marketing pushes for the product.

[0013] In one embodiment of this application, the respective probability distribution intervals can be set in the following manner:

[0014] Statistically analyze the probability distribution characteristics of the predicted probability value of each historical user purchasing the target product;

[0015] The number of probability distribution intervals and the probability range interval are set according to the probability distribution characteristics, so that the difference in the number of predicted probability values ​​contained in each probability distribution interval is less than a set threshold.

[0016] In one embodiment of this application, the respective probability distribution intervals can be set in the following manner:

[0017] Arrange the predicted probability values ​​of each historical user purchasing the target product in ascending or descending order;

[0018] The arranged predicted probability values ​​are divided into multiple probability combinations, and each probability combination contains a specified number of predicted probability values.

[0019] For each probability combination, its corresponding probability range is determined based on its minimum and maximum predicted probability values.

[0020] The probability range interval corresponding to each probability combination is determined as the respective probability distribution interval.

[0021] In one embodiment of this application, calculating the average conversion rate of each probability distribution interval based on the predicted probability value contained in each probability distribution interval may include:

[0022] For each probability distribution interval, the average value of all predicted probability values ​​contained in the probability distribution interval is calculated, and the average value is determined as the average conversion rate of the probability distribution interval.

[0023] In one embodiment of this application, calculating the predicted number of conversions for the target product based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval may include:

[0024] The predicted probability value of each target user purchasing the target product is divided into the respective probability distribution intervals;

[0025] Count the number of target users corresponding to each probability distribution interval;

[0026] Multiply the number of target users corresponding to each probability distribution interval by their respective average conversion rate to obtain the predicted number of conversions for each probability distribution interval.

[0027] The predicted conversion number corresponding to each probability distribution interval is added together to obtain the predicted conversion number of the target product.

[0028] In one embodiment of this application, determining whether to send an SMS message about the target product to each target user based on the predicted number of conversions may include:

[0029] Based on the predicted number of conversions and the unit profit of the target product, the estimated profit value of the target product is calculated.

[0030] If the estimated profit of the target product is greater than a set threshold, then it is determined to send a text message about the target product to each target user.

[0031] In one embodiment of this application, obtaining the test set data of the target product may include:

[0032] Retrieve historical information of sent product-related SMS messages;

[0033] Search the historical record information for the target product;

[0034] The first test set is extracted from the target record information.

[0035] Secondly, embodiments of this application provide an apparatus for sending text messages, comprising:

[0036] The test set data acquisition module is used to acquire test set data of the target product. The test set data includes a first test set and a second test set. The first test set includes user feature information of each historical user and product feature information of the target product. The second test set includes user feature information of each target user and product feature information of the target product.

[0037] The first probability prediction module is used to input the first test set into the trained probability prediction model for processing, and obtain the predicted probability value of each historical user purchasing the target product.

[0038] The probability partitioning module is used to divide the predicted probability value of each historical user purchasing the target product into various probability distribution intervals.

[0039] The conversion rate calculation module is used to calculate the average conversion rate of each probability distribution interval based on the predicted probability value contained in each probability distribution interval.

[0040] The second probability prediction module is used to input the second test set into the probability prediction model for processing, and obtain the predicted probability value of each target user purchasing the target product;

[0041] The conversion number calculation module is used to calculate the predicted conversion number of the target product based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval.

[0042] The SMS sending module is used to determine whether to send an SMS message about the target product to each target user based on the predicted number of conversions.

[0043] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for sending text messages as proposed in the first aspect of embodiments of this application.

[0044] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for sending SMS messages as proposed in the first aspect of embodiments of this application.

[0045] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the method for sending SMS messages as proposed in the first aspect of embodiments of this application.

[0046] The beneficial effects that can be achieved by the second to fifth aspects mentioned above can be referred to the relevant explanations of the first aspect mentioned above. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of one embodiment of a method for sending text messages provided in this application;

[0049] Figure 2 This is a schematic diagram illustrating the operational principle of a method for sending text messages provided in an embodiment of this application;

[0050] Figure 3 This is a structural diagram of one embodiment of a device for sending text messages provided in this application.

