A method, apparatus, electronic device, and storage medium for determining a service mode

Abstract

An embodiment of the present invention provides a service mode determination method, device, electronic device, and storage medium, which can obtain object information of an object to be served; the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served; perform mapping processing on the object information of the object to be served to obtain a feature vector of the object to be served as a first feature vector; input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model; determine the service mode corresponding to the service level of the object to be served in the preset correspondence between the service level and the service mode. Based on the above processing, the influence of personal subjective factors can be excluded, the accuracy of the determined service mode can be improved, the service quality can be improved, and the time cost and labor cost can be reduced, and the efficiency of determining the service mode can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a method, apparatus, electronic device, and storage medium for determining a service mode. Background Art

[0002] With the increasing aging of the population, the elderly care service problem has attracted much social attention. In related technologies, in order to better provide elderly care services, a questionnaire survey is conducted on the elderly manually, and the questionnaire survey results are evaluated manually to determine the service mode of the elderly. For example, the health status of the elderly is obtained through a questionnaire survey, and the cycle of health examinations for the elderly is determined by evaluating the health status of the elderly.

[0003] However, in the above process, the questionnaire survey results are evaluated manually to determine the service mode of the elderly. Due to the influence of personal subjective factors, the determined service mode may not meet the service needs of the elderly, which will reduce the service quality. Moreover, evaluating the questionnaire survey results manually requires a large amount of time cost and labor cost, resulting in low efficiency in determining the service mode. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a method, apparatus, electronic device, and storage medium for determining a service mode, so as to eliminate the influence of personal subjective factors, improve the accuracy of the determined service mode, improve the service quality, and can reduce the time cost and labor cost, and improve the efficiency of determining the service mode. The specific technical solutions are as follows:

[0005] In the first aspect of the present invention, a method for determining a service mode is provided, and the method includes:

[0006] Obtain the object information of the object to be served; wherein, the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served;

[0007] Perform mapping processing on the object information of the object to be served to obtain a feature vector of the object to be served as the first feature vector;

[0008] Input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model; wherein, the level prediction model is trained based on the sample object information and sample service level of the sample object;

[0009] Determine the service mode corresponding to the service level of the object to be served in the preset correspondence between the service level and the service mode.

[0010] Optionally, the level prediction model includes: a linear module, a non-linear module, a first multi-layer perceptron, a second multi-layer perceptron, and an output module;

[0011] Inputting the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model includes:

[0012] Performing a linear transformation on the first feature vector through the linear module to obtain a second feature vector;

[0013] Performing a non-linear transformation on the first feature vector through the non-linear module to obtain a third feature vector;

[0014] Performing a linear transformation on the third feature vector through the first multi-layer perceptron to obtain a fourth feature vector;

[0015] Performing feature fusion on the second feature vector and the fourth feature vector to obtain a fifth feature vector;

[0016] Performing a linear transformation on the fifth feature vector through the second multi-layer perceptron to obtain a first numerical value;

[0017] Performing normalization processing on the first numerical value through the output module to obtain a second numerical value representing the service level of the object to be served.

[0018] Optionally, the personal basic information of the object to be served includes at least one of the following: the age, gender, marital status, and number of children of the object to be served;

[0019] The physical health information of the object to be served includes at least one of the following: the height, weight, and current disease information of the object to be served;

[0020] The living environment information of the object to be served includes at least one of the following: the number of dining places and the number of activity centers within a preset geographical range of the residence of the object to be served;

[0021] The social activity information of the object to be served includes at least one of the following: the types and frequencies of community activities participated by the object to be served during a historical time period.

[0022] Optionally, performing a mapping process on the object information of the object to be served to obtain a feature vector of the object to be served as the first feature vector includes:

[0023] Determining the numerical information and non-numerical information in the object information of the object to be served; wherein, the numerical information is information represented by numerical values; the non-numerical information is other information except the numerical information;

[0024] Encode the non - numerical information to obtain a sixth feature vector;

[0025] Normalize the numerical information and perform a linear transformation on the normalized numerical information to obtain a seventh feature vector;

[0026] Concatenate the sixth feature vector and the seventh feature vector to obtain a feature vector of the object to be served, which is used as the first feature vector.

