Service acceptance method and device of conference tablet and electronic equipment

Abstract

The invention discloses a service acceptance method and device of a conference tablet and electronic equipment. The method comprises the following steps: determining a business acceptance result of a target object through a first model; under the condition that the service acceptance result exceeds a preset acceptance threshold value, determining a target service corresponding to the target object through a second model; and detecting the telephone charge balance when the target object handles the target service, determining the arrearage time limit of the target object through the third model under the condition of the arrearage state, and performing payment reminding processing on the conference tablet computer used by the target object according to the arrearage time limit. According to the method and the device, the technical problems that the default risk of a user cannot be effectively evaluated, a proper service package is difficult to accurately recommend and a personalized payment reminding strategy is lacked when an operator provides a conference tablet contract service in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the field of communication technologies, and in particular, to a service acceptance method, apparatus, and electronic device for a conference tablet. Background Art

[0002] With the deepening of digital transformation, the demand of each unit for improving the efficiency of internal meetings and remote collaboration capabilities is increasing day by day. As an intelligent device integrating multiple functions, the conference tablet has become a powerful tool to meet this demand because it integrates diverse functions such as a display screen, a projector, an electronic whiteboard, a sound system, a conference terminal, a camera, and an omnidirectional microphone.

[0003] In order to meet the customer's demand for high-quality products and services with cost-effectiveness, current operators have launched a conference tablet business model similar to the mobile phone contract package, that is, users can rent a conference tablet and other cloud computer services by paying a monthly rent fee, and obtain a high-performance intelligent conference solution at a discount lower than the market price. The innovation of this model lies in combining the tablet device with cloud services to form the "dual-system" advantage, which not only meets the low-cost demand for android system devices but also provides the high-performance experience of the windows system cloud computer. The goal is to provide customers with comprehensive services with both price advantages and brand guarantees. However, the implementation of this business model faces multiple challenges, including but not limited to the assessment of default risks, the recommendation of personalized packages, intelligent collection under the arrears status, and the prevention of device information.

[0004] Specifically, in the related art, operators mainly rely on static credit assessment and unified collection processes in business acceptance and collection strategies. This method may be insufficient when facing high-value devices such as conference tablets, unable to effectively predict and control default behaviors, and unable to provide customized services according to the consumption capabilities and creditworthiness of different users. At the same time, the business acceptance review process in the related art often relies on the personal credit assessment of financial institutions such as banks, and most of the arrears collection methods are after-the-fact processing, lacking forward-looking prediction and differentiated management. In addition, the purchase methods in the conference tablet market are mostly full payment or credit card installment, which is essentially different from the business acceptance and arrears management in the operator contract mode, especially in terms of dynamic business package recommendation and device information protection in combination with user portraits, and there is no mature solution.

[0005] In response to the above problems, no effective solution has been proposed yet. Summary of the Invention

[0006] An embodiment of the present application provides a method, apparatus, and electronic device for business acceptance of a conference tablet, so as to at least solve the technical problems that in the related art, when an operator provides a conference tablet contract service, it is impossible to effectively evaluate the user's default risk, difficult to accurately recommend a suitable business package, and lack of a personalized collection strategy.

[0007] According to one aspect of the embodiments of the present application, a method for business acceptance of a conference tablet is provided, including: determining a business acceptance result of a target object through a first model, where the first model is used to predict the default behavior of the target object by analyzing the historical business information of the target object, and the business acceptance result is used to represent the probability value that the target object meets the conference tablet handling conditions; in the case where the business acceptance result exceeds a preset acceptance threshold, determining a target business corresponding to the target object through a second model, where the second model is used to determine the matching degree between the target object and all the services in the conference tablet by analyzing the consumption behavior of the target object; detecting the phone bill balance of the target object when handling the target business, and in the case of an overdue state, determining the overdue time limit of the target object through a third model, and performing a collection process on the conference tablet used by the target object according to the overdue time limit, where the third model is used to predict the overdue behavior of the target object, and the overdue time limit is used to represent the time allowed for the target object to continue using the conference tablet in an overdue state.

[0008] Optionally, the first model is trained in the following manner: obtaining the historical business information of the target object; determining the attribute quantization score of the target object according to the historical business information, where the attribute quantization score at least includes the customer star rating score, consumption level score, overdue credit score, and complaint credit score of the target object; training the initial default prediction model through a convolutional neural network according to the attribute quantization score until the preset number of iterations is reached and then stopping the training to obtain the first model.

[0009] Optionally, the overdue credit score is determined in the following manner: determining the historical overdue times of the target object, and determining the cumulative overdue amount of the target object in the historical overdue times; determining a first quantization value corresponding to the historical overdue times, and determining a second quantization value corresponding to the cumulative overdue amount; determining the overdue credit score according to the first quantization value and the second quantization value.

[0010] Optionally, in the case where the historical overdue times exceed the preset overdue times, the first quantization value is determined to be 0; and in the case where the cumulative overdue amount exceeds the preset overdue amount, the second quantization value is determined to be 0.

[0011] Optionally, the complaint credit score is determined as follows: determining the historical number of complaints of the target object and determining the cumulative score of the complaint events of the target object; determining a third quantization value corresponding to the historical number of complaints and determining a fourth quantization value corresponding to the cumulative score of the complaint events; determining the complaint credit score based on the third quantization value and the fourth quantization value.

[0012] Optionally, when the historical number of complaints exceeds the preset number of complaints, the third quantization value is determined to be 0; and when the cumulative score of the complaint events exceeds the preset cumulative score, the fourth quantization value is determined to be 0.

[0013] Optionally, the overdue time limit of the target object is determined through a third model, including: obtaining the historical bill information of the target object; determining the estimated overdue days of the target object based on the attribute quantization score of the target object and the historical bill information, where the estimated overdue days are used to reflect the overdue behavior of the target object; determining the overdue time limit based on the estimated overdue days through the target activation function.

[0014] Optionally, the target service corresponding to the target object is determined through a second model, including: determining the location of the network point where the target object handles the meeting tablet service; determining the network point range corresponding to the target object based on the network point location and the preset range, and obtaining the number of service types and service consumption data within the network point range; determining a first customer group based on the number of service types and service consumption data, and determining a second customer group based on the service consumption data and the current service package of the target object; determining a third customer group that has handled the meeting tablet service and a fourth customer group that has not handled the meeting tablet service from the second customer group; determining the target service through the second model based on the distribution ratios of the third customer group and the fourth customer group.