[0051] Figure 4 This is a schematic diagram of a terminal device provided in an embodiment of this application. Detailed Implementation

[0052] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail. Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0053] SMS redemption services involve two main aspects: selecting users and selecting products. One of the key concerns is how to choose suitable products from a large pool for targeted marketing. The current practice is to conduct a small-scale test campaign before each SMS campaign, using the results to estimate profits and the number of recipients. However, conducting tests before each product is costly, involving multiple teams and lengthy communication processes. Therefore, this application proposes a new SMS marketing method that reduces the cost of pushing products via SMS.

[0054] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0055] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0056] It should be understood that the execution entity of the SMS sending method provided in this application embodiment can be a mobile phone, tablet computer, wearable device, in-vehicle device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), large-screen TV, or other terminal device or server. This application embodiment does not impose any restrictions on the specific type of terminal device and server. The server here can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery network (CDN), and big data and artificial intelligence platforms.

[0057] Please see Figure 1 The first embodiment of a method for sending text messages in this application includes:

[0058] 101. Obtain test set data for the target product, wherein the test set data includes a first test set and a second test set;

[0059] First, obtain test set data for the target product. The target product is any product to be pushed to. Understandably, each product can be processed in the same way as the target product to calculate the corresponding predicted conversion number (i.e., how many target users who receive the product SMS will buy the product), thereby assessing the corresponding profit and deciding whether to send marketing SMS messages for that product to the target users.

[0060] The test set data includes a first test set and a second test set. The first test set includes user characteristic information for each historical user and product characteristic information for the target product. The second test set includes user characteristic information for each target user and product characteristic information for the target product. Specifically, historical users refer to users who have previously received SMS messages about the target product, and target users refer to users who are about to receive SMS messages about the product. The system calculates the predicted conversion rate for the target product to determine whether to send SMS messages about the target product to target users. User characteristic information may include user ID, various user profile information, and redemption intention codes (used to indicate whether the target product has been purchased), etc. Product characteristic information may include product ID and other various product attribute information.

[0061] In one implementation of this application, obtaining the test set data of the target product may include:

[0062] (1) Obtain the historical information of sent product-related SMS messages;

[0063] (2) Search for the target record information corresponding to the target product from the historical record information;

[0064] (3) Extract the first test set from the target record information.

[0065] When acquiring the first test set, you can query the historical records of the products in the official event to obtain the historical information corresponding to the target product (i.e., the target record information), and then extract various feature information such as SMS sending date, product ID, user ID, and redemption intention code from it.

[0066] 102. Input the first test set into the trained probability prediction model for processing to obtain the predicted probability value of each historical user purchasing the target product;

[0067] Then, the first test set is input into a trained probability prediction model for processing to obtain the predicted probability value of each historical user purchasing the target product. In practice, a historical dataset of product SMS messages can be obtained and used as a training set to train a neural network model as the probability prediction model. This probability prediction model can employ various machine learning models, such as the XGBoost model, to process the input user feature information and product feature information, predicting the probability value of a user corresponding to the user feature information purchasing the product corresponding to the product feature information. That is, after inputting the first test set into the probability prediction model, it can output the predicted probability value of each historical user purchasing the target product. For a detailed explanation of this probability prediction model, refer to various existing machine learning models that predict the probability of a user purchasing a product based on user features and product features.

[0068] 103. Divide the predicted probability value of each historical user purchasing the target product into various probability distribution intervals;

[0069] After obtaining the predicted probability value of each historical user purchasing the target product, these predicted probability values ​​are divided into various preset probability distribution intervals. For example, five probability distribution intervals can be set: 0%-20%, 21%-40%, 41%-60%, 61%-80%, and 81%-100%. Then, each predicted probability value is assigned to a corresponding probability distribution interval based on its magnitude. For instance, if a historical user's predicted probability value is 58%, it would be assigned to the probability distribution interval 40%-59%, and so on.

