5G message complaint prediction method

Abstract

The invention discloses a 5G message complaint prediction method. The method comprises the steps of obtaining multi-dimensional feature information of a 5G message port in historical time and the 5G message issuing amount of the 5G message port; according to the multi-dimensional feature information, determining a multi-input ratio of the 5G message port in the historical time; performing prediction processing on the multi-input ratio of the 5G message port in the historical time through a trained prediction model to obtain a predicted multi-input ratio of the 5G message in a target time; and determining the predicted complaint amount of the 5G message port according to the predicted multi-thousand input ratio and the 5G message issuing amount.

Description

Technical Field

[0001] The present application belongs to the field of computer technology and specifically relates to a 5G message complaint prediction method. Background Art

[0002] With the vigorous development of 5G and mobile communication technology, the number of operators' users has also increased rapidly, and the business volume of 5G messages has also increased significantly. However, as the business volume of 5G messages has increased, the number of complaints about 5G messages has also increased. The current technologies for managing 5G messages are mainly concentrated in the mid-stage (during the 5G message sending process) or the post-stage (after the 5G message is sent), such as keyword filtering, information classification, and legal penalties, but there are fewer preventive measures in the pre-stage, lack of forward-looking and predictive methods, and the governance of 5G messages is not complete, which has affected the governance effect of 5G messages to a certain extent. Summary of the invention

[0003] The embodiment of the present application provides a 5G message complaint prediction method, which can predict 5G message complaints in advance, manage 5G messages in advance, and reduce the complaint rate of 5G messages.

[0004] In the first aspect, an embodiment of the present application provides a 5G message complaint prediction method, the method comprising: obtaining multi-dimensional feature information of a 5G message port at a historical time and the amount of 5G messages sent by the 5G message port; determining the complaint ratio of the 5G message port at the historical time based on the multi-dimensional feature information; performing prediction processing on the complaint ratio of the 5G message port at the historical time through a trained prediction model to obtain a predicted complaint ratio of the 5G message at the target time; determining the predicted complaint volume of the 5G message port based on the predicted complaint ratio and the amount of 5G messages sent.

[0005] In the second aspect, an embodiment of the present application provides a 5G message complaint prediction system, which includes: an acquisition module, used to obtain multi-dimensional feature information of a 5G message port at a historical time and the amount of 5G messages sent by the 5G message port; a determination module, used to determine the complaint ratio of the 5G message port at the historical time based on the multi-dimensional feature information; a first prediction module, used to predict the complaint ratio of the 5G message port at the historical time through a trained prediction model to obtain a predicted complaint ratio of the 5G message at the target time; a second prediction module, used to determine the predicted complaint volume of the 5G message port based on the predicted complaint ratio and the amount of 5G messages sent.

[0006] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0007] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0008] In a fifth aspect, an embodiment of the present application provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the steps of the method described in the first aspect.

[0009] In a sixth aspect, an embodiment of the present application provides a computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises a program or instructions, and when the program or instructions are executed, the steps of the method described in the first aspect are implemented.

[0010] In the embodiment of the present application, by obtaining the multi-dimensional feature information of the 5G message port at the historical time and the 5G message delivery volume of the 5G message port, the 10,000-vote ratio of the 5G message port at the historical time is determined according to the multi-dimensional feature information, and then the 10,000-vote ratio of the 5G message port at the historical time is predicted and processed by the trained prediction model to obtain the predicted 10,000-vote ratio of the 5G message at the target time, thereby determining the predicted complaint volume of the 5G message port according to the predicted 10,000-vote ratio and the 5G message delivery volume. The 5G message complaint situation is predicted in advance, so that the 5G message can be managed in advance and the complaint rate of 5G messages can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flowchart of a 5G message complaint prediction method provided in an embodiment of the present application; Figure 2 It is a flowchart of a training method of a prediction model provided in an embodiment of the present application; Figure 3 It is a structural diagram of a 5G message complaint prediction system provided in an embodiment of the present application; Figure 4 It is a structural diagram of another 5G message complaint prediction system provided in an embodiment of the present application; Figure 5 It is a flowchart of another 5G message complaint prediction method provided in an embodiment of the present application; Figure 6 It is a flowchart of a method for processing multi-dimensional feature information provided by an embodiment of the present application; Figure 7 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0013] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0014] The 5G message complaint prediction method provided in the embodiment of the present application is described in detail below through specific embodiments and application scenarios in combination with the accompanying drawings.

[0015] Figure 1 A flow chart of a 5G message complaint prediction method provided by an embodiment of the present application is shown, and the method can be executed by an electronic device. Figure 1 , the method may include the following steps.