[0015] Optionally, determining the target service based on the distribution ratios of the third customer group and the fourth customer group includes: determining the first distribution ratio of the third customer group compared to the second customer group and determining the second distribution ratio of the fourth customer group compared to the first customer group and the second customer group; determining the first weight of the first distribution ratio and determining the second weight of the second distribution ratio; determining the service recommendation probability of the target object based on the first distribution ratio, the second distribution ratio, the first weight, and the second weight; determining the target service based on the service recommendation probability.

[0016] Optionally, a collection process is performed on the meeting tablet used by the target object based on the overdue time limit, including: sending a payment reminder to the meeting tablet used by the target object when the remaining preset days of the overdue time limit; sending a shutdown instruction to the meeting tablet used by the target object when the overdue time limit is reached.

[0017] According to another aspect of the embodiments of the present application, there is also provided a business acceptance device for a conference tablet, including: a first determination module, configured to determine the business acceptance result of a target object through a first model, where the first model is used to predict the default behavior of the target object by analyzing the historical business information of the target object, and the business acceptance result is used to represent the probability value that the target object meets the conditions for handling the conference tablet; a second determination module, configured to determine the target business corresponding to the target object through a second model when the business acceptance result exceeds a preset acceptance threshold, where the second model is used to determine the matching degree between the target object and all services in the conference tablet by analyzing the consumption behavior of the target object; a third determination module, configured to detect the phone bill balance of the target object when handling the target business, and when in an arrears state, determine the arrears time limit of the target object through a third model, and perform a reminder process on the conference tablet used by the target object according to the arrears time limit, where the third model is used to predict the arrears behavior of the target object, and the arrears time limit is used to represent the time allowed for the target object to continue using the conference tablet in an arrears state.

[0018] According to yet another aspect of the embodiments of the present application, there is also provided an electronic device, including: a memory and a processor, where the memory is used to store program instructions; the processor is connected to the memory and is used to execute the business acceptance method of the above-mentioned conference tablet.

[0019] According to still another aspect of the embodiments of the present application, there is also provided a non-volatile storage medium, and the non-volatile storage medium includes a stored computer program, where the device where the non-volatile storage medium is located executes the business acceptance method of the above-mentioned conference tablet by running the computer program.

[0020] According to still another aspect of the embodiments of the present application, there is also provided a computer program product, including computer instructions, and when the computer instructions are executed by a processor, the business acceptance method of the above-mentioned conference tablet is implemented.

[0021] In an embodiment of the present application, a business acceptance result of a target object is determined by a first model, wherein the first model is used to predict the target object's default behavior by analyzing the target object's historical business information, and the business acceptance result is used to indicate the probability value that the target object meets the conditions for handling a conference tablet; when the business acceptance result exceeds a preset acceptance threshold, a target business corresponding to the target object is determined by a second model, wherein the second model is used to determine the degree of matching between the target object and all businesses in the conference tablet by analyzing the target object's consumption behavior; the target object's call balance when handling the target business is detected, and when the target object is in arrears, the target object's arrears time limit is determined by a third model, and the conference tablet used by the target object is urged to pay according to the arrears time limit, wherein the third model is used to predict the target object's arrears behavior, and the arrears time limit is used to indicate the time during which the target object is allowed to continue to use the conference tablet in an arrears state, thereby achieving the purpose of accurately identifying the user's default risk, personalized recommendation of business packages, and effective control of the arrears state, thereby achieving the technical effect of improving the operator's asset security, optimizing the user experience, and enhancing the service efficiency, thereby solving the technical problems in the related technology that the operator cannot effectively assess the user's default risk, it is difficult to accurately recommend suitable business packages, and lacks a personalized collection strategy when providing conference tablet contract services. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0023] Figure 1 It is a hardware structure diagram of a computer terminal for implementing a business acceptance method of a conference tablet according to an embodiment of the present application;

[0024] Figure 2 It is a flow chart of a business acceptance method of a conference tablet according to an embodiment of the present application;

[0025] Figure 3 It is a schematic diagram of a business acceptance process of a conference tablet according to an embodiment of the present application;

[0026] Figure 4 It is a structural diagram of a business acceptance device of a conference tablet according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] To enable those skilled in the art to better understand the solution of this application, the following will clearly and completely describe the technical solution in the embodiments of this application in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0028] It should be noted that the terms "first", "second", etc. in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0029] First, some nouns or terms that appear in the process of explaining the embodiments of this application are applicable to the following explanations:

[0030] ResNet18 and ResNet50: ResNet is a deep residual network designed to solve the problems of gradient vanishing and explosion in deep neural networks. Through residual learning, the network can be trained more effectively and deeper. ResNet18 and ResNet50 refer to two specific depth architectures of this network, and the numbers 18 and 50 represent the number of network layers respectively.

[0031] Sigmoid function: A mathematical function whose shape is similar to the letter S, which maps the input value to the range of 0 to 1. In machine learning, the Sigmoid function is often used in binary classification problems and can convert the output of the model into a probability value for easy understanding and decision-making.

[0032] Softmax function: A function used in multi-classification tasks that converts the raw output of each class into a probability distribution, making the sum of all outputs equal to 1, and the class corresponding to the highest output is selected.

[0033] To solve the problem of poor efficiency in business acceptance of conference tablets in the related art, the embodiments of this application provide a method for business acceptance of conference tablets, which can run on Figure 1 the computer terminal shown below. The following describes this computer terminal.

[0034] The method embodiments for business acceptance of the meeting flat panel provided by the embodiments of the present application can be executed on a mobile terminal, a computer terminal or a similar computing device. Figure 1 The following shows a hardware structure block diagram of a computer terminal for implementing the method for business acceptance of a meeting flat panel. As Figure 1 shown, the computer terminal 10 may include one or more processors (in the figure, 102a, 102b,..., 102n are used to illustrate), where the processor may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, a memory 104 for storing data, and a transmission module 106 for communication functions connected by a wired and / or wireless network. In addition, it may further include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those of ordinary skill in the art can understand that Figure 1 the structure shown is only schematic and does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 may further include more or fewer components than those Figure 1 shown, or have a different configuration from that Figure 1 shown.