[0070] Based on the principles of calculus, the average conversion rate calculated within a small probability distribution interval can be considered stable. Therefore, the number of probability distribution intervals can be set relatively large, resulting in a shorter range for each interval. For example, 50 probability distribution intervals can be created, with the first ranging from 0% to 2.0%, the second from 2.1% to 4.0%, and so on. Furthermore, the ranges of the various probability distribution intervals can be the same or different.

[0071] In one implementation of this application, the probability distribution intervals can be set in the following way:

[0072] (1) Statistically analyze the probability distribution characteristics of the predicted probability value of each historical user purchasing the target product;

[0073] (2) Set the number of each probability distribution interval and the probability range interval according to the probability distribution characteristics, so that the difference in the number of predicted probability values ​​contained in each probability distribution interval is less than a set threshold.

[0074] If a probability distribution interval contains too few predicted probability values, then setting that interval is largely meaningless. Therefore, to improve the accuracy of calculating the average conversion rate for each probability distribution interval, the basic principle is to ensure that the number of predicted probability values ​​contained in each interval is roughly balanced; that is, the difference in the number of predicted probability values ​​contained in any two intervals should be small. Based on this principle, we can first statistically analyze the probability distribution characteristics of predicted probability values ​​for each historical user purchasing the target product, and then set the corresponding probability distribution intervals according to these characteristics and the principle. Specifically, we can set the probability range for each interval. For example, if, based on the statistically obtained probability distribution characteristics, approximately 50% of the predicted probability values ​​fall within the 10%-30% interval, and another approximately 50% fall within the 50%-90% interval, then we can set two probability distribution intervals: 10%-30% and 50%-90%, and so on.

[0075] In another implementation of this application embodiment, the probability distribution intervals can be set in the following way:

[0076] (1) Arrange the predicted probability values ​​of each historical user purchasing the target product in ascending or descending order;

[0077] (2) Divide the arranged predicted probability values ​​into multiple probability combinations, and each probability combination contains a specified number of predicted probability values;

[0078] (3) For each probability combination, determine its corresponding probability range based on the minimum and maximum predicted probability values ​​it contains;

[0079] (4) The probability range interval corresponding to each probability combination is determined as the respective probability distribution interval.

[0080] The basic principle of this implementation is still to ensure that the number of predicted probability values ​​contained in each probability distribution interval is roughly balanced. Assume there are 100 predicted probability values ​​arranged in ascending order, denoted as P1-P2. 100 Let each probability combination contain 10 predicted probability values, then the 1st to 10th predicted probability values ​​P1-P 10 Divided into the first probability combination, the 11th-20th predicted probability values ​​P 11 -P 20 This is divided into a second probability combination… and so on, resulting in a total of 10 probability combinations. Then, for each probability combination, the corresponding probability range is determined based on its maximum and minimum predicted probability values. For example, the first probability combination contains a minimum predicted probability value P1 of 3%, and a maximum predicted probability value P…10 If the probability is 8%, then a probability range containing both the minimum and maximum predicted probability values ​​can be determined, such as 2%-10%, as the probability range for the first probability combination. Similarly, each probability combination can obtain a corresponding probability range. These probability ranges can be considered as the obtained probability distribution intervals, and each probability distribution interval contains the same number of predicted probability values, for example, 10.

[0081] 104. Calculate the average conversion rate of each probability distribution interval based on the predicted probability value contained in each probability distribution interval;

[0082] After dividing the predicted probability value of each historical user purchasing the target product into various probability distribution intervals, the average conversion rate of each probability distribution interval is calculated. Specifically, it can be calculated based on the average, maximum, or minimum value of the predicted probability values ​​contained in each probability distribution interval.

[0083] In one implementation of this application, calculating the average conversion rate of each probability distribution interval based on the predicted probability value contained in each probability distribution interval may include:

[0084] For each probability distribution interval, the average value of all predicted probability values ​​contained in the probability distribution interval is calculated, and the average value is determined as the average conversion rate of the probability distribution interval.

[0085] For example, for a probability distribution range of 61%-80%, if it contains four predicted probability values ​​for historical users, namely 65%, 70%, 70%, and 75%, then the average of these four predicted probability values, i.e., 70%, can be calculated and determined as the average conversion rate for that probability distribution range.