[0016] Step 102, obtaining multi-dimensional feature information of the 5G message port in historical time and the 5G message sending volume of the 5G message port.

[0017] Among them, the multi-dimensional feature information of the 5G message port in historical time includes 5G message content, port signature of the 5G message port, port behavior, port historical complaint information, agent historical complaint information, user complaint information and other dimensions.

[0018] Step 104: Determine the investment ratio of the 5G message port in the historical time according to the multi-dimensional feature information.

[0019] The 10,000-complaint ratio refers to the number of complaints per 10,000 paying users, which is used to reflect service quality and user satisfaction. The 10,000-complaint ratio of the 5G message port at the historical time is used to reflect the number of user complaints about 5G messages sent through the 5G message port.

[0020] Step 106, predicting the 10,000-th ratio of the 5G message port at the historical time through the trained prediction model to obtain the predicted 10,000-th ratio of the 5G message at the target time.

[0021] The prediction model may include a regression model, which uses the acquired time series data, that is, the investment ratio at historical time, to predict the predicted investment ratio at a target time through a trained prediction model, and the target time is a certain time in the future.

[0022] Step 108: Determine the predicted complaint volume of the 5G message port based on the predicted 10,000-vote ratio and the 5G message delivery volume.

[0023] In the embodiment of the present application, by obtaining the multi-dimensional feature information of the 5G message port at the historical time and the 5G message delivery volume of the 5G message port, the 10,000-vote ratio of the 5G message port at the historical time is determined according to the multi-dimensional feature information, and then the 10,000-vote ratio of the 5G message port at the historical time is predicted and processed by the trained prediction model to obtain the predicted 10,000-vote ratio of the 5G message at the target time, thereby determining the predicted complaint volume of the 5G message port according to the predicted 10,000-vote ratio and the 5G message delivery volume. The 5G message complaint situation is predicted in advance, so that the 5G message can be managed in advance and the complaint rate of 5G messages can be reduced.

[0024] In one implementation, the above-mentioned step 104 determines the investment ratio of the 5G message port in the historical time according to the multi-dimensional feature information, including the following steps.

[0025] Step 1041, determining the ten thousand investment ratio of each dimension respectively according to the multi-dimensional feature information.

[0026] Among them, for the 5G message content dimension, the Chinese content of each 5G message is deduplicated, and the deduplicated content is labeled and classified through a text classification algorithm. For example, it can be divided into multiple label categories such as e-commerce, education, and finance. The issuance volume and complaint volume of each label category are counted separately to determine the 10,000-vote ratio of the 5G message content dimension.

[0027] For the port signature dimension of the 5G message port, keywords of the 5G message content sent through the 5G message port are extracted, and signatures are classified according to the keywords, for example, signature categories including A bank, B platform, etc. The issuance volume and complaint volume of each signature category are counted separately, and the ten-thousand-vote ratio of the port signature dimension of the 5G message port is determined.

[0028] For the port behavior dimension of the 5G message port, obtain the abnormal behavior of the port, such as scattered, random, continuous sending, etc., count the sending volume and complaint volume of each behavior respectively, and determine the ten thousand vote ratio of the port behavior dimension of the 5G message port.

[0029] For the port historical complaint information dimension of the 5G message port, the complaint volume and the sent volume of the port historical complaint information are counted to determine the ten thousand vote ratio of the port historical complaint information dimension of the 5G message port.

[0030] For the dimension of agents' historical complaint information, the number of complaints and the number of complaints sent out for each agent's historical complaint information are counted separately to determine the ratio of 10,000 complaints to agents' historical complaint information.

[0031] For the user complaint information dimension, the number of complaints and the number of successful mass messaging for each user complaint information are counted to determine the 10,000-to-1 ratio of user complaint information.

[0032] Step 1042: Determine the ratio of the 5G message port to the historical time according to the ratio of each dimension.

[0033] In the embodiment of the present application, the acquired multi-dimensional feature information is analyzed to determine the ten thousand votes ratio of each dimension respectively, and then the ten thousand votes ratio of each dimension is counted and integrated to determine the ten thousand votes ratio of the 5G message port at the historical time. The ten thousand votes ratio of the 5G message port comprehensively considers the ten thousand votes ratio corresponding to the feature information of multiple dimensions, thereby improving the accuracy of the prediction model and determining the ten thousand votes ratio of the 5G message port more comprehensively and accurately.

[0034] In one implementation, the prediction model includes an elastic network model. In the above step 106, the 5G message ratio at the historical time is predicted by the trained prediction model to obtain the predicted ratio of the 5G message port at the target time. The method also includes the following steps.