[0027] Optionally, the training process of the level prediction model includes the following steps:

[0028] Obtain the sample object information of the sample object and the sample value representing the sample service level of the sample object;

[0029] Perform a mapping process on the sample object information to obtain a feature vector of the sample object, which is used as the sample feature vector;

[0030] Use the sample feature vector as the input data of the level prediction model with the initial structure, and use the sample value representing the sample service level of the sample object as the output data of the level prediction model with the initial structure. Adjust the model parameters of the level prediction model with the initial structure until the level prediction model with the initial structure reaches a preset convergence condition to obtain a trained level prediction model.

[0031] In the second aspect of the implementation of the present invention, a service mode determination device is provided. The device includes:

[0032] An acquisition module for acquiring the object information of the object to be served; where the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served;

[0033] A mapping module for performing a mapping process on the object information of the object to be served to obtain a feature vector of the object to be served, which is used as the first feature vector;

[0034] A prediction module for inputting the first feature vector into a pre - trained level prediction model to obtain the service level of the object to be served output by the level prediction model; where the level prediction model is trained based on the sample object information and sample service level of the sample object;

[0035] A determination module for determining the service mode corresponding to the service level of the object to be served in the preset correspondence between the service level and the service mode.

[0036] Optionally, the level prediction model includes: a linear module, a non-linear module, a first multi-layer perceptron, a second multi-layer perceptron, and an output module;

[0037] The prediction module is specifically configured to perform a linear transformation on the first feature vector through the linear module to obtain a second feature vector;

[0038] Perform a non-linear transformation on the first feature vector through the non-linear module to obtain a third feature vector;

[0039] Perform a linear transformation on the third feature vector through the first multi-layer perceptron to obtain a fourth feature vector;

[0040] Perform feature fusion on the second feature vector and the fourth feature vector to obtain a fifth feature vector;

[0041] Perform a linear transformation on the fifth feature vector through the second multi-layer perceptron to obtain a first value;

[0042] Perform a normalization process on the first value through the output module to obtain a second value representing the service level of the object to be served.

[0043] Optionally, the personal basic information of the object to be served includes at least one of the following: the age, gender, marital status, and number of children of the object to be served;

[0044] The physical health information of the object to be served includes at least one of the following: the height, weight, and current disease information of the object to be served;

[0045] The living environment information of the object to be served includes at least one of the following: the number of dining places and the number of activity centers within a preset geographical range of the residence of the object to be served;

[0046] The social activity information of the object to be served includes at least one of the following: the types and frequencies of community activities participated by the object to be served within a historical time period.

[0047] Optionally, the mapping module is specifically configured to determine the numerical information and non-numerical information in the object information of the object to be served; wherein, the numerical information is information represented by a numerical value; the non-numerical information is other information except the numerical information;

[0048] Perform an encoding process on the non-numerical information to obtain a sixth feature vector;

[0049] Perform a normalization process on the numerical information, and perform a linear transformation on the normalized numerical information to obtain a seventh feature vector;

[0050] Concatenate the sixth eigenvector and the seventh eigenvector to obtain the eigenvector of the object to be served, which is used as the first eigenvector.

[0051] Optionally, the apparatus further includes a training module, configured to obtain the sample object information of the sample object and a sample value representing the sample service level of the sample object.

[0052] Perform a mapping process on the sample object information to obtain the eigenvector of the sample object, which is used as the sample eigenvector.

[0053] Use the sample eigenvector as the input data of the rank prediction model with an initial structure, and use the sample value representing the sample service level of the sample object as the output data of the rank prediction model with the initial structure. Adjust the model parameters of the rank prediction model with the initial structure until the rank prediction model with the initial structure reaches a preset convergence condition, and obtain a trained rank prediction model.

[0054] An embodiment of the present invention further provides an electronic device, including a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus.

[0055] The memory is used to store a computer program.

[0056] When the processor is configured to execute the program stored in the memory, it implements the steps of the service mode determination method described in any one of the above.

[0057] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the service mode determination method described in any one of the above.

[0058] An embodiment of the present invention further provides a computer program product including instructions. When it runs on a computer, it causes the computer to execute the service mode determination method described in any one of the above.