[0086] 105. Input the second test set into the probability prediction model for processing to obtain the predicted probability value of each target user purchasing the target product;

[0087] The second test set obtained in step 101 is also input into the probability prediction model in step 102 for processing to obtain the predicted probability value of each target user purchasing the target product.

[0088] 106. Calculate the predicted number of conversions for the target product based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval;

[0089] Next, based on the predicted probability of each target user purchasing the target product and the average conversion rate for each probability distribution interval, the predicted number of conversions for the target product is calculated, which is the number of users among these target users who may purchase the target product after the SMS message for the target product is sent to them.

[0090] In one implementation of this application, calculating the predicted number of conversions for the target product based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval may include:

[0091] (1) Divide the predicted probability value of each target user purchasing the target product into the respective probability distribution intervals;

[0092] (2) Count the number of target users corresponding to each probability distribution interval respectively;

[0093] (3) Multiply the number of target users corresponding to each probability distribution interval by their respective average conversion rate to obtain the predicted number of conversions corresponding to each probability distribution interval;

[0094] (4) Add up the predicted conversion number corresponding to each probability distribution interval to obtain the predicted conversion number of the target product.

[0095] For example, assuming there are 100 target users, and the probability distribution intervals are divided into 5 ranges: 0%-19%, 20%-39%, 40%-59%, 60%-79%, and 80%-100%. The 0%-19% interval corresponds to 17 target users (meaning the predicted probability of these 17 users purchasing the target product falls within this interval), and the 20%-39% interval corresponds to 23 target users (the number of target users in each interval can be considered as the predicted probability of those within that interval purchasing the target product). The target user count is as follows: 40%-59% corresponds to 30 users, 60%-79% to 20 users, and 80%-100% to 10 users. The average conversion rate is 15% for the 0%-19% range, 30% for the 20%-39% range, 50% for the 40%-59% range, 68% for the 60%-79% range, and 90% for the 80%-100% range. Therefore, for the 0%-19% range, the predicted number of conversions is 17 * 15% = 2.55 people; for the 20%-39% range, the predicted number of conversions is 23 * 30% = 6.9 people; for the 40%-59% range, the predicted number of conversions is 30 * 50% = 15 people; for the 60%-79% range, the predicted number of conversions is 20 * 68% = 13.6 people; and for the 80%-100% range, the predicted number of conversions is 10 * 90% = 9 people. Finally, adding the predicted number of conversions for each probability distribution range, we get 2.55 + 6.9 + 15 + 13.6 + 9 = 47.05 people (this result can be rounded down), which is taken as the predicted number of conversions for this target product.

[0096] 107. Based on the predicted number of conversions, determine whether to send a text message about the target product to each target user.

[0097] After calculating the predicted number of conversions for the target product, the system can determine whether to send an SMS message about the target product to each target user based on this predicted number. For example, if the predicted conversion number is high, it means that sending an SMS message about the target product to the target user may generate higher profits, and the system will send the SMS message about the target product to the target user. Otherwise, it means that sending an SMS message about the target product to the target user may generate lower profits or even result in a loss, so the system will not send an SMS message about the target product to the target user to reduce SMS sending costs.

[0098] In one implementation of this application, determining whether to send an SMS message about the target product to each target user based on the predicted number of conversions may include:

[0099] (1) Calculate the estimated profit value of the target product based on the predicted number of conversions and the unit profit of the target product;

[0100] (2) If the estimated profit of the target product is greater than the set threshold, then determine to send the SMS message of the target product to each target user.

[0101] The profit per unit of a target product is the profit that can be obtained from selling a single target product. Multiplying the profit per unit by the predicted number of conversions yields the estimated profit value of the target product. If the estimated profit value is greater than a certain set threshold, it means that the return on SMS sending meets expectations, and at this point, it can be determined to send SMS messages about the target product to the target users. In practice, the estimated profit values ​​for a large number of products to be pushed can be determined in the same way as for the target product. Then, a certain number of products with the highest estimated profit values ​​can be selected for push, i.e., sending corresponding product SMS messages to each target user.

