Prediction Method, Device, Medium and Equipment for Elderly Care Service Information

Abstract

The present disclosure relates to the technical field of data processing, and provides a prediction method and device for elderly care service information, a computer storage medium, and an electronic device. Among them, the method includes: obtaining the insurance characteristics and user characteristics of a target user; inputting the insurance characteristics and user characteristics into the input layer of a prediction model; performing feature fusion processing on the insurance characteristics and user characteristics through a first weight matrix between the input layer and the hidden layer of the prediction model to obtain fused features; performing prediction processing on the fused features through a second weight matrix between the hidden layer and the output layer of the prediction model to obtain a prediction value of the elderly care service information about the target user, and displaying the identifier of the target user and the prediction value of the elderly care service information of the target user. This technical solution provides a technical solution for analyzing insurance customers and associating them with elderly care community members. At the same time, this technical solution has a relatively high prediction accuracy for the elderly care service information of the target user.

Description

Background Art

[0002] Insurance is an important industry in modern economy and a basic means of risk management. With its unique business advantages, commercial insurance companies have improved the old-age insurance system by building old-age communities and developing new old-age products, which is the integrated development with the traditional old-age service industry, and is also of great significance to the adjustment of the asset-liability structure of commercial insurance and the extension of the industrial chain. At the same time, with the continuous expansion of the demand for old-age services and the enhancement of the innovation ability of the insurance industry, insurance products targeting the needs of the elderly are gradually enriched.

[0003] In the prior art, relevant companies have all provided relevant value-added services for insurance products and old-age service rights. However, due to the lack of in-depth analysis of insurance customers and the failure to associate insurance customers with old-age community services (such as old-age membership levels). It can be seen that with the deepening of business and the intensification of competition, it is necessary to consider providing better services for customers. Therefore, a solution for analyzing insurance customers and associating them with old-age community members is needed to improve the prediction accuracy of old-age service information.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure. Summary of the Invention

[0005] The purpose of the present disclosure is to provide a prediction method for old-age service information, a prediction device for old-age service information, a computer storage medium and an electronic device, and further provide a technical solution for analyzing insurance customers and associating them with old-age community members, which is beneficial to improving the prediction accuracy of old-age service information.

[0006] Other features and advantages of the present disclosure will become apparent through the following detailed description, or be learned in part through the practice of the present disclosure.

[0007] According to the first aspect of the present disclosure, a prediction method for old-age service information is provided, and the method includes:

[0008] Obtain the insurance characteristics and user characteristics of the target user;

[0009] Input the above insurance characteristics and the above user characteristics into the input layer of the prediction model;

[0010] Perform feature fusion processing on the above insurance characteristics and the above user characteristics through the first weight matrix between the above input layer and the hidden layer of the prediction model to obtain fusion features;

[0011] Perform a prediction process on the above fusion features through a second weight matrix between the above hidden layer and the output layer of the above prediction model to obtain a predicted value of the elderly care service information for the above target user, and display the identifier of the above target user and the predicted value of the elderly care service information for the above target user.

[0012] In an exemplary embodiment of the present disclosure, based on the foregoing solution, before the above inputting the above insurance features and the above user features into the input layer of the prediction model, the above method further includes:

[0013] Obtain the insurance features, user features, and actual elderly care service information of historical users as a set of samples to obtain a sample set;

[0014] Obtain the above prediction model through the first part of the samples in the above sample set.

[0015] In an exemplary embodiment of the present disclosure, based on the foregoing solution, the above obtaining the above prediction model through the first part of the samples in the above sample set includes:

[0016] Input the insurance features and user features of each group of samples in the above first part of the samples into the above prediction model to obtain predicted elderly care service information, a first intermediate weight matrix between the above input layer and the above hidden layer, and a second intermediate weight matrix between the above hidden layer and the above output layer;

[0017] Perform iterative calculations on the above first intermediate weight matrix through the above actual elderly care service information and the above predicted elderly care service information to obtain a first weight matrix that meets a preset error threshold, and perform iterative calculations on the above second intermediate weight matrix to obtain a second weight matrix that meets the above preset error threshold.

[0018] In an exemplary embodiment of the present disclosure, based on the foregoing solution, after the above obtaining the above prediction model through the first part of the samples in the above sample set, the above method further includes:

[0019] Test the above prediction model through the second part of the samples in the above sample set to make the prediction model meet preset test metrics, where the above test metrics include at least one of accuracy, recall rate, and AUC.