[0059] A service mode determination method provided by an embodiment of the present invention may obtain object information of an object to be served; the object information includes at least one of personal basic information, physical health information, living environment information, and social activity information of the object to be served; perform mapping processing on the object information of the object to be served to obtain a feature vector of the object to be served as a first feature vector; input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model; the level prediction model is trained based on sample object information and sample service levels of sample objects; in a preset correspondence between service levels and service modes, determine the service mode corresponding to the service level of the object to be served.

[0060] Based on the above processing, the level prediction model is trained based on the object information and sample service levels of sample objects, and the level prediction model can learn the mapping relationship between object information and service levels. Accordingly, the service level of the object to be served can be determined through the level prediction model and the object information of the object to be served, and then the service mode corresponding to the service level of the object to be served can be determined. It is not necessary to evaluate the questionnaire survey results of the elderly manually to determine the service mode of the elderly, which can exclude the influence of personal subjective factors, improve the accuracy of the determined service mode, improve service quality, and can reduce time costs and labor costs, and improve the efficiency of determining the service mode.

[0061] Of course, when implementing any product or method of the present invention, it is not necessarily required to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other embodiments based on these drawings.

[0063] Figure 1 It is a flowchart of a service mode determination method provided by an embodiment of the present invention;

[0064] Figure 2 It is a flowchart of another service mode determination method provided by an embodiment of the present invention;

[0065] Figure 3 It is a flowchart of another service mode determination method provided by an embodiment of the present invention;

[0066] Figure 4 It is a flowchart of another service mode determination method provided by an embodiment of the present invention;

[0067] Figure 5 Structural diagram of a service mode determination device provided by an embodiment of the present invention;

[0068] Figure 6 Structural diagram of an electronic device provided by an embodiment of the present invention. Specific implementation manners

[0069] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art based on this application belong to the protection scope of the present invention.

[0070] See Figure 1 , Figure 1 which is a flowchart of a service mode determination method provided by an embodiment of the present invention. This method can be applied to an electronic device, and the electronic device can be a terminal or a server. This method can include the following steps:

[0071] S101: Obtain the object information of the object to be served.

[0072] Among them, the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served.

[0073] S102: Perform mapping processing on the object information of the object to be served to obtain the feature vector of the object to be served as the first feature vector.

[0074] S103: Input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model.

[0075] Among them, the level prediction model is trained based on the sample object information and sample service level of the sample object.

[0076] S104: Determine the service mode corresponding to the service level of the object to be served in the preset correspondence between the service level and the service mode.

[0077] Based on the service mode determination method provided by the embodiments of the present invention, the level prediction model is trained based on the object information of the sample object and the sample service level. The level prediction model can learn the mapping relationship between the object information and the service level. Accordingly, the service level of the object to be served can be determined through the level prediction model and the object information of the object to be served, and then the service mode corresponding to the service level of the object to be served can be determined. It is not necessary to manually evaluate the questionnaire survey results of the elderly to determine the service mode of the elderly, which can exclude the influence of personal subjective factors, improve the accuracy of the determined service mode, improve the service quality, and can reduce the time cost and labor cost, and improve the efficiency of determining the service mode.

[0078] Regarding step S101, the object to be served can be the elderly, the disabled, etc. When the object to be served is the elderly, the service mode for the object to be served is the way of providing elderly care services. Elderly care services can include: physical health examination services, psychological counseling services, rehabilitation nursing services, housekeeping services, etc.

[0079] The object information of the object to be served can include at least one of the following: the personal basic information of the object to be served, the physical health information, the living environment information where the object is located, and the social activity information.

[0080] The personal basic information of the object to be served includes at least one of the following: the age, gender, marital status, and the number of children of the object to be served.

[0081] The physical health information of the object to be served includes at least one of the following: the height, weight, and current disease information of the object to be served. The current disease information can include: whether suffering from a disease, the type of the disease suffered, the time of onset, etc.

[0082] The living environment information where the object to be served is located includes at least one of the following: the number of dining places and the number of activity centers within a preset geographical range of the residence of the object to be served. The preset geographical range can be the range of the community to which the residence of the object to be served belongs.