[0102] For example, assuming the predicted conversion rate for a target product is 500 people and the profit per unit sold is 100 yuan, the estimated profit for that target product is 500 * 100 = 50,000 yuan. If this estimated profit is among the top 10 of all product estimated profit values, a text message about the target product will be sent to the target users; otherwise, no text message about the target product will be sent.

[0103] This embodiment first acquires test set data for the target product, including a first test set corresponding to historical users and a second test set corresponding to the target user. Then, the first test set is input into a trained probability prediction model for processing to obtain the predicted probability value for each historical user to purchase the target product, and each predicted probability value is divided into a preset probability distribution interval. Next, based on the predicted probability values ​​contained in each probability distribution interval, the average conversion rate for each probability distribution interval is calculated. Based on the principles of calculus, the average conversion rate calculated within a smaller probability distribution interval can be considered stable. Then, the second test set is also input into the probability prediction model for processing to obtain the predicted probability value for each target user to purchase the target product. Based on the obtained predicted probability values ​​and the average conversion rate of each probability distribution interval, the predicted number of conversions for the target product is calculated, i.e., how many target users are likely to purchase the target product. Finally, based on the calculated predicted number of conversions, it is determined whether to send an SMS message about the target product to the target users. This process obtains the predicted number of conversions for the product without performing any SMS sending test activities, thus reducing the cost of SMS marketing pushes for the product.

[0104] like Figure 2The diagram shown illustrates the operational principle of the SMS sending method proposed in this application. Figure 2 First, test set data for the target product is acquired. This test set consists of two parts: a first test set corresponding to historical users and a second test set corresponding to target users. Then, a trained probability prediction model is used to predict the probabilities of historical users and the target user purchasing the target product, respectively. Next, the predicted probabilities of historical users purchasing the target product are divided into N preset probability distribution intervals, and the average conversion rate for each interval is calculated. Finally, based on the predicted probability of the target user purchasing the target product and the average conversion rate for each interval, the predicted number of users who will purchase the target product is calculated. Based on this predicted number, a decision is made regarding whether to send an SMS message about the target product to the target user.

[0105] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0106] Corresponding to the method of sending text messages described in the above embodiments, Figure 3 This diagram illustrates a structural block diagram of an apparatus for sending text messages according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiments of this application are shown.

[0107] Reference Figure 3 The device includes:

[0108] The test set data acquisition module 301 is used to acquire test set data of the target product. The test set data includes a first test set and a second test set. The first test set includes user feature information of each historical user and product feature information of the target product. The second test set includes user feature information of each target user and product feature information of the target product.

[0109] The first probability prediction module 302 is used to input the first test set into the trained probability prediction model for processing, and obtain the predicted probability value of each historical user purchasing the target product.

[0110] The probability partitioning module 303 is used to partition the predicted probability value of each historical user purchasing the target product into various probability distribution intervals.

[0111] The conversion rate calculation module 304 is used to calculate the average conversion rate of each probability distribution interval based on the predicted probability value contained in each probability distribution interval.

[0112] The second probability prediction module 305 is used to input the second test set into the probability prediction model for processing, and obtain the predicted probability value of each target user purchasing the target product.

[0113] The conversion number calculation module 306 is used to calculate the predicted conversion number of the target product based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval.

[0114] The SMS sending module 307 is used to determine whether to send an SMS message about the target product to each target user based on the predicted number of conversions.

[0115] In one embodiment of this application, the device for sending text messages may further include:

[0116] The distribution feature statistics module is used to statistically analyze the probability distribution characteristics of the predicted probability value of each historical user purchasing the target product.

[0117] The first distribution interval setting module is used to set the number of each probability distribution interval and the probability range interval according to the probability distribution characteristics, so that the difference in the number of predicted probability values ​​contained in each probability distribution interval is less than a set threshold.

[0118] In one embodiment of this application, the device for sending text messages may further include:

[0119] The probability sorting module is used to sort the predicted probability values ​​of each historical user purchasing the target product in ascending or descending order;

[0120] The probability combination module is used to divide the permuted predicted probability values ​​into multiple probability combinations, and each probability combination contains a specified number of predicted probability values.