[0035] It should be noted that the above one or more processors and / or other data processing circuits are generally referred to as "data processing circuits" in this article. The data processing circuit may be embodied in whole or in part as software, hardware, firmware, or any combination thereof. In addition, the data processing circuit may be a single independent processing module, or be incorporated in whole or in part into any one of the other elements in the computer terminal 10. As involved in the embodiments of the present application, the data processing circuit is used for processor control (such as the selection of a variable resistance terminal path connected to an interface).

[0036] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for business acceptance of the meeting flat panel in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the above-mentioned method for business acceptance of the meeting flat panel. The memory 104 may include a high-speed random access memory, and may further include a non-volatile memory, such as one or more magnetic storage devices, a flash memory, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely set relative to the processor, and these remote memories can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an enterprise internal network, a local area network, a mobile communication network, and combinations thereof.

[0037] The transmission module 106 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission module 106 can be a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0038] The display can be, for example, a touch-screen liquid crystal display (LCD), which enables users to interact with the user interface of the computer terminal 10.

[0039] It should be noted here that in some alternative embodiments, the above Figure 1 The computer terminal shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware elements and software elements. It should be pointed out that Figure 1 is only an example of a specific specific instance and is intended to show the types of components that may exist in the above computer terminal.

[0040] Under the above operating environment, an embodiment of a method for accepting services of a conference tablet is provided in an embodiment of the present application. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0041] Figure 2 is a flowchart of a method for accepting services of a conference tablet according to an embodiment of the present application. As Figure 2 shown, the method includes the following steps:

[0042] Step S202, determining the service acceptance result of the target object through a first model, where the first model is used to predict the default behavior of the target object by analyzing the historical service information of the target object, and the service acceptance result is used to represent the probability value that the target object meets the conditions for handling the conference tablet.

[0043] In the above step S202, the first model (such as the default behavior prediction model) can be used to predict the default behavior of the user (i.e., the above-mentioned target object). Specifically, the first model is trained using a convolutional neural network (such as ResNet18). By inputting the user attribute data into the first model, the business acceptance result of the user can be obtained. Among them, the business acceptance result is a probability value used to determine whether the user meets the handling conditions for the meeting tablet business. If the business acceptance result exceeds the preset acceptance threshold, it indicates that the system predicts that the default risk of this user is relatively low, and the application for the meeting tablet business can be accepted.

[0044] Step S204, in the case where the business acceptance result exceeds the preset acceptance threshold, determine the target business corresponding to the target object through the second model, where the second model is used to determine the matching degree between the target object and all the businesses in the meeting tablet by analyzing the consumption behavior of the target object.

[0045] In the above step S204, when the business acceptance result of the user meets the conditions, the second model (such as the regional user consumption model) can be used to determine the meeting tablet business that best matches this user, that is, the above-mentioned target business. The purpose of this second model is to recommend the best business package for the user through multi-dimensional user screening and similar user group analysis. Specifically, the system will count the consumption records of nearby outlets according to the geographical location of the user; combine the current package level and monthly bill level of the user to select a similar user group. Subsequently, comprehensively analyze the consumption distributions of these two types of user groups, and combine the personal consumption habits and regional consumption characteristics of the user to recommend the meeting tablet business package that best meets the user's expectations, realizing the personalization and precision of package recommendation.

[0046] Step S206, detect the phone bill balance of the target object when handling the target business. In the case of an overdue state, determine the overdue time limit of the target object through the third model, and conduct a reminder process on the meeting tablet used by the target object according to the overdue time limit, where the third model is used to predict the overdue behavior of the target object, and the overdue time limit is used to represent the time allowed for the target object to continue using the meeting tablet in an overdue state.

[0047] In the above step S206, after the user has handled the above target service, the cloud-side management system will continuously detect the user's phone bill balance. When it detects that the user is in an overdue state, the system will enable the third model (such as the overdue days prediction model) to predict the user's overdue behavior and determine the overdue time limit. Among them, the third model is a deep learning-based overdue days prediction model trained using ResNet50. This model can predict the possible overdue duration of the user based on the consumption behavior data of the target object and the local overdue situation. The overdue time limit is determined by the cloud-side management system based on the user's creditworthiness and the overdue prediction result, indicating the maximum time the user can continue to use the meeting tablet in an overdue state. Users with a higher credit may enjoy a longer overdue time limit, while users with a lower credit will be immediately shut down until the payment is completed. This differential management strategy not only ensures the asset security of the operator but also takes into account the user experience, avoiding customer dissatisfaction that may be caused by a one-size-fits-all shutdown process.

[0048] Through the above steps S202 to S206, the purposes of accurately identifying the user's default risk, personalized recommending business packages, and effectively controlling the overdue state are achieved, thus realizing the technical effects of improving the operator's asset security, optimizing the user experience, and enhancing the service efficiency, and further solving the technical problems in the related technologies that the operator cannot effectively evaluate the user's default risk, accurately recommend appropriate business packages, and lack personalized collection strategies when providing meeting tablet contract services. The following is a detailed description.

[0049] In the above step S202, the first model is trained in the following way: Obtain the historical business information of the target object; Determine the attribute quantization score of the target object based on the historical business information, where the attribute quantization score at least includes the customer star rating score, consumption level score, overdue credit score, and complaint credit score of the target object; Train the initial default prediction model through a convolutional neural network based on the attribute quantization score until the preset number of iterations is reached and then stop training to obtain the first model.

[0050] In the embodiments of the present application, the training process of the first model reflects the in-depth mining and intelligent analysis of the user's historical business information. Specifically, first, multi-dimensional information such as the user's customer star rating, consumption level, arrears history, and complaint history can be obtained from the cloud business management system, and attribute quantization scores are calculated for it, converting abstract user characteristics into specific numerical indicators for easy understanding and learning by the model. Secondly, a convolutional neural network (such as ResNet18) is used to train these scoring data, and the network can automatically extract and learn the complex relationship between user attribute scores and default behaviors to form a default prediction model. Through a large number of iterative trainings, the model is gradually optimized until it stops after reaching the preset number of iterations, and finally a stable default behavior prediction model, that is, the first model, is obtained. This first model can predict the default probability of the target object, provide a scientific basis for business acceptance, ensure that only users who meet the credit conditions can handle the meeting tablet business, effectively reduce the default risk, and protect the asset security of the operator. The following explains each attribute quantization score separately.