[0035] Step 1051, obtaining sample multi-dimensional feature information of the 5G message port at the first time and the sample 5G message delivery volume of the 5G message port; Step 1052: determining the sample-to-sample ratio of the 5G message at the first time according to the multi-dimensional feature information of the sample; Step 1053, predicting the sample-to-vote ratio through the elastic network model to obtain the sample-to-vote ratio predicted for the 5G message at the second time; Step 1054, determining the sample predicted complaint volume of the 5G message port according to the sample predicted 10,000-vote ratio and the sample 5G message delivery volume; Step 1055, determining the loss function of the elastic network model according to the sample predicted 10,000 investment ratio, the sample predicted complaint volume, and the parameter vector of the elastic network model; Step 1056, updating the parameters of the elastic network model according to the loss function until the loss function converges.

[0036] In the embodiment of the present application, a method for training a prediction model is provided, see Figure 2 , obtain sample data for training the prediction model, including sample multi-dimensional feature information of the 5G message port at the first time and the sample 5G message delivery volume of the 5G message port, determine the sample 10,000-vote ratio of the 5G message at the first time by analyzing the sample multi-dimensional feature information, and then predict and process the sample 10,000-vote ratio through the elastic network model to obtain the sample predicted 10,000-vote ratio of the 5G message at the second time, thereby determining the sample predicted complaint volume of the 5G message port according to the sample predicted 10,000-vote ratio and the sample 5G message delivery volume, and then update the parameters of the elastic network model based on the loss function until the loss function converges, and the elastic network model training is completed, so as to predict the predicted 10,000-vote ratio of the 5G message port at the target time, thereby predicting the predicted complaint volume of the 5G message port according to the predicted 10,000-vote ratio and the 5G message delivery volume.

[0037] In one implementation, the above-mentioned step 1055 determines the loss function of the elastic network model according to the sample predicted 10,000 investment ratio, the sample predicted complaint volume and the parameter vector of the elastic network model, including the following steps.

[0038] Step 1055a, determining a data fitting term of the loss function according to the sample predicted 10,000 investment ratio, the sample predicted complaint volume, and the parameter vector of the elastic network model; Step 1055b, determining a first regularization term and a second regularization term of the loss function according to the parameter vector, the regularization coefficient, and the regularization trade-off coefficient of the elastic network model; Step 1055c, determining a first improvement term and a second improvement term of the loss function according to the sample prediction ten thousand investment ratio; Step 1055d, determining the loss function according to the data fitting term, the first regularization term, the second regularization term, the first improvement term and the second improvement term.

[0039] In the embodiment of the present application, the expression of the loss function is as follows: .

[0040] in, is the parameter vector consisting of the parameters of the elastic network model, is the sample prediction ratio for the i-th second time period, Predict the number of complaints for a sample, is the regularization coefficient, is the regularization trade-off coefficient. is the data fitting term, is the first regularization term, is the second regularization term, As the first improvement, Compared with the loss function of the existing elastic network model, the loss function of the elastic network model provided in the embodiment of the present application has two new improvement items, which can better characterize the impact of the sample prediction 10,000 investment ratio on the sample prediction complaint volume.

[0041] In one implementation, after determining the predicted complaint volume of the 5G message port according to the predicted 10,000-vote ratio and the 5G message delivery volume in the above step 108, the method further includes: adjusting the 5G message delivery of the 5G message port according to the predicted 10,000-vote ratio or the predicted complaint volume.

[0042] In an embodiment of the present application, after obtaining the predicted complaint volume of the 5G message port by predicting the vote-to-complaint ratio and the 5G message sending volume, the 5G message sending of the 5G message port is managed based on the predicted vote-to-complaint ratio or the predicted complaint volume. For example, the 5G message sending is adjusted to intercept or allow. When the predicted vote-to-complaint ratio or the predicted complaint volume meets the corresponding threshold or range, the sending of the 5G message is intercepted or the sending of the 5G message is allowed. This achieves early prediction of the complaint situation and management of the sending of 5G messages in advance based on the predicted complaint situation, thereby effectively reducing the complaint rate of 5G messages.

[0043] In one implementation, the above-mentioned adjustment of the 5G message sending of the 5G message port according to the predicted investment ratio and the predicted complaint volume may include the following steps.

[0044] Step 110, when the predicted 10,000-vote ratio is greater than a first 10,000-vote ratio threshold, intercepting the 5G message sent by the 5G message port; Step 112, when the predicted 10,000-vote ratio is less than a second 10,000-vote ratio threshold, allowing the 5G message of the 5G message port to be sent; Step 114, when the predicted complaint ratio is greater than or equal to the second complaint ratio threshold and less than or equal to the first complaint ratio threshold, adjust the 5G message sending of the 5G message port according to the predicted complaint volume.