[0020] In an exemplary embodiment of the present disclosure, based on the foregoing solution, the above obtaining the insurance features of the target user includes:

[0021] Obtain at least one type of insurance of the above target user and the premium amount corresponding to the at least one type of insurance to determine the type of rights and interests of the above target user to obtain the above insurance features;

[0022] The predicted value of the elderly care service information of the above target user is the membership level of the above elderly care service.

[0023] In an exemplary embodiment of the present disclosure, based on the foregoing solution, the above user characteristics include one or more of age, gender, education level, occupation, and consumption level.

[0024] In an exemplary embodiment of the present disclosure, based on the foregoing solution, the above prediction model is a backpropagation neural network model.

[0025] According to a second aspect of the present disclosure, there is provided a prediction device for elderly care service information, the device including:

[0026] A feature acquisition module; configured to: acquire the insurance features and user features of a target user;

[0027] A feature input module; configured to: input the above insurance features and the above user features into the input layer of the prediction model;

[0028] A feature processing module; configured to: perform feature fusion processing on the above insurance features and the above user features through a first weight matrix between the above input layer and the hidden layer of the prediction model to obtain fused features;

[0029] An information prediction module; configured to: perform prediction processing on the above fused features through a second weight matrix between the above hidden layer and the output layer of the prediction model to obtain a prediction value of the elderly care service information about the above target user, and display the identifier of the above target user and the prediction value of the elderly care service information of the above target user.

[0030] According to a third aspect of the present disclosure, there is provided a computer storage medium having a computer program stored thereon, where the computer program, when executed by a processor, implements the prediction method of the elderly care service information described in the first aspect, and the computer program, when executed by a processor, implements the prediction method of the elderly care service information described in the second aspect.

[0031] According to a fourth aspect of the present disclosure, there is provided an electronic device, including: a processor; and a memory for storing executable instructions of the processor; wherein, the processor is configured to execute the prediction method of the elderly care service information described in the first aspect by executing the executable instructions, and the processor is configured to execute the prediction method of the elderly care service information described in the second aspect by executing the executable instructions.

[0032] As can be seen from the above technical solutions, the prediction method of the elderly care service information, the prediction device of the elderly care service information, the computer storage medium, and the electronic device in the exemplary embodiments of the present disclosure at least have the following advantages and positive effects:

[0033] In the technical solutions provided by some embodiments of the present disclosure, on the one hand, the present technical solution realizes prediction through a machine learning model obtained by big data, which is beneficial to improving the prediction accuracy. On the other hand, displaying the identifier of the target user and the predicted value of the elderly care service information of the target user is beneficial for the user to conveniently view the prediction result and improve the convenience; the predicted value and the corresponding user identifier can also be sent to a remote device to further improve the convenience of obtaining the predicted value of the target user. On yet another hand, the elderly care service information of the target user is predicted based on the insurance characteristics and user characteristics of the target user. It can be seen that in the present technical solution, by appropriately analyzing the existing data converted from insurance customers to elderly care customers, the membership intention and membership level of potential elderly care customers in the future when they move into an elderly care community can be effectively estimated, so as to better provide dual value-added services of insurance and elderly care for customers.

[0034] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0036] Figure 1 Schematic diagram of the system architecture for implementing the prediction method of elderly care service information in an exemplary embodiment of the present disclosure;

[0037] Figure 2 Schematic flowchart of the prediction method of elderly care service information according to an embodiment of the present disclosure;

[0038] Figure 3 Schematic flowchart of the method for obtaining a prediction model according to an embodiment of the present disclosure;

[0039] Figure 4 Schematic diagram of the structure of a prediction model according to an embodiment of the present disclosure;

[0040] Figure 5 Schematic flowchart of the method for obtaining a model and making a prediction according to an embodiment of the present disclosure;

[0041] Figure 6 Schematic diagram of the structure of a prediction device for elderly care service information according to an embodiment of the present disclosure;

[0042] Figure 7A schematic structural diagram of a computer storage medium in an exemplary embodiment of the present disclosure is shown; and,

[0043] Figure 8 A schematic structural diagram of an electronic device in an exemplary embodiment of the present disclosure is shown. Detailed implementation manners

[0044] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art.

[0045] In addition, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that the technical solutions of the present disclosure can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be used. In other cases, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the present disclosure.

[0046] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices. In addition, the drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities.

[0047] The flowcharts shown in the drawings are only exemplary illustrations and do not necessarily include all the contents and operations / steps, nor do they necessarily have to be executed in the described order. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may be changed according to the actual situation. In this specification, the terms "a", "one", "the", and "said" are used to indicate the presence of one or more elements / components / etc.; the terms "comprising" and "having" are used to mean an open inclusion and mean that there may be additional elements / components / etc. in addition to the listed elements / components / etc.; the terms "first" and "second", etc. are only used as labels and are not a limitation on the quantity of their objects.