[0083] The social activity information of the object to be served includes at least one of the following: the types and frequencies of community activities participated by the object to be served within a historical time period. The historical time period can be set by technicians according to experience. For example, the historical time period can be the most recent week from the current moment, or the historical time period can also be the most recent month from the current moment, but it is not limited thereto. The types of community activities participated can include: cultural performance activities, physical exercise activities, safety education activities, etc.

[0084] For step S102, after obtaining the object information of the object to be served, the electronic device can perform mapping processing on the object information of the object to be served to obtain the feature vector of the object to be served (i.e., the first feature vector).

[0085] In one embodiment of the present invention, based on Figure 1 , referring to Figure 2 , step S102 may include the following steps:

[0086] S1021: Determine the numerical information and non-numerical information in the object information of the object to be served.

[0087] Among them, the numerical information is the information represented by numerical values; the non-numerical information is other information except the numerical information.

[0088] S1022: Perform encoding processing on the non-numerical information to obtain the sixth feature vector.

[0089] S1023: Perform normalization processing on the numerical information, and perform linear transformation on the normalized numerical information to obtain the seventh feature vector.

[0090] S1024: Concatenate the sixth feature vector and the seventh feature vector to obtain the feature vector of the object to be served as the first feature vector.

[0091] The numerical information is the information represented by numerical values. For example, the age, height, weight, etc. of the object to be served are numerical information. The non-numerical information is other information except the numerical information. For example, the marital status, gender, disease information, etc. of the object to be served are non-numerical information.

[0092] The electronic device can determine the numerical information and non-numerical information in the object information of the object to be served. Then, according to the preset encoding method, encode the non-numerical information to convert the non-numerical information into a numerical value that can be processed by the rank prediction model to obtain the sixth feature vector. The preset encoding method can be hash encoding, or the preset encoding method can also be one-hot encoding, or the preset encoding method can also be mean encoding, etc. This embodiment does not make specific limitations. The electronic device can also perform linear transformation on the numerical information based on the preset function to obtain the seventh feature vector. The preset function can be the sigmoid function.

[0093] Furthermore, the electronic device can concatenate the sixth feature vector and the seventh feature vector to obtain the feature vector of the object to be served (i.e., the first feature vector).

[0094] For step S103, the level prediction model can be any one of an ANN (Artificial Neural Network) model, a DeepFM (Deep Factorization Machine) model, an NCF (Neural Collaborative Filtering) model, and a Wide&Deep (width and depth network for classification and regression) model.

[0095] After obtaining the first feature vector, the electronic device can input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model.

[0096] In an embodiment of the present invention, the level prediction model includes: a linear module, a non-linear module, a first multi-layer perceptron, a second multi-layer perceptron, and an output module. Correspondingly, on the Figure 1 basis, referring to Figure 3 , step S103 may include the following steps:

[0097] S1031: Perform a linear transformation on the first feature vector through the linear module to obtain a second feature vector.

[0098] S1032: Perform a non-linear transformation on the first feature vector through the non-linear module to obtain a third feature vector.

[0099] S1033: Perform a linear transformation on the third feature vector through the first multi-layer perceptron to obtain a fourth feature vector.

[0100] S1034: Perform feature fusion on the second feature vector and the fourth feature vector to obtain a fifth feature vector.

[0101] S1035: Perform a linear transformation on the fifth feature vector through the second multi-layer perceptron to obtain a first value.

[0102] S1036: Perform normalization processing on the first value through the output module to obtain a second value representing the service level of the object to be served.

[0103] The level prediction model includes: a linear module, a non-linear module, a first multi-layer perceptron, a second multi-layer perceptron, and an output module. Among them, the linear module can be an MLP (Multilayer Perceptron). The non-linear module can be a DNN (Deep Neural Networks). The DNN contains multiple fully connected layers, and each fully connected layer contains a preset activation function, which is used to perform a non-linear transformation on the feature vector input to the fully connected layer. The preset activation function can be a ReLU (Rectified Linear Unit). The output module contains a preset normalization function. For example, the output module can contain a sigmoid function, or the output module can also contain a Softmax function, but it is not limited to this.