[0045] In an embodiment of the present application, when the predicted 10,000-vote ratio is greater than the first 10,000-vote ratio threshold, it means that the predicted 10,000-vote ratio of the 5G message port at the target time has exceeded the preset first 10,000-vote ratio threshold, and the service quality and user satisfaction of the 5G message are poor, so the 5G message delivery of the 5G message port is intercepted. When the predicted 10,000-vote ratio is less than the second 10,000-vote ratio threshold, it means that the service quality and user satisfaction of the 5G message delivered by the 5G message port at the target time are good, there are fewer complaints, and the complaints are controllable or acceptable, so the 5G message delivery of the 5G message port is allowed. When the predicted 10,000-vote ratio is greater than or equal to the second 10,000-vote ratio threshold, and less than or equal to the first 10,000-vote ratio threshold, the predicted 10,000-vote ratio at this time is between the second 10,000-vote ratio threshold and the first 10,000-vote ratio threshold, it is necessary to further adjust the 5G message delivery of the 5G message port according to the predicted complaint volume.

[0046] In one implementation, the above-mentioned step 114, when the predicted complaint ratio is greater than or equal to the second complaint ratio threshold and less than or equal to the first complaint ratio threshold, adjusts the 5G message sending of the 5G message port according to the predicted complaint volume, including the following steps.

[0047] Step 1141, when the predicted complaint volume is greater than the first complaint volume threshold, intercepting the 5G message sent by the 5G message port; Step 1142: When the predicted complaint volume is less than the second complaint volume threshold, allow the 5G message of the 5G message port to be sent; Step 1143, when the predicted complaint volume is greater than or equal to the second complaint volume threshold and less than or equal to the first complaint volume threshold, adjust the 5G message sending of the 5G message port according to the alternative review method.

[0048] Among them, alternative review methods may include manual review.

[0049] In an embodiment of the present application, when the predicted complaint volume is greater than the first complaint volume threshold, it means that the predicted complaint volume of the 5G message port at the target time has exceeded the preset first complaint volume threshold, and it is predicted that there are many complaints about 5G messages, and the service quality and user satisfaction of 5G messages are poor, so the 5G message delivery of the 5G message port is intercepted. When the predicted complaint volume is less than the second complaint volume threshold, it means that the predicted complaints about 5G messages are less, so the 5G message delivery of the 5G message port is allowed. When the predicted complaint volume is greater than or equal to the second complaint volume threshold, and less than or equal to the first complaint volume threshold, the predicted complaint volume at this time is between the second complaint volume threshold and the first complaint volume threshold, then the 5G message delivery of the 5G message port is further adjusted according to the alternative review method. Thereby, the delivery of 5G messages is managed in advance based on the predicted 10,000-vote ratio or the predicted complaint volume, and the interception or permission of the 5G message port for the delivery of 5G messages is adjusted.

[0050] In one implementation, the above step 106 predicts the 10,000-thousand ... Step 1061, selecting a preset dimension's 10,000-vote ratio from the 10,000-vote ratios of each dimension.

[0051] Among them, since the multi-dimensional feature information includes feature information of multiple dimensions, there may be feature information with poor quality or small contribution, and there may also be predictions of the vote-to-volume ratio and complaint volume for one or multiple dimensions. Therefore, according to the specific situation of the multi-dimensional feature information or the specific requirements of the prediction, the vote-to-volume ratio of the preset dimension is selected as the input of the prediction model to improve the prediction accuracy of the prediction model.

[0052] Step 1062, predicting the ten thousand-vote ratio of the preset dimension through a trained prediction model to obtain a predicted ten thousand-vote ratio of the preset dimension of the 5G message at the target time; Thus, according to the predicted 10,000-vote ratio and the 5G message delivery volume, the predicted complaint volume of the 5G message port is determined, including: Step 1063: Determine the predicted complaint volume of the preset dimension of the 5G message port based on the predicted 10,000-vote ratio of the preset dimension and the 5G message delivery volume.

[0053] In an embodiment of the present application, based on the quality requirements for multi-feature information and the requirements for prediction, the vote ratio of a preset dimension is selected from the vote ratios of each dimension, and the vote ratio of the preset dimension is predicted and processed through a trained prediction model to obtain the predicted vote ratio of the preset dimension of the 5G message at the target time, thereby determining the predicted complaint volume of the preset dimension of the 5G message port based on the predicted vote ratio of the preset dimension and the 5G message delivery volume, thereby achieving targeted prediction of the predicted vote ratio and predicted complaint volume of 5G messages, which can better utilize multi-feature information and make the prediction more in line with the prediction requirements.