[0051] (1) Customer star rating score

[0052] The customer star rating score is a quantitative evaluation method of the operator for the user's credit and service level. It converts different service levels of users from 1 star to 7 stars into specific numerical scores. A higher star rating corresponds to a higher score, indicating that the user enjoys a higher level of credit and service. The specific expression is as follows:

[0053]

[0054] In the formula, x1 represents the customer star rating score, and d1 represents the customer star rating.

[0055] For example: The customer star rating is 5 stars, that is, d1 = 5, then the quantified customer star rating score x1 = 85.

[0056] (2) Consumption level score

[0057] The consumption level score is a quantitative method for evaluating the user's average monthly consumption bill in the past year. It compares the individual consumption behavior in the context of the group consumption behavior in the same region to reflect whether the user's consumption level is above or below the average level of the region. The specific expression is as follows:

[0058]

[0059] In the formula, x2 represents the consumption level score, and d2 represents the user's consumption level ranking.

[0060] For example: A certain user's phone bill consumption level ranks in the top 30% in the region, that is, d2 = 0.3, then the quantified consumption level score x2 = 84.

[0061] (3) Overdue Credit Score

[0062] The overdue credit score is a credit assessment of a user's past payment behavior. It is based on the user's historical overdue records, including the number of overdue times and the cumulative overdue amount. Through a quantitative method, this information is converted into a score to intuitively reflect the user's credit level and reliability in paying bills.

[0063] Specifically, the overdue credit score is determined as follows: determining the historical number of overdue times of the target object and determining the cumulative overdue amount of the target object among the historical number of overdue times; determining the first quantitative value corresponding to the historical number of overdue times and determining the second quantitative value corresponding to the cumulative overdue amount; determining the overdue credit score based on the first quantitative value and the second quantitative value. Further, in the case where the historical number of overdue times exceeds the preset number of overdue times, the first quantitative value is determined to be 0; and in the case where the cumulative overdue amount exceeds the preset overdue amount, the second quantitative value is determined to be 0. The specific expression is as follows:

[0064] The quantization formula for the historical number of overdue times of a user is as follows, and in the case where the historical number of overdue times exceeds 4 times (i.e., the above-mentioned preset number of overdue times), the corresponding first quantitative value is determined to be 0:

[0065]

[0066] In the formula, x 31 represents the first quantitative value, and d 31 represents the historical number of overdue times.

[0067] The quantization formula for the cumulative overdue amount is as follows, and in the case where the cumulative overdue amount exceeds 1000 yuan (i.e., the above-mentioned preset overdue amount), the corresponding second quantitative value is determined to be 0:

[0068]

[0069] In the formula, x 32 represents the second quantitative value, and d 32 represents the cumulative overdue amount.

[0070] The final overdue credit score is the product of the historical number of overdue times and the cumulative overdue amount:

[0071]

[0072] In the formula, x3 represents the overdue credit score.

[0073] For example: A certain user has 2 overdue times and a cumulative overdue amount of 300 yuan, that is, d 31 = 2, d 32= 300, the quantization results of the historical overdue times and the cumulative overdue amount are respectively: the first quantization value x 31 = 77, the second quantization value x 32 = 84, then the final overdue credit score, after rounding, is x3 = 65.

[0074] (4) Complaint credit score

[0075] The complaint credit score is a quantitative assessment of the complaint frequency and nature that a user experiences during the business service process, aiming to reflect whether the user's behavior complies with service specifications and the impact on other users or the operator.

[0076] Specifically, the complaint credit score is determined as follows: determine the historical complaint times of the target object and determine the cumulative score of the complaint events of the target object; determine the third quantization value corresponding to the historical complaint times and determine the fourth quantization value corresponding to the cumulative score of the complaint events; determine the complaint credit score based on the third quantization value and the fourth quantization value. Further, when the historical complaint times exceed the preset complaint times, the third quantization value is determined to be 0; and when the cumulative score of the complaint events exceeds the preset cumulative score, the fourth quantization value is determined to be 0. The specific expression is as follows:

[0077] The quantization formula for the user's historical complaint times is as follows, and when the historical complaint times exceed 2 times (i.e., the above-mentioned preset complaint times), the corresponding third quantization value is determined to be 0:

[0078]

[0079] In the formula, x 41 represents the third quantization value, and d 41 represents the historical complaint times.

[0080] The complaint content is divided into 4 levels according to the severity, accounting for 0.8, 0.6, 0.4, and 0.2 points respectively. The quantization formula for the cumulative score of the user's complaint events is as follows, and when the cumulative score of the complaint events exceeds 1 point (i.e., the above-mentioned preset cumulative score), the corresponding fourth quantization value is determined to be 0:

[0081]

[0082] In the formula, x 42 represents the fourth quantization value, and d 42 represents the cumulative score of all the user's complaint events.

[0083] The final complaint credit score is the product of the historical complaint times and the cumulative score.

[0084]

[0085] In the formula, x4 represents the complaint credit score.

[0086] For example: A certain user has been complained twice, once for advertising promotion and once for other complaints, that is, d 41 = 2, d 42 = 0.6. The quantization results of the historical complaint times and the cumulative score are: the third quantization value x 41 = 58, the fourth quantization value x 42 = 63. Then the final complaint credit score after rounding is x4 = 37.

[0087] After integrating the above-obtained attribute quantization scores, it is shown in Table 1 as follows:

[0088] Table 1 User Attribute Quantization Score

[0089] Customer star rating Consumption level rating Arrears credit rating Complaint credit rating User 1 85 84 65 37

[0090] Subsequently, taking the user attribute quantization score as the training data, use the convolutional neural network resnet18 for training to obtain the default behavior prediction model A, that is, the above-mentioned first model. The customer star rating score x1, consumption level score x2, overdue credit score x3, and complaint credit score x4 in a total of four dimensions are used as the input of model A, and the input format is [x1, x2, x3, x4]. Whether there has been a default behavior is used as the output, and the output format is default y = [0, 1], no default y = [1, 0].

[0091] In order to maintain the accuracy and timeliness of model A, the training data of the model will be updated regularly every month, collecting the latest user historical information, including the latest consumption bills, overdue records, complaint situations, etc., and retraining model A through the new data set to update its weights and parameters, ensuring that the model can reflect the latest user behavior trends and market changes, thereby improving the accuracy and reliability of default prediction and providing solid data support for the business decisions of the operator.