[0104] The electronic device can perform a linear transformation on the first feature vector through the linear module to obtain a second feature vector, and perform a non-linear transformation on the first feature vector through the non-linear module to obtain a third feature vector. The electronic device can also perform a linear transformation on the third feature vector through the first multi-layer perceptron (MLP) to obtain a fourth feature vector.

[0105] Then, the electronic device can perform feature fusion on the second feature vector and the fourth feature vector to obtain a fifth feature vector. For example, the electronic device can splice the second feature vector and the fourth feature vector to obtain a fifth feature vector. Or the electronic device can also add the elements at the corresponding positions of the second feature vector and the fourth feature vector to obtain a fifth feature vector.

[0106] Furthermore, the electronic device can perform a linear transformation on the fifth feature vector through the second multi-layer perceptron to obtain a first value, and perform a normalization process on the first value through the output module to obtain a second value representing the service level of the object to be served.

[0107] The second value output by the level prediction model belongs to [0, 1], and different second values represent different service levels. The smaller the second value, the higher the corresponding service level. For example, when the second value output by the level prediction model belongs to [0.1, 0.2), it means that the service level of the object to be served is level one. When the second value output by the level prediction model belongs to [0.2, 0.3), it means that the service level of the object to be served is level two. When the second value output by the level prediction model belongs to [0.3, 0.4), it means that the service level of the object to be served is level three. And the service level of level one is higher than that of level two, and the service level of level two is higher than that of level three, and so on. Different service levels represented by different second values can be determined.

[0108] In an embodiment of the present invention, before determining the service level of the object to be served based on the level prediction model, the electronic device may also train the level prediction model with the initial structure based on a preset training sample to obtain a trained level prediction model.

[0109] Correspondingly, the training process of the level prediction model may include the following steps:

[0110] Step 1, obtain the sample object information of the sample object and the sample value representing the sample service level of the sample object.

[0111] Step 2, perform a mapping process on the sample object information to obtain the feature vector of the sample object as the sample feature vector.

[0112] Step 3, use the sample feature vector as the input data of the level prediction model with the initial structure, and use the sample value representing the sample service level of the sample object as the output data of the level prediction model with the initial structure, and adjust the model parameters of the level prediction model with the initial structure until the level prediction model with the initial structure reaches a preset convergence condition to obtain a trained level prediction model.

[0113] The electronic device may obtain the sample object information of the sample object and the sample value representing the sample service level of the sample object, and perform a mapping process on the sample object information to obtain the feature vector of the sample object (i.e., the sample feature vector). The manner in which the electronic device performs the mapping process on the object information of the sample object may refer to the relevant introduction in the foregoing embodiment.

[0114] To improve the accuracy of the trained level prediction model, the selected sample objects may include sample objects in different regions, different age groups, different physical conditions, etc., to ensure the diversity and randomness of the samples.

[0115] The electronic device may input the sample feature vector into the level prediction model with the initial structure to obtain the value representing the service level of the sample object output by the level prediction model with the initial structure (which may be referred to as the predicted value). Then, the electronic device may calculate the loss function value representing the difference between the sample value and the predicted value, and adjust the model parameters of the level prediction model with the initial structure based on the calculated loss function value until the level prediction model with the initial structure reaches a preset convergence condition to obtain a trained level prediction model.

[0116] The preset convergence condition can be that the number of training times of the level prediction model for the initial structure reaches a preset number of times. For example, the number of training times of the level prediction model for the initial structure reaches 200 times. Or, the preset condition can also be that the loss function values calculated continuously for a preset number of times are all less than a preset value. For example, the loss function values calculated continuously for 5 times are all less than 0.01.

[0117] Regarding step S104, the object to be served can be an elderly person, and the service method can be the cycle of providing elderly care services (such as physical health examination services, psychological counseling services, and housekeeping services, etc.) to the object to be served. The time interval between adjacent cycles of service methods with higher service levels is shorter.

[0118] Furthermore, after determining the service level of the object to be served, the service method corresponding to the service level of the object to be served can be determined in the preset correspondence between service levels and service methods.