[0048] In this exemplary embodiment, a system architecture for implementing a prediction method for elderly care service information is first provided, which can be applied to various data processing scenarios. Refer to Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0049] Users can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send request instructions, etc. Various communication client applications may be installed on the terminal devices 101, 102, 103, such as picture processing applications, shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0050] The terminal devices 101, 102, 103 may be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop computers, and desktop computers, etc.

[0051] In an exemplary embodiment, the terminal devices 101, 102, 103 send the insurance characteristics and user characteristics of the target user to the server 105. Thus, the server 105 inputs the received insurance characteristics and user characteristics of the target user into the input layer of the prediction model. Then, the server 105 performs feature fusion processing on the above-mentioned insurance characteristics and user characteristics through the first weight matrix between the input layer and the hidden layer of the prediction model to obtain fusion characteristics. Further, the server 105 performs prediction processing on the above-mentioned fusion characteristics through the second weight matrix between the hidden layer and the output layer of the prediction model, thereby obtaining a prediction value of the elderly care service information about the above-mentioned target user, and displaying the identifier of the above-mentioned target user and the prediction value of the elderly care service information of the above-mentioned target user.

[0052] In an exemplary embodiment, the server 105 may also send the prediction value of the elderly care service information about the above-mentioned target user to the terminal devices 101, 102, 103, so that the terminal devices 101, 102, 103 can remotely and conveniently view the prediction information of the elderly care service information of the relevant target user.

[0053] Figure 2 Shows a schematic flowchart of a method for predicting elderly care service information according to an embodiment of the present disclosure. Refer to Figure 2 , the method provided in this embodiment includes:

[0054] Step S210, obtain the insurance features and user features of the target user;

[0055] Step S220, input the insurance features and the user features into the input layer of the prediction model;

[0056] Step S230, perform feature fusion processing on the insurance features and the user features through a first weight matrix between the input layer and the hidden layer of the prediction model to obtain fused features; and,

[0057] Step S240, perform prediction processing on the fused features through a second weight matrix between the hidden layer and the output layer of the prediction model to obtain a predicted value of the elderly care service information about the target user, and display the identifier of the target user and the predicted value of the elderly care service information of the target user.

[0058] In Figure 2 In the technical solution provided by the illustrated embodiment, on the one hand, the prediction of this technical solution is realized through a machine learning model obtained by big data, which is beneficial to improving the prediction accuracy. On the other hand, displaying the identifier of the target user and the predicted value of the elderly care service information of the target user is beneficial for the user to conveniently view the prediction result and improve convenience; the predicted value and the corresponding user identifier can also be sent to a remote device to further improve the convenience of obtaining the predicted value of the target user. On the other hand, the elderly care service information of the target user is predicted based on the insurance features and user features of the target user. It can be seen that in this technical solution, by appropriately analyzing the data of existing insurance customers converted into elderly care customers, the membership willingness and membership level of potential elderly care customers in the future when they move into an elderly care community can be effectively estimated, so as to better provide dual value-added services of insurance and elderly care for customers.

[0059] The following Figure 2 elaborates in detail on the specific implementation manners of each step in the illustrated embodiment:

[0060] In an exemplary embodiment, the acquisition process of the prediction model in this technical solution is first introduced. Specifically, Figure 3 shows a flowchart of a method for obtaining a prediction model according to an embodiment of the present disclosure. Refer to Figure 3 This embodiment provides a method that includes:

[0061] Step S310, obtain the insurance features, user features, and actual elderly care service information of historical users as a set of samples to obtain a sample set.

[0062] In an exemplary embodiment, the technical solution analyzes the characteristics of pension community customers from insurance customers, and extracts the relevant characteristics of this part of customers as a sample set. Specifically: for each pension community customer from insurance customers, extract their insurance characteristics, user characteristics, and actual pension service information as a set of samples for this customer (exemplarily, the user identification of this customer can be used as the identification of this set of samples); further, after vectorizing each set of samples, a sample set is obtained.

[0063] In an exemplary embodiment, the user characteristics in each set of samples include one or more of: age, gender, education level, occupation, consumption level. Exemplarily, the determination method of user characteristics can be: determined according to the basic information registered and collected when the user logs in to platform software such as instant messaging software, financial management platform, or shopping software. For example, information such as age, gender, education level, occupation, political status, etc.