[0054] In one implementation, before the above-mentioned step 102 obtains the multi-dimensional feature information of the 5G message port in historical time and the 5G message sending volume, the method also includes the following steps.

[0055] Step 1011, obtaining basic sending and receiving information of each port; wherein the basic sending and receiving information includes agent information.

[0056] Step 1012, determining the port type of each port according to the basic sending and receiving information; Step 1013: determine the port whose port type is 5G message or 5G message uplink channel as a 5G message port.

[0057] In an embodiment of the present application, by acquiring basic information on the reception and transmission of each port, the port type of each port is determined according to the basic reception and transmission information, and the port whose port type is 5G message or 5G message uplink channel is determined as a 5G message port, thereby obtaining multi-dimensional feature information of the 5G message port whose port type is 5G message or 5G message uplink channel, and predicting the predicted investment ratio and predicted complaint volume of the 5G port at the target time, instead of directly predicting through massive 5G messages, thereby improving the accuracy of the prediction and saving the computing resources of the prediction model.

[0058] In one implementation, the above step 1012 determines the port type of each port according to the basic sending and receiving information, including: (1) when the basic information for sending and receiving indicates that the port includes a 5G message, determining that the port type is the 5G message; (2) when the basic information for sending and receiving indicates that the port includes a 5G message uplink channel, determining that the port type is the 5G message uplink and downlink channel; (3) when the basic information for sending and receiving indicates that the port includes 5G messages and other messages, determining that the port type is a multiplexed message; (4) When the basic information for sending and receiving indicates that the port does not include the 5G message, the 5G message up and down channels, and the other messages, the port type is a 2G message.

[0059] In an embodiment of the present application, for the basic transceiver information of each port obtained, when the basic transceiver information indicates that the port includes a 5G message, the port is determined to be a 5G message type port. When the basic transceiver information indicates that the port includes a 5G message uplink channel, the port is determined to be a 5G message uplink and downlink channel type port. When the basic transceiver information indicates that the port includes 5G messages and other messages, the port is determined to be a multiplexed message type port. When the basic transceiver information indicates that the port does not include the 5G message, the 5G message uplink and downlink channels, and the other messages, the port is determined to be a 2G message type port. In this way, multi-dimensional feature information of a 5G message port of a 5G message type or a 5G message uplink channel type is obtained, and the predicted investment ratio and predicted complaint volume of the 5G port at the target time are predicted, instead of directly predicting through massive 5G messages, thereby improving the accuracy of the prediction and saving the computing resources of the prediction model.

[0060] Figure 3 The structure diagram of a 5G message complaint prediction system provided by an embodiment of the present application is shown. Figure 3 The system 300 includes: an acquisition module 31, which is used to obtain multi-dimensional feature information of the 5G message port at a historical time and the 5G message delivery volume of the 5G message port; a determination module 32, which is used to determine the complaint ratio of the 5G message port at the historical time according to the multi-dimensional feature information; a first prediction module 33, which is used to predict the complaint ratio of the 5G message port at the historical time through a trained prediction model to obtain the predicted complaint ratio of the 5G message at the target time; a second prediction module 34, which is used to determine the predicted complaint volume of the 5G message port according to the predicted complaint ratio and the 5G message delivery volume.

[0061] In one implementation, the determination module 32 can be used to determine the vote ratio of each dimension respectively according to the multi-dimensional feature information; and determine the vote ratio of the 5G message port at the historical time according to the vote ratio of each dimension.

[0062] In one implementation, the prediction model includes an elastic network model. The above-mentioned system 300 may also include a prediction model training module, which is used to obtain the sample multi-dimensional feature information of the 5G message port at the first time and the sample 5G message delivery volume of the 5G message port; determine the sample ten thousand vote ratio of the 5G message at the first time according to the sample multi-dimensional feature information; predict and process the sample ten thousand vote ratio through the elastic network model to obtain the sample predicted ten thousand vote ratio of the 5G message at the second time; determine the sample predicted complaint volume of the 5G message port according to the sample predicted ten thousand vote ratio and the sample 5G message delivery volume; determine the loss function of the elastic network model according to the sample predicted ten thousand vote ratio, the sample predicted complaint volume and the parameter vector of the elastic network model; update the parameters of the elastic network model according to the loss function until the loss function converges.

[0063] In one implementation, the prediction model training module can be used to determine the data fitting term of the loss function based on the sample predicted investment ratio, the sample predicted complaint volume, and the parameter vector of the elastic network model; determine the first regularization term and the second regularization term of the loss function based on the parameter vector, regularization coefficient, and regularization trade-off coefficient of the elastic network model; determine the first improvement term and the second improvement term of the loss function based on the sample predicted investment ratio; determine the loss function based on the data fitting term, the first regularization term, the second regularization term, the first improvement term, and the second improvement term.