[0092] Generally speaking, the training process of the above-mentioned first model combines historical business information quantization, convolutional neural network training, and iterative optimization, provides accurate default prediction for business acceptance, realizes the intelligent evaluation of the user's credit status, and ensures the scientificity and rationality of business acceptance decisions.

[0093] In the above step S204, determining the target service corresponding to the target object through the second model includes: determining the location of the network point where the target object handles the meeting tablet service; determining the network point range corresponding to the target object according to the network point location and the preset range, and obtaining the number of service types and service consumption data within the network point range; determining the first customer group according to the number of service types and service consumption data, and determining the second customer group according to the service consumption data and the current service package of the target object; determining the third customer group that has handled the meeting tablet service and the fourth customer group that has not handled the meeting tablet service from the second customer group; determining the target service through the second model according to the distribution ratios of the third customer group and the fourth customer group.

[0094] Optionally, determining the target service according to the distribution ratios of the third customer group and the fourth customer group includes: determining the first distribution ratio of the third customer group compared to the second customer group, and determining the second distribution ratio of the fourth customer group compared to the first customer group and the second customer group; determining the first weight of the first distribution ratio, and determining the second weight of the second distribution ratio; determining the service recommendation probability of the target object according to the first distribution ratio, the second distribution ratio, the first weight, and the second weight; determining the target service according to the service recommendation probability.

[0095] In the embodiment of the present application, the second model can not only intelligently determine the recommended service for the target user, but also ensure that the recommendation result is not only based on personal consumption habits, but also in line with the regional consumption level, so as to provide a more accurate and practical business selection, optimize the user service experience, and improve the efficiency and success rate of service acceptance. The specific steps can be as follows:

[0096] S1: First, determine the specific location of the network point (business hall) where the user handles the meeting tablet service. Based on this network point location, define a preset range (such as within e kilometers nearby, where the reference value of e is 5) to cover other network points in the surrounding area of the business hall. According to the above network point range, collect and analyze the number of service types and service consumption data in this area.

[0097] S2: Sort the monthly bill amounts in the service consumption data, and divide the total population Q geometrically in combination with the number of service types a to form multiple subdivided first customer groups {Q1, Q2... Q a}. At the same time, select the second customer group P with similar current service packages and average monthly phone bill levels of the user within the past six months, with a similar range of ±b%, and the reference value of b is 30, to further refine the group division of the data. It should be noted that Q and P are two ways of dividing and screening the customer population.

[0098] S3: Further distinguish within the second customer group P. According to whether the user has ever handled the conference tablet business, the group is subdivided into the third customer group P1 that has handled the conference tablet business and the fourth customer group P2 that has not handled the conference tablet business.

[0099] S4: For the third customer group P1, calculate its first distribution proportion O1 in each business type, and obtain {O 11 , O 12 … O 1a}, which represents the proportion of people with similar consumption habits and consumption levels who have handled the tablet business in the 1st to th business types respectively. And for the fourth customer group P2, calculate its second distribution proportion O2 in each business type, and obtain {O 21 , O 22 … O 2a}, which represents the proportion of people with similar consumption habits and consumption levels who have not handled the tablet business in {Q1, Q2… Q a}. It should be noted that since the fourth customer group P2 does not have the reference condition for handling the tablet business, it is only possible to combine the package amount and monthly phone bill of the fourth customer group P2 in the distribution of {Q1, Q2… Q a} and take the intersection to calculate the second distribution proportion O2 of different businesses.

[0100] S5: Determine the first weight z1 corresponding to the first distribution proportion O1; and determine the second weight z2 corresponding to the second distribution proportion O2. It should be noted that the setting of the weight is based on the distribution proportion of P1 in the P group to ensure that the contribution degree of the user group directly with conference tablet business experience to the recommendation is reasonably enhanced. The specific expression is as follows:

[0101]

[0102] z2 = 1 - z1

[0103] S6: Determine the business recommendation probability W of the target object based on the first distribution proportion O1, the second distribution proportion O2, the first weight z1 and the second weight z2, and determine the target business based on the business recommendation probability W. The specific expression is as follows:

[0104] W = z1×O1 + z2×O2

[0105] This business recommendation probability not only takes into account the consumption habits and business needs of the target user personally, but also combines the business preferences of similar user groups in the region, better realizing the organic combination of personalization and regional adaptability.

[0106] In a specific embodiment, assume that there are currently 4 tiers of meeting tablet services, i.e., a = 4. After sorting by monthly bill amount, Q is divided into (0 - 25%, 25% - 50%, 50% - 75%, 75% - 100%) to obtain {Q1, Q2, Q3, Q4}. Assume that the user package is 100 yuan per month and the monthly phone bill is 120 yuan per month. Then, user P is selected simultaneously according to two screening conditions: the package amount is 70 - 130 yuan per month and the monthly phone bill is 84 - 156 yuan per month.

[0107] Assume that the distribution ratios O1 of P1 in the 4 types of tablet services are (10%, 30%, 40%, 20%) respectively, that is, 10% of the people in P have subscribed to the first-tier meeting tablet service. Assume that the ranges of P2 in the total population Q according to the two methods of package amount and monthly phone bill are 30% - 50% and 40% - 60% respectively. After merging, the intersection range is 30% - 60%. With respect to the distribution O2 of {Q1, Q2, Q3, Q4}, the ratios are (0, 66.7%, 33.3%, 0). The calculation method is that 66.7% of the range of 30% - 60% is included in the second-tier meeting tablet service of 25% - 50% and 33.3% of the third-tier meeting tablet service of 50% - 75%;

[0108] Assume that the proportion of P1 in P is approximately 30%. According to the formula calculation, z1 = 0.61 and z2 = 0.39. After O1 and O2 are distributed according to their respective weights and added together:

[0109] W = 0.61×(10%, 30%, 40%, 20%) + 0.39×(0, 66.7%, 33.3%, 0)

[0110] Finally, the recommended probabilities for the four types of services are (6.1%, 44.3%, 37.4%, 12.2%). That is, according to the analysis of the user and regional consumption levels, the probability of 44.3% is the highest, and the second-tier service package is recommended for the user first.

[0111] In the above step S206, determining the overdue time limit of the target object through the third model includes: obtaining the historical bill information of the target object; determining the estimated overdue days of the target object through the third model based on the attribute quantitative score and historical bill information of the target object, where the estimated overdue days are used to reflect the overdue behavior of the target object; determining the overdue time limit based on the estimated overdue days through the target activation function.