[0121] The probability range determination module is used to determine the corresponding probability range interval for each probability combination based on its minimum and maximum predicted probability values.

[0122] The second distribution interval setting module is used to determine the probability range interval corresponding to each probability combination as the respective probability distribution interval.

[0123] In one embodiment of this application, the conversion rate calculation module may include:

[0124] The average conversion rate calculation unit is used to calculate the average value of all predicted probability values ​​contained in each probability distribution interval, and to determine the average value as the average conversion rate of the probability distribution interval.

[0125] In one embodiment of this application, the conversion number calculation module may include:

[0126] A probability partitioning unit is used to divide the predicted probability value of each target user purchasing the target product into the respective probability distribution intervals;

[0127] The user count unit is used to count the number of target users corresponding to each probability distribution interval.

[0128] The first conversion number calculation unit is used to multiply the number of target users corresponding to each probability distribution interval by their respective average conversion rate to obtain the predicted conversion number corresponding to each probability distribution interval.

[0129] The second conversion number calculation unit is used to add up the predicted conversion number corresponding to each of the probability distribution intervals to obtain the predicted conversion number of the target product.

[0130] In one embodiment of this application, the SMS sending module may include:

[0131] The profit estimation unit is used to calculate the estimated profit value of the target product based on the predicted number of conversions and the unit profit of the target product.

[0132] The SMS sending unit is used to determine to send an SMS message about the target product to each target user if the estimated profit value of the target product is greater than a set threshold.

[0133] In one embodiment of this application, the test set data acquisition module may include:

[0134] The history acquisition unit is used to acquire the history information of sent product-related SMS messages;

[0135] The target record lookup unit is used to search for target record information corresponding to the target product from the historical record information.

[0136] The test data extraction unit is used to extract the first test set from the target record information.

[0137] This application embodiment also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement... Figure 1 This represents any method of sending a text message.

[0138] This application also provides a computer program product that, when run on a server, causes the server to execute the implementation as described above. Figure 1 This represents any method of sending a text message.

[0139] Figure 4 This is a schematic diagram of a terminal device provided in an embodiment of this application. For example... Figure 4 As shown, the terminal device 4 in this embodiment includes: a processor 40, a memory 41, and computer-readable instructions 42 stored in the memory 41 and executable on the processor 40. When the processor 40 executes the computer-readable instructions 42, it implements the steps in the various methods for sending text messages described above, for example... Figure 1 Steps 101 to 107 are shown. Alternatively, when the processor 40 executes the computer-readable instructions 42, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 3 The functions of modules 301 to 307 are shown.

[0140] For example, the computer-readable instructions 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to complete this application. The one or more modules / units may be a series of computer-readable instruction segments capable of performing a specific function, which describe the execution process of the computer-readable instructions 42 in the terminal device 4.

[0141] The terminal device 4 may be a computing device such as a smartphone, laptop, PDA, or cloud terminal device. The terminal device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of terminal device 4 and does not constitute a limitation on terminal device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, terminal device 4 may also include input / output devices, network access devices, buses, etc.

[0142] The processor 40 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0143] The memory 41 can be an internal storage unit of the terminal device 4, such as a hard disk or memory of the terminal device 4. The memory 41 can also be an external storage device of the terminal device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the terminal device 4. Furthermore, the memory 41 can include both internal and external storage units of the terminal device 4. The memory 41 is used to store the computer-readable instructions and other programs and data required by the terminal device. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0144] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0145] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0146] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0147] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0148] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for sending text messages, characterized in that, include: Acquire test set data for the target product. The test set data includes a first test set and a second test set. The first test set includes user characteristic information of each historical user and product characteristic information of the target product. The second test set includes user characteristic information of each target user and product characteristic information of the target product. The first test set is input into the trained probability prediction model for processing to obtain the predicted probability value of each historical user purchasing the target product. The predicted probability value of each historical user purchasing the target product is divided into various probability distribution intervals; The average conversion rate of each probability distribution interval is calculated based on the predicted probability value contained in each probability distribution interval. The second test set is input into the probability prediction model for processing to obtain the predicted probability value of each target user purchasing the target product; The predicted number of conversions for the target product is calculated based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval. Based on the predicted number of conversions, determine whether to send an SMS message containing the target product to each target user; The step of calculating the predicted number of conversions for the target product based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval includes: The predicted probability value of each target user purchasing the target product is divided into the respective probability distribution intervals; Count the number of target users corresponding to each probability distribution interval; Multiply the number of target users corresponding to each probability distribution interval by their respective average conversion rate to obtain the predicted number of conversions for each probability distribution interval. The predicted conversion number corresponding to each probability distribution interval is added together to obtain the predicted conversion number of the target product.