[0064] In one implementation, the above-mentioned system 300 may also include an adjustment module for adjusting the 5G message sending of the 5G message port according to the predicted investment ratio or the predicted complaint volume.

[0065] In one implementation, the adjustment module can be used to (1) intercept the 5G message delivery of the 5G message port when the predicted complaint ratio is greater than a first complaint ratio threshold; (2) allow the 5G message delivery of the 5G message port when the predicted complaint ratio is less than a second complaint ratio threshold; (3) adjust the 5G message delivery of the 5G message port according to the predicted complaint volume when the predicted complaint ratio is greater than or equal to the second complaint ratio threshold and less than or equal to the first complaint ratio threshold.

[0066] In one implementation, the adjustment module can be used to (3a) intercept the 5G message delivery of the 5G message port when the predicted complaint volume is greater than a first complaint volume threshold; (3b) allow the 5G message delivery of the 5G message port when the predicted complaint volume is less than a second complaint volume threshold; (3c) adjust the 5G message delivery of the 5G message port according to an alternative review method when the predicted complaint volume is greater than or equal to the second complaint volume threshold and less than or equal to the first complaint volume threshold.

[0067] In one implementation, the first prediction module 33 can be used to select the ratio of a preset dimension from the ratios of each dimension; the ratio of the preset dimension is predicted and processed by a trained prediction model to obtain the predicted ratio of the preset dimension of the 5G message at the target time. Thus, the second prediction module 34 can be used to determine the predicted complaint volume of the preset dimension of the 5G message port based on the predicted ratio of the preset dimension and the 5G message delivery volume.

[0068] In one implementation, the above-mentioned system 300 may also include a port determination module for obtaining basic sending and receiving information of each port; determining the port type of each port based on the basic sending and receiving information; and determining the port whose port type is 5G message or 5G message uplink channel as a 5G message port.

[0069] In one implementation, the port determination module may be used to (1) determine that the port type is the 5G message when the basic transceiver information indicates that the port includes a 5G message; (2) determine that the port type is the 5G message uplink channel when the basic transceiver information indicates that the port includes a 5G message; (3) determine that the port type is a multiplexed message when the basic transceiver information indicates that the port includes a 5G message and other messages; and (4) determine that the port type is a 2G message when the basic transceiver information indicates that the port does not include the 5G message, the 5G message uplink channel, and the other messages.

[0070] Figure 4 FIG. 1 shows a schematic diagram of the structure of another 5G message complaint prediction system provided by an embodiment of the present application in one implementation manner, see Figure 4 The system 400 includes a port type classification module 41, a factor-to-investment ratio calculation module 42, a factor-to-investment ratio regression prediction module 43, a complaint volume prediction and statistics module 44, and a complaint prediction and management module 45.

[0071] Among them, the port type classification module 41 is used to classify ports based on the basic information of port reception and transmission, and determine the 5G message port. The factor-to-vote ratio calculation module 42 is used to obtain the multi-dimensional feature information of the 5G message port in historical time and the amount of 5G message sent, and calculate the 5G message ratio of the 5G message port in historical time. The 5G message ratio regression prediction module 43 is used to obtain the predicted 5G message ratio through the trained prediction model of the 5G message port in historical time. The complaint volume prediction and statistics module 44 is used to predict the predicted complaint volume of the 5G message port based on the predicted 5G message ratio and the amount of 5G message sent. The complaint prediction management module 45 is used to manage the sending of 5G messages according to the predicted 5G message ratio or the predicted complaint volume.

[0072] Figure 5 A flow chart of another 5G message complaint prediction method provided in an embodiment of the present application is shown, which can be applied to Figure 3 or Figure 4 The 5G message complaint prediction system shown in Figure 5 , the method may include the following steps.

[0073] Step 501, obtain the basic sending and receiving information of each port, and classify the ports based on the basic sending and receiving information of the ports to determine the 5G message port.

[0074] Among them, the port types include four types: 5G messages, 5G message uplink channels, 2G messages, and multiplexed messages.

[0075] Step 502, obtain the multi-dimensional feature information of the 5G message port in the historical time and the 5G message delivery volume, and calculate the 10,000-thousand investment ratio of the 5G message port in the historical time.

[0076] Among them, see Figure 6 , the investment ratio of each dimension is determined through multi-dimensional feature information, and then the investment ratio of 5G message ports in historical time is comprehensively determined through statistics and integration.

[0077] Step 503, the 10,000-vote ratio of the 5G message port in historical time is obtained through the trained prediction model to obtain the predicted 10,000-vote ratio.