[0064] In an exemplary embodiment, the acquisition method of insurance characteristics in each set of samples can be: obtain at least one insurance type of pension community customers from insurance customers and the premium amount corresponding to the at least one insurance type, so as to determine the rights and interests type of this customer according to the insurance type and premium amount, and obtain its insurance characteristics.

[0065] In an exemplary embodiment, the prediction target of the prediction model can be the membership level of pension services. For example, the actual pension service information in each of the above sets of samples includes: the membership level of the pension service of pension community customer A from insurance customers is level one, and the membership level of the pension service of pension community customer B from insurance customers is level four, etc.

[0066] In the technical solution provided in this embodiment, for pension community customers from insurance customers, extract the insurance type and the premium amount corresponding to each insurance type, so as to determine the insurance characteristics of each customer, then combine the user characteristics of each customer, and match the membership system of each customer in the pension community to determine the above sample set.

[0067] In an exemplary embodiment, for the sample set determined in step S310, the above sample set can be randomly divided into a first part of samples and a second part of samples. Among them, the first part of samples is used to obtain the prediction model, and the second part of samples is used to verify whether the prediction model in the acquisition process reaches the preset prediction index. Exemplarily, for a prediction model that reaches the preset test index, it can be used to perform relevant base predictions on the target users in the Figure 2 shown embodiment, and it can be realized to guide the industry on how to provide more value-added services (such as pension services, etc.) for subsequent insurance customers.

[0068] In an exemplary embodiment, continue to refer to Figure 3, in step S320: obtaining the prediction model by using the first part of samples in the sample set; and, in step S330: testing the prediction model by using the second part of samples in the sample set so that the prediction model meets a preset test index.

[0069] In this embodiment, the above prediction model adopts a backpropagation neural network model. Specifically, Figure 4 shows a schematic structural diagram of a prediction model according to an embodiment of the present disclosure, Figure 5 shows a schematic flowchart of a model acquisition and prediction method according to an embodiment of the present disclosure. The following embodiments will be combined with Figure 4 and Figure 5 to explain the specific implementation manner of the acquisition process in step S320:

[0070] In an exemplary embodiment, referring to Figure 4 and Figure 5 , the insurance features and user features (i.e., feature 40) of each group of samples in the above first part of samples are input into the prediction model 400 to obtain predicted elderly care service information, a first intermediate weight matrix 51 between the input layer 41 and the hidden layer 42, and a second intermediate weight matrix 52 between the hidden layer 42 and the output layer 43.

[0071] Exemplarily, first, a group of inputs x1, x2,..., x p come to the input layer 41, and then a group of data S1, S2,..., S n is generated through the connection weight 51 (i.e., the above first intermediate weight matrix) with the hidden layer 42 and used as the input of the hidden layer 42. Then, after passing through the activation function f() of the nodes in the hidden layer 42, it is converted into f(S j ), where j takes values in [1, n], indicating the output generated by the jth node in the hidden layer 42. Further, these outputs will generate the input of the output layer 43 through the connection weight 52 (i.e., the above second intermediate weight matrix) between the hidden layer 42 and the output layer 43.

[0072] Exemplarily, the processing process of the output layer 43 is similar to that of the hidden layer 42. Finally, outputs y1, y2,..., y q will be generated in the output layer 43. Thus, the forward propagation process of the backpropagation model is completed. The following embodiments will explain the backpropagation of errors, where the purpose of the backpropagation of errors is to adjust the above first intermediate weight matrix and the second intermediate weight matrix.

[0073] Exemplarily, the first intermediate weight matrix is iteratively calculated using the actual elderly care service information in each group of first - part samples and the predicted elderly care service information obtained from the first - part samples of this group to obtain a first weight matrix that meets the preset error threshold, and the second intermediate weight matrix is iteratively calculated to obtain a second weight matrix that meets the preset error threshold.

[0074] Reference Figure 5 , during the backpropagation of errors, for a given first - part sample, on the one hand, its actual elderly care service information is already determined. Exemplarily, it is y’1, y’2, …, y’ q . On the other hand, the insurance characteristics and user characteristics of this group of first - part samples are predicted by the output values of the neural network (i.e., the predicted elderly care service information y1, y2, …, y q ), and there will be an error when compared with the above - mentioned actual elderly care service information. Exemplarily, the error is: e1 = y’1 - y1, e2 = y’2 - y2, …, e q = y’ q - y q . During the model acquisition process, the above - mentioned errors e1, e2, …, e q are backpropagated in the prediction model to iteratively calculate the first intermediate weight matrix and the second intermediate weight matrix. Thus, the smaller the above - mentioned errors e1, e2, …, e q , the closer the output result of the prediction model is to the actual result, that is to say, the higher the prediction accuracy of the prediction model. And finally, the iterative result of the first intermediate weight matrix is determined as the first weight matrix, and the iterative result of the second intermediate weight matrix is determined as the second weight matrix.