[0112] In the embodiment of the present application, a third model closely related to the user's consumption behavior and credit status is pre-constructed to predict the estimated overdue days of the target object, that is, the user, and based on this prediction result, a personalized overdue time limit is determined, a reasonable shutdown strategy is formulated, and the risk control of the operator and the user service experience are effectively balanced. The specific expression can be as follows:

[0113] D1 = C(M, N, L, x1, x2, x3, x4) + Δd

[0114] Wherein, D1 represents the estimated overdue days, C represents the overdue days prediction model (i.e., the above third model), M represents the monthly bill, N represents the average overdue days in the overdue events within the local area, L represents the proportion of the number of people with overdue records in the total number of people within the local area, Δd represents the date correction factor, and the expected value of D1 is D1 ≥ 0.

[0115] According to the preset rules, the longer the estimated overdue days D1, the higher the user risk, and the shorter the allowable overdue time, with the shortest being 0 days; conversely, the longer the allowable overdue days D1, the longest allowable overdue being 30 days. Therefore, the longest overdue time limit D allowed for overdue can be inferred through the estimated overdue days D1, and the calculation formula is as follows:

[0116]

[0117] Wherein, D represents the overdue time limit, represents the Sigmoid function, which is used to normalize the estimated overdue days.

[0118] The distribution interval of D is [0, 30], that is, high-credit users can be overdue for up to 30 days, and low-credit users can be overdue for up to 0 days.

[0119] According to the predicted longest overdue time limit D allowed for overdue, the cloud-side management system will initiate a shutdown strategy for the device. For example, for users with lower credit and longer estimated overdue days, the system may immediately restrict the device function after the overdue to reduce the risk of operator asset losses; while for users with higher credit and shorter estimated overdue days, a certain buffer time will be provided to allow users to resolve the bill problem within a certain time, thereby optimizing the user experience and avoiding the negative impact on customer relationships that overly strict strategies may bring.

[0120] Optionally, a reminder for payment is sent to the meeting tablet used by the target object when the remaining preset days of the overdue time limit are reached, and a shutdown instruction is sent to the meeting tablet used by the target object when the overdue time limit is reached.

[0121] In the embodiments of the present application, not only can the overdue behavior of users be predicted, but also a reasonable reminder process can be formulated according to the prediction results, including early warning and timely shutdown, ensuring the intelligence, efficiency, and humanization of the reminder process.

[0122] Specifically, the shutdown strategy can be divided into reminder and forced shutdown according to the status of the user's phone bill balance. Among them, reminder means that when it comes to 7 days before the estimated overdue payment date, a mobile phone text message reminder and a tablet timed pop-up reminder will be sent every day on the turned-on tablet to remind the user to pay the bill. The pop-up window can be closed manually; forced shutdown means that when the maximum time limit D allowed for arrears is reached, the tablet local control system will forcibly intervene and disable the tablet device. After detecting that it is turned on, a forced pop-up window will appear and cannot be closed, and it will be in a "locked" state. When the user's phone bill balance is sufficient, the cloud side will send a "unlock" command to the local side to complete the unlocking.

[0123] Figure 3 Schematic diagram of a conference tablet service process according to an embodiment of the present application. Figure 3 As shown in the figure, the entire process of conference tablet business review and acceptance and overdue payment collection is shown in detail, involving multiple key links such as user business application, credit evaluation, business recommendation, equipment control and overdue payment management. The specific steps can be as follows:

[0124] S1: Users go to offline business halls to apply for conference tablet services.

[0125] S2: After the user applies for the service, the default risk of the user is first assessed through the default behavior prediction model A (the first model mentioned above) to determine whether the user meets the conditions for applying for the conference tablet service.

[0126] Model A can generate a normalized output based on customer star rating, consumption level, arrears reputation, and complaint reputation, indicating the possibility of user default. The process will only proceed if the user is assessed by the model as meeting the business processing conditions.

[0127] S3: Regional User Consumption Model B: For users who do not meet the requirements for conference tablets, the conference tablet service will not be provided. For users who meet the requirements for conference tablets, their personal and regional consumption levels are analyzed through regional user consumption model B (i.e. the second model mentioned above) to recommend the best service package.

[0128] Among them, model B can combine the historical consumption data of the user's region and the user's personal consumption habits to generate recommended service packages and probability values, ensuring that the package recommendations are in line with regional consumption trends and meet personal needs.

[0129] S4: After confirming the recommended package, the user signs an agreement to rent the conference tablet.

[0130] During this process, the salesperson needs to fill in the tablet SN code and other relevant information. The SN code is the unique serial number that comes with the device when it leaves the factory. Each user's personal registration information, service package, tablet SN code, user attributes and other information are bound and stored in the system database table.

[0131] S5: Detect whether the device is connected to the network.

[0132] The tablet device needs to be used while connected to the network. When it is powered on for the first time, it will automatically send the device information to the cloud - side business management system to automatically activate the device, including important information such as the SN code and device IP. After the system verifies the owner of the device SN code, it queries the relevant information pre - stored in the cloud - side database and returns it. The tablet will automatically log in to the tablet using the user's account. Except for the first power - on, the tablet device can be powered on and used in the offline state, but most functions are disabled. When the device is connected to the network and the cloud - side business management system and the account balance is verified to be sufficient, all functions can be unlocked.

[0133] S6: Based on the user consumption model and historical arrears records, predict the maximum allowed arrears days for each user through the arrears days prediction model C (i.e., the above - mentioned third model).

[0134] S7: Determine whether to enter the arrears collection stage according to the user's bill status. If the user has arrears, the arrears collection process will be triggered; otherwise, the user can continue to use the meeting tablet device normally.

[0135] S8: For users with arrears, the system will take corresponding collection measures according to the estimated arrears days. When the remaining preset days of the arrears time limit are reached, the system sends payment reminders, including text messages and pop - up reminders, to remind users to pay in time. If the user still has not paid when the allowed arrears time limit is reached, the system will send a shutdown command to the meeting tablet to forcibly stop the device function until the user completes the payment. The shutdown strategy is divided into reminder and forced shutdown, which is personalized according to the user's creditworthiness and allowed arrears days.