[0119] For example, the service levels can include level one, level two, level three, and level four. The service method is the cycle of providing psychological counseling services to the object to be served. The preset correspondence between service levels and service objects includes: level one corresponds to providing psychological counseling services once a week; level two corresponds to providing psychological counseling services once every two weeks; level three corresponds to providing psychological counseling services once every three weeks; level four corresponds to providing psychological counseling services once every four weeks.

[0120] If the service level of the object to be served is level two, it can be determined that psychological counseling will be provided to the object to be served once every two weeks.

[0121] In an embodiment of the present invention, since the object information of the object to be served changes over time, in order to improve the accuracy of determining the service level and ensure the timeliness of the determined service method, the object information of the object to be served can be updated periodically. For example, the object information of the object to be served is re-obtained every other week.

[0122] In addition, for the object to be served whose service level has been determined by the level prediction model, the service level of the object to be served can be determined manually through a questionnaire survey. Furthermore, based on the object information of the object to be served and the service level of the object to be served determined manually again, the level prediction model can be trained again to optimize the level prediction model and improve the accuracy of the service level determined by the level prediction model.

[0123] See Figure 4 , Figure 4 which is the flowchart of another service method determination method provided by the embodiment of the present invention.

[0124] When the object to be served is an elderly person, obtain the demographic data, personal information data, health-related data, and surrounding environment data of the elderly person, that is, obtain the object information of the object to be served. The demographic data is the social activity information in the foregoing embodiment, the personal information data is the personal basic information in the foregoing embodiment, the health-related data is the physical health information in the foregoing embodiment, and the surrounding environment data is the living environment information in the foregoing embodiment.

[0125] Then, perform mapping processing on the object information of the elderly person to obtain the first feature vector of the elderly person, and input the first feature vector into a pre-trained level prediction model to obtain the service level of the elderly person output by the level prediction model. Furthermore, in the preset correspondence between the service level and the service method, determine the service method corresponding to the service level of the elderly person.

[0126] Based on the above processing, the level prediction model is trained based on the object information of the sample object and the sample service level. The level prediction model can learn the mapping relationship between the object information and the service level. Accordingly, the service level of the object to be served can be determined through the level prediction model and the object information of the object to be served, and then the service method corresponding to the service level of the object to be served can be determined. It is not necessary to manually evaluate the questionnaire survey results of the elderly person to determine the service method of the elderly person, which can reduce the time cost and labor cost, improve the efficiency of determining the service method, and can exclude the influence of personal subjective factors, improve the accuracy of the determined service method, improve the service quality, can provide targeted elderly care services for the elderly, achieve precise elderly care services, and improve the sustainability of the elderly care results.

[0127] And Figure 1 the method embodiment of Figure 5 , Figure 5 is the structural diagram of a service method determination device provided by an embodiment of the present invention. The device includes:

[0128] An acquisition module 501, configured to acquire the object information of the object to be served; wherein, the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served;

[0129] A mapping module 502, configured to perform mapping processing on the object information of the object to be served to obtain the feature vector of the object to be served as the first feature vector;

[0130] A prediction module 503, configured to input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model; wherein, the level prediction model is trained based on the sample object information and the sample service level of the sample object;

[0131] A determination module 504, configured to determine a service mode corresponding to the service level of the object to be served in a preset correspondence between service levels and service modes.

[0132] Optionally, the level prediction model includes: a linear module, a non-linear module, a first multi-layer perceptron, a second multi-layer perceptron, and an output module;

[0133] The prediction module 503 is specifically configured to perform a linear transformation on the first feature vector through the linear module to obtain a second feature vector;

[0134] Perform a non-linear transformation on the first feature vector through the non-linear module to obtain a third feature vector;

[0135] Perform a linear transformation on the third feature vector through the first multi-layer perceptron to obtain a fourth feature vector;

[0136] Perform feature fusion on the second feature vector and the fourth feature vector to obtain a fifth feature vector;

[0137] Perform a linear transformation on the fifth feature vector through the second multi-layer perceptron to obtain a first value;

[0138] Perform normalization processing on the first value through the output module to obtain a second value for representing the service level of the object to be served.