[0075] In an exemplary embodiment, the membership levels of the elderly care service community can be divided into 4 levels. Exemplarily, each level is numbered from 1 to 4. Then, the output of the prediction model is used with a normalization function (such as the softmax function), so as to assign probabilities to different objects to increase the accuracy of the final decision for the probabilities of possible membership levels. For example, for a certain target user K, in the first case of the prediction model, the probability of belonging to membership level 1 is 80%, the probability of belonging to membership level 2 is 15%, and then smaller values are given to the probabilities of other membership levels.

[0076] In an exemplary embodiment, the technical solution of the present invention can be implemented based on an open - source software library (tensorflow). The prediction model is implemented and obtained through steps such as data - set preparation, setting placeholders, constructing a graph, initializing weight and bias variables, minimizing loss using gradient descent, and providing feedback to the graph.

[0077] The following embodiments are combined with Figure 4 andFigure 5 An explanation of the specific implementation of the testing process in step S330 will be given as follows:

[0078] In an exemplary embodiment, in step S330, the second part of the sample is used to test the prediction model so that the prediction model meets at least one of the following test metrics: accuracy, recall rate, and AUC (a model evaluation metric specifically used to evaluate the prediction value of a model; abbreviated as Area Under Curve).

[0079] Exemplarily, the second part of the sample is used to test the prediction model (denoted as the model to be tested), and at least one test metric is used to verify the test results of the model to be tested, and the prediction model that meets the test metrics is Figure 2 used to predict the elderly care service information in the illustrated embodiment to determine the membership level of the elderly care service for the target user in step S210. Among them, the target user should be a user with insurance characteristics, that is, the existing insurance customers are predicted to determine the possibility of converting these insurance customers into elderly care community customers.

[0080] In an exemplary embodiment, the specific way to test the model to be tested can be:

[0081] First, based on the second part of the sample and the output data after substituting the second part of the sample into the model to be tested, the following are obtained: true positive (TP), true negative (TN), false negative (FN), and false positive (FP). Among them, TP is the number of positive classes in the second part of the sample that are still judged as positive classes by the model to be tested, TN is the number of negative classes in the second part of the sample that are still judged as negative classes by the model to be tested, FN is the number of negative classes in the second part of the sample that are judged as positive classes by the model to be tested, and FP is the number of positive classes in the second part of the sample that are judged as negative classes by the model to be tested. Positive classes and negative classes refer to two categories manually labeled for the first part of the sample, that is, if a sample is manually labeled as belonging to a specific class, then the sample belongs to the positive class, and the sample that does not belong to the specific class belongs to the negative class.

[0082] Second, calculate the test results of the model to be tested based on true positive (TP), true negative (TN), false negative (FN), and false positive (FP).

[0083] In an exemplary embodiment, the test metrics will be introduced taking accuracy and recall rate as examples. Specifically:

[0084] Calculate the accuracy p and recall rate r according to Formula 1 and Formula 2 respectively;

[0085] p = TP / (TP + FP) Formula 1

[0086] r = TP / (TP + FN), Formula 2.

[0087] If the test results obtained after testing x models to be tested are: accuracy test results p1, p2,..., px, and recall test results r1, r2,..., rx.

[0088] The set conditions corresponding to the test metrics are: if the accuracy test result is greater than p', it meets the accuracy set condition, otherwise it does not meet the accuracy set condition, and if the recall test result is greater than r', it meets the recall set condition, otherwise it does not meet the recall set condition.

[0089] In an exemplary embodiment, if the test results meet the set conditions corresponding to the test metrics, the model to be tested can be used as a prediction model for determining the pension service information of the above target users; if the test results do not meet the set conditions, the above model to be tested continues to iterate until the test results of the model to be tested meet the set conditions.

[0090] In an exemplary embodiment, when determining whether the test results meet the set conditions corresponding to the test metrics, it can be based only on accuracy or recall as the test metric, that is, the accuracy / recall meets the set conditions; it can also be based on both accuracy and recall as the test metrics at the same time, that is, both the accuracy and recall meet the set conditions.

[0091] It should be noted that the specific test method is formulated according to actual needs, and it is not limited to using the above accuracy and / or recall as the test metrics for testing.

[0092] In an exemplary embodiment, the test metric can also be AUC. Specifically:

[0093] In an exemplary embodiment, the false positive rate FPR and the true positive rate TPR are determined using Formula 3 and Formula 4.