[0078] Among them, the loss function of the prediction model newly added the first improvement item and the second improvement item to better characterize the impact of the sample prediction 10,000-investment ratio on the sample prediction complaint volume.

[0079] Step 504, calculate the predicted complaint volume of the 5G message port based on the predicted 10,000-vote ratio and the 5G message delivery volume.

[0080] Among them, the predicted investment ratio and the predicted complaint volume of the preset dimension can be determined based on the characteristic information of the preset dimension, and more detailed complaint information of each dimension can be obtained to carry out targeted management of the distribution of 5G messages.

[0081] Step 505: Manage the delivery of 5G messages based on the predicted 10,000-vote ratio or the predicted complaint volume.

[0082] Among them, the management of the sending of 5G messages includes intercepting the sending of 5G messages and allowing the sending of 5G messages. The specific situations include (1) when the predicted 10,000-vote ratio is greater than the first 10,000-vote ratio threshold, intercepting the sending of 5G messages from the 5G message port; (2) when the predicted 10,000-vote ratio is less than the second 10,000-vote ratio threshold, allowing the sending of 5G messages from the 5G message port; (3) when the predicted 10,000-vote ratio is greater than or equal to the second 10,000-vote ratio threshold and less than or equal to the first 10,000-vote ratio threshold, adjusting the sending of 5G messages from the 5G message port according to the predicted complaint volume. (3a) when the predicted complaint volume is greater than the first complaint volume threshold, intercepting the sending of 5G messages from the 5G message port; (3b) when the predicted complaint volume is less than the second complaint volume threshold, allowing the sending of 5G messages from the 5G message port; (3c) when the predicted complaint volume is greater than or equal to the second complaint volume threshold and less than or equal to the first complaint volume threshold, adjusting the sending of 5G messages from the 5G message port according to the alternative review method.

[0083] Through the embodiments of the present application, 5G message complaints can be predicted in advance, so that 5G messages can be managed in advance and the complaint rate of 5G messages can be reduced.

[0084] Based on the same technical concept, an embodiment of the present application further provides an electronic device, which is used to execute the above-mentioned 5G message complaint prediction method. Figure 7 A schematic diagram of the structure of an electronic device for implementing each embodiment of the present application. Electronic devices may have relatively large differences due to different configurations or performances, and may include a processor (processor) 701, a communication interface (Communications Interface) 702, a memory (memory) 703 and a communication bus 704, wherein the processor 701, the communication interface 702, and the memory 703 communicate with each other through the communication bus 704. The processor 701 can call a computer program stored in the memory 703 and can be run on the processor 701 to execute the various steps of the above-mentioned 5G message complaint prediction method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0085] The above electronic device structure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, the input unit may include a graphics processing unit (GPU) and a microphone, and the display unit may be configured with a display panel in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit includes a touch panel and at least one of other input devices. The touch panel is also called a touch screen. Other input devices may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0086] The memory can be used to store software programs and various data. The memory may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory may include a volatile memory or a non-volatile memory, or the memory may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct memory bus random access memory (DRRAM).

[0087] The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to the operating system, user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor.

[0088] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the above-mentioned 5G message complaint prediction method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0089] The processor is a processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0090] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned 5G message complaint prediction method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0091] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0092] An embodiment of the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes a program or instructions. When the program or instructions are executed, the various processes of the above-mentioned 5G message complaint prediction method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0093] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0095] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A 5G message complaint prediction method, characterized in that: include: Obtain multi-dimensional feature information of a 5G message port at a historical time and the amount of 5G messages sent by the 5G message port; Determine, according to the multi-dimensional feature information, the investment ratio of the 5G message port in the historical time; The 10,000-thousandth ratio of the 5G message port at the historical time is predicted by a trained prediction model to obtain a predicted 10,000-thousandth ratio of the 5G message at the target time; The predicted complaint volume of the 5G message port is determined based on the predicted 10,000-vote ratio and the 5G message delivery volume.

2. The method according to claim 1, characterized in that The determining, according to the multi-dimensional feature information, the investment ratio of the 5G message port in the historical time includes: According to the multi-dimensional feature information, respectively determine the ten thousand investment ratio of each dimension; According to the ten thousand investment ratios of the various dimensions, the ten thousand investment ratios of the 5G message port at the historical time are determined.