2. The method as described in claim 1, characterized in that, The probability distribution intervals are set in the following manner: Statistically analyze the probability distribution characteristics of the predicted probability value of each historical user purchasing the target product; The number of probability distribution intervals and the probability range interval are set according to the probability distribution characteristics, so that the difference in the number of predicted probability values ​​contained in each probability distribution interval is less than a set threshold.

3. The method as described in claim 1, characterized in that, The probability distribution intervals are set in the following manner: Arrange the predicted probability values ​​of each historical user purchasing the target product in ascending or descending order; The arranged predicted probability values ​​are divided into multiple probability combinations, and each probability combination contains a specified number of predicted probability values. For each probability combination, its corresponding probability range is determined based on its minimum and maximum predicted probability values. The probability range interval corresponding to each probability combination is determined as the respective probability distribution interval.

4. The method as described in claim 1, characterized in that, The step of calculating the average conversion rate for each probability distribution interval based on the predicted probability value contained in each probability distribution interval includes: For each probability distribution interval, the average value of all predicted probability values ​​contained in the probability distribution interval is calculated, and the average value is determined as the average conversion rate of the probability distribution interval.

5. The method as described in claim 1, characterized in that, The step of determining whether to send an SMS message containing the target product to each target user based on the predicted number of conversions includes: Based on the predicted number of conversions and the unit profit of the target product, the estimated profit value of the target product is calculated. If the estimated profit of the target product is greater than a set threshold, then it is determined to send a text message about the target product to each target user.

6. The method according to any one of claims 1 to 5, characterized in that, The acquisition of test set data for the target product includes: Retrieve historical information of sent product-related SMS messages; Search the historical record information for the target product; The first test set is extracted from the target record information.

7. A device for sending text messages, characterized in that, include: The test set data acquisition module is used to acquire test set data of the target product. The test set data includes a first test set and a second test set. The first test set includes user feature information of each historical user and product feature information of the target product. The second test set includes user feature information of each target user and product feature information of the target product. The first probability prediction module is used to input the first test set into the trained probability prediction model for processing, and obtain the predicted probability value of each historical user purchasing the target product. The probability partitioning module is used to divide the predicted probability value of each historical user purchasing the target product into various probability distribution intervals. The conversion rate calculation module is used to calculate the average conversion rate of each probability distribution interval based on the predicted probability value contained in each probability distribution interval. The second probability prediction module is used to input the second test set into the probability prediction model for processing, and obtain the predicted probability value of each target user purchasing the target product; The conversion number calculation module is used to calculate the predicted conversion number of the target product based on the predicted probability value of each target user purchasing the target product and the average conversion rate of each probability distribution interval. The SMS sending module is used to determine whether to send an SMS message about the target product to each target user based on the predicted number of conversions. The conversion rate calculation module includes: A probability partitioning unit is used to divide the predicted probability value of each target user purchasing the target product into the respective probability distribution intervals; The user count unit is used to count the number of target users corresponding to each probability distribution interval. The first conversion number calculation unit is used to multiply the number of target users corresponding to each probability distribution interval by their respective average conversion rate to obtain the predicted conversion number corresponding to each probability distribution interval. The second conversion number calculation unit is used to add up the predicted conversion number corresponding to each of the probability distribution intervals to obtain the predicted conversion number of the target product.

8. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for sending SMS messages as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for sending SMS messages as described in any one of claims 1 to 6.

Images