[0136] In the embodiment of the present application, a comprehensive meeting tablet business management framework is constructed. Through the default behavior prediction model, regional user consumption model, and arrears days prediction model, accurate assessment of user risks, personalized business package recommendation, and differentiated arrears collection strategies are realized. Using deep learning technology, especially the ResNet architecture, deep mining of user historical data not only improves the accuracy of default behavior prediction but also can recommend the most suitable business package according to the user's consumption habits and the consumption level of the region where the user is located. By quantifying user attributes, dynamically updating the prediction model, and using the Sigmoid function to calculate the arrears time limit, while protecting the asset security of the operator, the user experience is also optimized, avoiding customer dissatisfaction that may be brought by the "one - size - fits - all" shutdown strategy. This intelligent management solution significantly reduces the business risks of the operator, improves service efficiency, brings economic benefits to the operator, and provides a more personalized and flexible service experience for users, achieving a win - win beneficial effect.

[0137] According to an embodiment of the present application, a service acceptance device for a conference tablet is provided. It should be noted that the service acceptance device for the conference tablet in the embodiment of the present application can be used to execute the service acceptance method for the conference tablet provided in the embodiment of the present application. The following introduces the service acceptance device for the conference tablet provided in the embodiment of the present application.

[0138] Figure 4 It is a structural diagram of a service acceptance device for a conference tablet provided according to an embodiment of the present application. As Figure 4 shown, the device includes:

[0139] A first determination module 40, configured to determine the service acceptance result of a target object through a first model, where the first model is used to predict the default behavior of the target object by analyzing the historical service information of the target object, and the service acceptance result is used to represent the probability value that the target object meets the conference tablet handling conditions;

[0140] A second determination module 42, configured to determine a target service corresponding to the target object through a second model when the service acceptance result exceeds a preset acceptance threshold, where the second model is used to determine the matching degree between the target object and all services in the conference tablet by analyzing the consumption behavior of the target object;

[0141] A third determination module 44, configured to detect the phone bill balance of the target object when handling the target service, and when in an overdue state, determine the overdue time limit of the target object through a third model, and perform a reminder process on the conference tablet used by the target object according to the overdue time limit, where the third model is used to predict the overdue behavior of the target object, and the overdue time limit is used to represent the time allowed for the target object to continue using the conference tablet in an overdue state.

[0142] Through the first determination module, the second determination module, and the third determination module in the above-mentioned service acceptance device for a conference tablet,

[0143] In the service acceptance device for a conference tablet provided in the embodiment of the present application, a training module 48 is further included. The training module is configured to obtain the historical service information of the target object; determine the attribute quantization score of the target object according to the historical service information, where the attribute quantization score at least includes the customer star rating score, consumption level score, overdue credit score, and complaint credit score of the target object; train the initial default prediction model through a convolutional neural network according to the attribute quantization score until the preset number of iterations is reached and then stop training to obtain the first model.

[0144] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the training module is further configured to determine the historical overdue times of the target object, and determine the cumulative overdue amount of the target object in the historical overdue times; determine a first quantization value corresponding to the historical overdue times, and determine a second quantization value corresponding to the cumulative overdue amount; and determine an overdue credit score based on the first quantization value and the second quantization value.

[0145] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the training module is further configured to, when the historical overdue times exceed the preset overdue times, determine the first quantization value as 0; and when the cumulative overdue amount exceeds the preset overdue amount, determine the second quantization value as 0.

[0146] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the training module is further configured to determine the historical complaint times of the target object, and determine the cumulative score of the complaint events of the target object; determine a third quantization value corresponding to the historical complaint times, and determine a fourth quantization value corresponding to the cumulative score of the complaint events; and determine a complaint credit score based on the third quantization value and the fourth quantization value.

[0147] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the training module is further configured to, when the historical complaint times exceed the preset complaint times, determine the third quantization value as 0; and when the cumulative score of the complaint events exceeds the preset cumulative score, determine the fourth quantization value as 0.

[0148] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the second determination module is further configured to determine the network point location where the target object handles the conference tablet service; determine a network point range corresponding to the target object according to the network point location and a preset range, and obtain the number of service types and service consumption data within the network point range; determine a first customer group according to the number of service types and the service consumption data, and determine a second customer group according to the service consumption data and the current service package of the target object; determine a third customer group that has handled the conference tablet service and a fourth customer group that has not handled the conference tablet service from the second customer group; and determine the target service according to the distribution ratios of the third customer group and the fourth customer group through a second model.

[0149] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the second determination module is further configured to determine a first distribution ratio of the third customer group compared to the second customer group, and determine a second distribution ratio of the fourth customer group compared to the first customer group and the second customer group; determine a first weight of the first distribution ratio, and determine a second weight of the second distribution ratio; determine a service recommendation probability of the target object based on the first distribution ratio, the second distribution ratio, the first weight, and the second weight; and determine the target service according to the service recommendation probability.

[0150] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the third determination module is further configured to obtain the historical bill information of the target object; determine the estimated overdue days of the target object through the third model according to the attribute quantization score and the historical bill information of the target object, where the estimated overdue days are used to reflect the overdue behavior of the target object; determine the overdue time limit through the target activation function according to the estimated overdue days.

[0151] In the service acceptance device of the conference tablet provided in the embodiment of the present application, the third determination module is further configured to send a payment reminder to the conference tablet used by the target object when the remaining preset days of the overdue time limit; send a shutdown instruction to the conference tablet used by the target object when the overdue time limit is reached.

[0152] The embodiment of the present application further provides an electronic device, including: a memory and a processor, where the memory is used to store program instructions; the processor is connected to the memory and is used to execute the service acceptance method of the above-mentioned conference tablet.

[0153] It should be noted that the above-mentioned electronic device is used to execute Figure 2 the service acceptance method of the conference tablet shown, so the relevant explanations in the service acceptance method of the above-mentioned conference tablet also apply to this electronic device, and will not be elaborated here.

[0154] The embodiment of the present application further provides a non-volatile storage medium, which includes a stored computer program, where the device where the non-volatile storage medium is located executes the service acceptance method of the above-mentioned conference tablet by running the computer program.

[0155] It should be noted that the above-mentioned non-volatile storage medium is used to execute Figure 2 the service acceptance method of the conference tablet shown, so the relevant explanations in the service acceptance method of the above-mentioned conference tablet also apply to this non-volatile storage medium, and will not be elaborated here.