[0094] FPR = FP / (FP + TN), Formula 3.

[0095] TPR = TP / (TP + FN), Formula 4.

[0096] Further, with the FPR as the abscissa and the TPR as the ordinate, a Receiver Operating Characteristic curve (ROC curve for short) is plotted. Among them, the ROC curve is the characteristic curve of each obtained index, used to display the relationship between each index, and further calculate the area under the ROC curve, AUC. The ROC curve is the characteristic curve of each obtained index, used to display the relationship between each index. AUC is the area under the ROC curve. The larger the AUC, the higher the prediction value of the model. Furthermore, the model to be tested can be tested through the AUC. And the model with the largest AUC value in the evaluation result is used as the prediction model to determine the membership level of the elderly care service for the target user in step S210.

[0097] In the technical solution provided by the above embodiment, on the one hand, the elderly care service information of the target user is predicted based on the insurance characteristics and user characteristics of the target user. It can be seen that in this technical solution, by appropriately analyzing the data of existing insurance customers converted into elderly care customers, the membership willingness and membership level of potential elderly care customers in the future when they move into the elderly care community can be effectively estimated, so as to better provide dual value-added services of insurance and elderly care for customers. On the other hand, the prediction in this technical solution is realized through a machine learning model obtained by big data, which is beneficial to improving the prediction accuracy.

[0098] Those skilled in the art can understand that all or part of the steps to implement the above implementation manner are implemented as a computer program executed by a GPU or a CPU. For example, the process of obtaining the above prediction model is implemented through a GPU, or the process of predicting the membership level of the elderly care service for the target user through the obtained prediction model by a CPU. When the computer program is executed by a GPU or a CPU, it executes the above functions defined by the above method provided by the present disclosure. The program can be stored in a computer-readable storage medium, and the storage medium can be a read-only memory, a disk, an optical disc, etc.

[0099] In addition, it should be noted that the above drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, rather than for limiting purposes. It is easy to understand that the processes shown in the above drawings do not indicate or limit the time sequence of these processes. Additionally, it is also easy to understand that these processes can be executed synchronously or asynchronously in, for example, multiple modules.

[0100] The following introduces the device embodiment of the present disclosure, which can be used to execute the above-mentioned recommended method for driving routes of the present disclosure.

[0101] Figure 6 The structural schematic diagram of the prediction device for elderly care service information in the exemplary embodiment of the present disclosure is shown. As Figure 6As shown in the figure, the prediction device 600 for the above-mentioned elderly care service information includes: an acquisition module 601, a feature input module 602, a feature processing module 603, and an information prediction module 604. Among them:

[0102] The above-mentioned feature acquisition module 601 is configured to: acquire the insurance features and user features of the target user;

[0103] The above-mentioned feature input module 602 is configured to: input the above-mentioned insurance features and the above-mentioned user features into the input layer of the prediction model;

[0104] The above-mentioned feature processing module 603 is configured to: perform feature fusion processing on the above-mentioned insurance features and the above-mentioned user features through a first weight matrix between the above-mentioned input layer and the hidden layer of the above-mentioned prediction model to obtain fused features;

[0105] The above-mentioned information prediction module 604 is configured to: perform prediction processing on the above-mentioned fused features through a second weight matrix between the above-mentioned hidden layer and the output layer of the above-mentioned prediction model to obtain a prediction value of the elderly care service information about the above-mentioned target user, and display the identifier of the above-mentioned target user and the prediction value of the elderly care service information of the above-mentioned target user.

[0106] In an exemplary embodiment, based on the foregoing solution, the prediction device 600 for the above-mentioned elderly care service information includes: a model acquisition module. Among them:

[0107] The above-mentioned model acquisition module is configured to: before the above-mentioned feature input module 602 inputs the above-mentioned insurance features and the above-mentioned user features into the input layer of the prediction model, acquire the insurance features, user features, and actual elderly care service information of historical users as a set of samples to obtain a sample set; and acquire the above-mentioned prediction model through the first part of the samples in the above-mentioned sample set.

[0108] In an exemplary embodiment, based on the foregoing solution, the above-mentioned model acquisition module is specifically configured to:

[0109] Input the insurance features and user features of each group of samples in the above-mentioned first part of the samples into the above-mentioned prediction model to obtain predicted elderly care service information, a first intermediate weight matrix between the above-mentioned input layer and the above-mentioned hidden layer, and a second intermediate weight matrix between the above-mentioned hidden layer and the above-mentioned output layer; and perform iterative calculation on the above-mentioned first intermediate weight matrix through the above-mentioned actual elderly care service information and the above-mentioned predicted elderly care service information to obtain a first weight matrix that meets a preset error threshold, and perform iterative calculation on the above-mentioned second intermediate weight matrix to obtain a second weight matrix that meets the above-mentioned preset error threshold.