3. The method according to claim 1, characterized in that: The prediction model includes an elastic network model; before predicting the 5G message delivery ratio at the historical time through the trained prediction model to obtain the predicted delivery ratio of the 5G message port at the target time, the method also includes: Obtain sample multi-dimensional feature information of the 5G message port at the first time and the sample 5G message delivery volume of the 5G message port; Determine, according to the multi-dimensional feature information of the samples, the sample-to-sample ratio of the 5G message at the first time; The sample ten thousand vote ratio is predicted by the elastic network model to obtain the sample predicted ten thousand vote ratio of the 5G message at the second time; Determine the sample predicted complaint volume of the 5G message port according to the sample predicted 10,000-vote ratio and the sample 5G message delivery volume; Determine the loss function of the elastic network model according to the sample predicted 10,000 investment ratio, the sample predicted complaint volume, and the parameter vector of the elastic network model; The parameters of the elastic network model are updated according to the loss function until the loss function converges.

4. The method according to claim 3, characterized in that The step of determining the loss function of the elastic network model according to the sample predicted 10,000 investment ratio, the sample predicted complaint volume, and the parameter vector of the elastic network model includes: Determining a data fitting term of the loss function according to the sample predicted 10,000 investment ratio, the sample predicted complaint volume, and a parameter vector of the elastic network model; Determining a first regularization term and a second regularization term of the loss function according to a parameter vector, a regularization coefficient, and a regularization trade-off coefficient of the elastic network model; Determining a first improvement term and a second improvement term of the loss function according to the sample prediction 10,000-investment ratio; The loss function is determined according to the data fitting term, the first regularization term, the second regularization term, the first improvement term and the second improvement term.

5. The method according to claim 1, characterized in that After determining the predicted complaint volume of the 5G message port according to the predicted 10,000-vote ratio and the 5G message delivery volume, the method further includes: According to the predicted investment-to-volume ratio or the predicted complaint volume, the 5G message sending of the 5G message port is adjusted.

6. The method according to claim 5, characterized in that The adjusting, according to the predicted 10,000-investment ratio and the predicted complaint volume, the sending of the 5G message of the 5G message port includes: When the predicted 10,000-vote ratio is greater than a first 10,000-vote ratio threshold, intercepting the 5G message sent by the 5G message port; When the predicted 10,000-vote ratio is less than a second 10,000-vote ratio threshold, allowing the 5G message of the 5G message port to be sent; When the predicted complaint ratio is greater than or equal to the second complaint ratio threshold and less than or equal to the first complaint ratio threshold, the 5G message delivery of the 5G message port is adjusted according to the predicted complaint volume.

7. The method according to claim 6, characterized in that When the predicted complaint ratio is greater than or equal to the second complaint ratio threshold and less than or equal to the first complaint ratio threshold, adjusting the 5G message delivery of the 5G message port according to the predicted complaint amount, including: When the predicted complaint volume is greater than a first complaint volume threshold, intercepting the 5G message sent by the 5G message port; When the predicted complaint amount is less than a second complaint amount threshold, allowing the 5G message of the 5G message port to be sent; When the predicted complaint volume is greater than or equal to the second complaint volume threshold and less than or equal to the first complaint volume threshold, the 5G message sending of the 5G message port is adjusted according to the alternative review method.

8. The method according to claim 2, characterized in that: The predicting process of using a trained prediction model to predict the 10,000-thousandth ratio of the 5G message port at the historical time to obtain the predicted 10,000-thousandth ratio of the 5G message at the target time includes: Selecting a preset dimension's ten thousand investment ratio from the ten thousand investment ratios of each dimension; The ten thousand-vote ratio of the preset dimension is predicted by a trained prediction model to obtain a predicted ten thousand-vote ratio of the preset dimension of the 5G message at the target time; The step of determining the predicted complaint volume of the 5G message port according to the predicted 10,000-vote ratio and the 5G message delivery volume includes: According to the predicted 10,000-vote ratio of the preset dimension and the 5G message delivery volume, the predicted complaint volume of the preset dimension of the 5G message port is determined.

9. The method according to claim 1, characterized in that: Before obtaining the multi-dimensional feature information of the 5G message port at the historical time and the 5G message delivery volume, the method further includes: Get the basic sending and receiving information of each port; Determine the port type of each port according to the basic sending and receiving information; The port whose port type is 5G message or 5G message uplink channel is determined as a 5G message port.

10. The method according to claim 9, characterized in that Determining the port type of each port according to the basic sending and receiving information includes: When the basic information for sending and receiving indicates that the port includes a 5G message, determining that the port type is the 5G message; When the basic information for sending and receiving indicates that the port includes a 5G message uplink channel, determining that the port type is the 5G message uplink and downlink channel; When the basic information for sending and receiving indicates that the port includes 5G messages and other messages, determining that the port type is a multiplexed message; When the basic information for sending and receiving indicates that the port does not include the 5G message, the 5G message up and down channels, and the other messages, the port type is a 2G message.

Images