[0156] The embodiment of the present application further provides a computer program product, including computer instructions, and the computer instructions implement the service acceptance method of the above-mentioned conference tablet when executed by a processor.

[0157] It should be noted that the above-mentioned computer program product is used to execute Figure 2 the service acceptance method of the conference tablet shown, so the relevant explanations in the service acceptance method of the above-mentioned conference tablet also apply to this computer program product, and will not be elaborated here.

[0158] The serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0159] In the above embodiments of the present application, the descriptions of the various embodiments each have their own emphasis. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0160] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units can be a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of units or modules can be in an electrical or other form.

[0161] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0162] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0163] If the above-mentioned integrated unit is implemented in the form of 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, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. And the aforementioned storage medium includes: USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks or optical disks and other various media that can store program codes.

[0164] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A business acceptance method for a conference tablet, characterized in that, Including: Determine the business acceptance result of the target object through a first model, where the first model is used to predict the default behavior of the target object by analyzing the historical business information of the target object, and the business acceptance result is used to represent the probability value that the target object meets the meeting tablet handling conditions; When the business acceptance result exceeds a preset acceptance threshold, determine the target business corresponding to the target object through a second model, where the second model is used to determine the matching degree between the target object and all services in the meeting tablet by analyzing the consumption behavior of the target object; Detect the phone bill balance of the target object when handling the target business. When in an overdue state, determine the overdue time limit of the target object through a third model, and perform a reminder process on the meeting tablet used by the target object according to the overdue time limit, where the third model is used to predict the overdue behavior of the target object, and the overdue time limit is used to represent the time allowed for the target object to continue using the meeting tablet in an overdue state.

2. The method according to claim 1, characterized in that, The first model is trained through the following method: Obtain the historical business information of the target object; Determine the attribute quantization score of the target object according to the historical business information, where the attribute quantization score at least includes the customer star rating score, consumption level score, overdue credit score, and complaint credit score of the target object; Train the initial default prediction model through a convolutional neural network according to the attribute quantization score until the training stops after reaching the preset number of iterations, and obtain the first model.

3. The method according to claim 2, wherein The overdue credit score is determined through the following method: Determine the historical overdue times of the target object, and determine the cumulative overdue amount of the target object in the historical overdue times; Determine the first quantization value corresponding to the historical overdue times, and determine the second quantization value corresponding to the cumulative overdue amount; Determine the overdue credit score according to the first quantization value and the second quantization value.

4. The method according to claim 3, characterized in that When the historical overdue times exceed the preset overdue times, determine the first quantization value as 0; and when the cumulative overdue amount exceeds the preset overdue amount, determine the second quantization value as 0.

5. The method according to claim 2, wherein The complaint credit score is determined through the following method: Determine the historical complaint times of the target object, and determine the cumulative score of the complaint events of the target object; Determine the third quantization value corresponding to the historical complaint times, and determine the fourth quantization value corresponding to the cumulative score of the complaint events; Determine the complaint credit score according to the third quantization value and the fourth quantization value.

6. The method according to claim 5, characterized in that, When the historical complaint times exceed the preset complaint times, determine the third quantization value as 0; and when the cumulative score of the complaint events exceeds the preset cumulative score, determine the fourth quantization value as 0.

7. The method according to claim 2, characterized in that, Determining the overdue time limit of the target object through the third model includes: Obtain the historical bill information of the target object; Quantify the score based on the attributes of the target object and the historical bill information, and determine the estimated overdue days of the target object through the third model, where the estimated overdue days are used to reflect the overdue behavior of the target object; Determine the overdue time limit through a target activation function based on the estimated overdue days.

8. The method according to claim 1, wherein Determine the target service corresponding to the target object through the second model, including: Determine the location of the network point where the target object handles the meeting tablet service; Determine the network point range corresponding to the target object according to the network point location and a preset range, and obtain the number of service types and service consumption data within the network point range; Determine the first customer group according to the number of service types and the service consumption data, and determine the second customer group according to the service consumption data and the current service package of the target object; Determine the third customer group that has handled the meeting tablet service and the fourth customer group that has not handled the meeting tablet service from the second customer group; Determine the target service through the second model according to the distribution ratios of the third customer group and the fourth customer group.

9. The method according to claim 8, wherein Determine the target service according to the distribution ratios of the third customer group and the fourth customer group, including: Determine the first distribution ratio of the third customer group compared to the second customer group, and determine the second distribution ratio of the fourth customer group compared to the first customer group and the second customer group; Determine the first weight of the first distribution ratio, and determine the second weight of the second distribution ratio; Determine the service recommendation probability of the target object according to the first distribution ratio, the second distribution ratio, the first weight, and the second weight; Determine the target service according to the service recommendation probability.

10. The method according to claim 1, characterized in that, Perform a reminder process on the meeting tablet used by the target object according to the overdue time limit, including: When the remaining preset days of the overdue time limit are reached, send a payment reminder to the meeting tablet used by the target object; When the overdue time limit is reached, send a shutdown instruction to the meeting tablet used by the target object.

11. A business acceptance device for a conference tablet, characterized in that, Including: The first determination module is used to determine the service acceptance result of the target object through the first model, where the first model is used to predict the default behavior of the target object by analyzing the historical service information of the target object, and the service acceptance result is used to represent the probability value that the target object meets the meeting tablet handling conditions; The second determination module is used to determine the target service corresponding to the target object through the second model when the service acceptance result exceeds the preset acceptance threshold, where the second model is used to determine the matching degree of the target object with all services in the meeting tablet by analyzing the consumption behavior of the target object; A third determination module, configured to detect the balance of the target object's phone bill when handling the target service. In the case of an overdue status, determine the overdue time limit of the target object through a third model, and perform a reminder process on the conference tablet used by the target object according to the overdue time limit. The third model is used to predict the overdue behavior of the target object, and the overdue time limit is used to represent the time allowed for the target object to continue using the conference tablet in an overdue status.

12. An electronic device, characterized in that, Comprising: A memory and a processor. The memory is configured to store program instructions. The processor is connected to the memory and is configured to execute the service acceptance method of the conference tablet according to any one of claims 1 to 10.

13. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program. The device where the non-volatile storage medium is located executes the service acceptance method of the conference tablet according to any one of claims 1 to 10 by running the computer program.

14. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, the service acceptance method of the conference tablet according to any one of claims 1 to 10 is implemented.

Images