[0110] In an exemplary embodiment, based on the foregoing solution, the prediction device 600 for the above-mentioned elderly care service information includes: a model testing module. Among them:

[0111] The above model testing module is configured to: after the above model acquisition module obtains the above prediction model through the first part of the samples in the above sample set, test the above prediction model through the second part of the samples in the above sample set, so that the prediction model meets the preset test metrics.

[0112] Wherein, the above test metrics include at least one of accuracy rate, recall rate, and AUC.

[0113] In an exemplary embodiment, based on the foregoing solution, the feature acquisition module 601 is specifically configured to: obtain at least one type of insurance coverage of the above target user and the premium amount corresponding to the at least one type of insurance coverage to determine the rights and interests type of the above target user, and obtain the above insurance features;

[0114] The predicted value of the above old-age service information of the above target user is the membership level of the above old-age service.

[0115] In an exemplary embodiment, based on the foregoing solution, the above user features include one or more of age, gender, education level, occupation, and consumption level.

[0116] In an exemplary embodiment, based on the foregoing solution, the above prediction model is a backpropagation neural network model.

[0117] The specific details of each module in the above prediction device for old-age service information have been described in detail in the above embodiments of the prediction method for old-age service information, and thus will not be elaborated here.

[0118] It should be noted that although several modules or units of a device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0119] In addition, although the steps of the methods in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0120] Based on the descriptions of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (such as a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0121] In an exemplary embodiment of the present disclosure, there is also provided a computer storage medium capable of implementing the above method. A program product capable of implementing the method described in this specification is stored thereon. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program code. When the above program product runs on a terminal device, the above program code is used to enable the above terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section of this specification.

[0122] Refer to Figure 7 As shown, a program product 700 for implementing the above method according to an embodiment of the present disclosure is described. It can be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device.

[0123] The above program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0124] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, carrying readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable signal medium may also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0125] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0126] The program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0127] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

[0128] Those skilled in the art can understand that various aspects of the present disclosure can be implemented as a system, method, or program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuitry", "module", or "system" here.

[0129] The following refers to Figure 8 to describe the electronic device 800 according to such an embodiment of the present disclosure. Figure 8 The illustrated electronic device 800 is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of the present disclosure.

[0130] As Figure 8As shown, the electronic device 800 is presented in the form of a general computing device. The components of the electronic device 800 may include, but are not limited to: at least one of the above-mentioned processing units 810, at least one of the above-mentioned storage units 820, and a bus 830 that connects different system components (including the storage unit 820 and the processing unit 810).

[0131] Among them, the above storage unit stores program code, and the above program code can be executed by the above processing unit 810, so that the above processing unit 810 executes the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section of this specification. For example, the above processing unit 810 can execute as Figure 2 shown in: Step S210, obtaining the insurance characteristics and user characteristics of the target user; Step S220, inputting the insurance characteristics and the user characteristics into the input layer of the prediction model; Step S230, performing feature fusion processing on the insurance characteristics and the user characteristics through the first weight matrix between the input layer and the hidden layer of the prediction model to obtain fused features; and Step S240, performing prediction processing on the fused features through the second weight matrix between the hidden layer and the output layer of the prediction model to obtain a predicted value of the elderly care service information about the target user, and displaying the identifier of the target user and the predicted value of the elderly care service information of the target user.

[0132] Exemplarily, the above processing unit 810 can also execute any steps of the embodiments as Figure 3 shown.

[0133] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 8201 and / or a cache storage unit 8202, and may further include a read-only storage unit (ROM) 8203.

[0134] The storage unit 820 may further include a program / utility 8204 having a set (at least one) of program modules 8205. Such program modules 8205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0135] The bus 830 may represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any bus structure in a variety of bus structures.

[0136] The electronic device 800 can also communicate with one or more external devices 900 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication can be carried out through the input / output (I / O) interface 850. Moreover, the electronic device 800 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 880. As shown in the figure, the network adapter 860 communicates with other modules of the electronic device 800 through the bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0137] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or can be implemented by the way of software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0138] In addition, the above drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, rather than for limiting purposes. It is easy to understand that the processes shown in the above drawings do not indicate or limit the time sequence of these processes. Additionally, it is also easy to understand that these processes can be executed synchronously or asynchronously in, for example, multiple modules.

[0139] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include well-known common general knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.

Claims

1. A prediction method for elderly care service information, characterized in that The method includes: Obtaining the insurance features and user features of the target user; the insurance features refer to the existing insurance types of the user and the premium amounts corresponding to the insurance types, and the user features include one or more of age, gender, education level, occupation, and consumption level; Inputting the insurance features and the user features into the input layer of the prediction model; Performing feature fusion processing on the insurance features and the user features through a first weight matrix between the input layer and the hidden layer of the prediction model to obtain fused features; Performing prediction processing on the fused features through a second weight matrix between the hidden layer and the output layer of the prediction model to obtain a predicted value of the elderly care service information about the target user, and displaying the identifier of the target user and the predicted value of the elderly care service information of the target user; Wherein, before inputting the insurance features and the user features into the input layer of the prediction model, the method further includes: Obtaining the insurance features, user features, and actual elderly care service information of historical users as a set of samples to obtain a sample set; Obtaining the prediction model through the first part of the samples in the sample set, wherein inputting the insurance features and user features of each group of samples in the first part of the samples into the prediction model to obtain predicted elderly care service information, a first intermediate weight matrix between the input layer and the hidden layer, and a second intermediate weight matrix between the hidden layer and the output layer; performing iterative calculation on the first intermediate weight matrix through the actual elderly care service information and the predicted elderly care service information to obtain a first weight matrix that meets a preset error threshold, and performing iterative calculation on the second intermediate weight matrix to obtain a second weight matrix that meets the preset error threshold; The obtaining of the insurance features of the target user includes: Obtaining at least one insurance type of the target user, and obtaining the premium amount corresponding to the at least one insurance type to determine the rights and interests type of the target user, so as to obtain the insurance features; The predicted value of the elderly care service information of the target user is the membership level of the elderly care service.

2. The prediction method of elderly care service information according to claim 1, characterized in that, After obtaining the prediction model through the first part of the samples in the sample set, the method further includes: Testing the prediction model through the second part of the samples in the sample set to make the prediction model meet preset test metrics, where the test metrics include at least one of accuracy rate, recall rate, and AUC.

3. The prediction method of elderly care service information according to any one of claims 1 to 2, characterized in that The prediction model is a backpropagation neural network model.

4. A prediction device for elderly care service information, characterized in that, The device includes: A feature acquisition module; configured to: obtain the insurance features and user features of the target user; the insurance features refer to the existing insurance types of the user and the premium amounts corresponding to the insurance types, and the user features include one or more of age, gender, education level, occupation, and consumption level; A feature input module; configured to: input the insurance features and the user features into the input layer of the prediction model; A feature processing module; configured to: perform feature fusion processing on the insurance features and the user features through a first weight matrix between the input layer and the hidden layer of the prediction model to obtain fused features; An information prediction module; configured to: perform prediction processing on the fused features through a second weight matrix between the hidden layer and the output layer of the prediction model to obtain a predicted value of the elderly care service information about the target user, and display the identifier of the target user and the predicted value of the elderly care service information of the target user; A model acquisition module, configured to: before the feature input module inputs the above-mentioned insurance features and the above-mentioned user features into the input layer of the prediction model, acquire the insurance features, user features, and actual elderly care service information of historical users as a set of samples to obtain a sample set; and, obtain the above-mentioned prediction model through the first part of the samples in the sample set, wherein the insurance features and user features of each group of samples in the first part of the samples are input into the prediction model to obtain predicted elderly care service information, a first intermediate weight matrix between the input layer and the hidden layer, and a second intermediate weight matrix between the hidden layer and the output layer; perform iterative calculation on the first intermediate weight matrix through the actual elderly care service information and the predicted elderly care service information to obtain a first weight matrix that meets a preset error threshold, and perform iterative calculation on the second intermediate weight matrix to obtain a second weight matrix that meets the preset error threshold; Wherein, acquiring the insurance features of the target user includes: acquiring at least one type of insurance of the target user and the premium amount corresponding to the at least one type of insurance to determine the type of rights and interests of the target user to obtain the insurance features; the predicted value of the elderly care service information of the target user is the membership level of the elderly care service.

5. A computer storage medium, on which a computer program is stored, characterized in that, When the computer program is executed by a processor, it implements the prediction method of the elderly care service information according to any one of claims 1 to 3.

6. An electronic device, characterized in that, Including: A processor; And A memory for storing executable instructions of the processor; Wherein, the processor is configured to execute the prediction method of the elderly care service information according to any one of claims 1 to 3 by executing the executable instructions.

Images