[0139] Optionally, the personal basic information of the object to be served includes at least one of the following: the age, gender, marital status, and number of children of the object to be served;

[0140] The physical health information of the object to be served includes at least one of the following: the height, weight, and current disease information of the object to be served;

[0141] The living environment information of the object to be served includes at least one of the following: the number of dining places and the number of activity centers within a preset geographical range of the residence of the object to be served;

[0142] The social activity information of the object to be served includes at least one of the following: the types and frequencies of community activities participated by the object to be served within a historical time period.

[0143] Optionally, the mapping module 502 is specifically configured to determine the numerical information and non-numerical information in the object information of the object to be served; wherein, the numerical information is information represented by numerical values; the non-numerical information is other information except the numerical information;

[0144] Encode the non-numerical information to obtain a sixth feature vector;

[0145] Normalize the numerical information and perform a linear transformation on the normalized numerical information to obtain a seventh feature vector;

[0146] Concatenate the sixth feature vector and the seventh feature vector to obtain a feature vector of the object to be served, which is used as the first feature vector.

[0147] Optionally, the device further includes a training module, configured to obtain sample object information of the sample object and a sample value representing the sample service level of the sample object;

[0148] Perform a mapping process on the sample object information to obtain a feature vector of the sample object, which is used as a sample feature vector;

[0149] Use the sample feature vector as input data of a rank prediction model with an initial structure, and use the sample value representing the sample service level of the sample object as output data of the rank prediction model with the initial structure. Adjust the model parameters of the rank prediction model with the initial structure until the rank prediction model with the initial structure reaches a preset convergence condition to obtain a trained rank prediction model.

[0150] Based on the service method determination device provided in the embodiments of the present invention, the rank prediction model is trained based on the object information and sample service level of the sample object. The rank prediction model can learn the mapping relationship between the object information and the service level. Accordingly, the service level of the object to be served can be determined through the rank prediction model and the object information of the object to be served, and then the service method corresponding to the service level of the object to be served can be determined. It is not necessary to manually evaluate the questionnaire survey results of the elderly to determine the service method of the elderly, which can exclude the influence of personal subjective factors, improve the accuracy of the determined service method, improve the service quality, and can reduce the time cost and labor cost, and improve the efficiency of determining the service method.

[0151] Embodiments of the present invention further provide an electronic device, as Figure 6 shown, including a processor 601, a communication interface 602, a memory 603, and a communication bus 604. Among them, the processor 601, the communication interface 602, and the memory 603 communicate with each other through the communication bus 604.

[0152] The memory 603 is used to store a computer program;

[0153] When the processor 601 is configured to execute the program stored in the memory 603, the following steps are implemented:

[0154] Obtain the object information of the object to be served; wherein, the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served;

[0155] Perform mapping processing on the object information of the object to be served to obtain a feature vector of the object to be served as the first feature vector;

[0156] Input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model; wherein, the level prediction model is trained based on the sample object information and sample service levels of sample objects;

[0157] Determine the service method corresponding to the service level of the object to be served in the preset correspondence between service levels and service methods.

[0158] The communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity, only a thick line is used in the figure to represent it, but it does not mean that there is only one bus or one type of bus.

[0159] The communication interface is used for communication between the above electronic device and other devices.

[0160] The memory may include a Random Access Memory (RAM), and may also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0161] The above-mentioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0162] Based on the electronic device provided by the embodiment of the present invention, the level prediction model is trained based on the object information of the sample object and the sample service level. The level prediction model can learn the mapping relationship between the object information and the service level. Correspondingly, the service level of the object to be served can be determined through the level prediction model and the object information of the object to be served, and then the service method corresponding to the service level of the object to be served can be determined. It is not necessary to evaluate the questionnaire survey results of the elderly manually to determine the service method of the elderly, which can exclude the influence of personal subjective factors, improve the accuracy of the determined service method, improve the service quality, and can reduce the time cost and labor cost, and improve the efficiency of determining the service method.

[0163] In another embodiment provided by the present invention, a computer-readable storage medium is also provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above service method determination methods are implemented.

[0164] In another embodiment provided by the present invention, a computer program product containing instructions is also provided. When it runs on a computer, it causes the computer to execute any of the service method determination methods in the above embodiments.

[0165] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0166] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements but also other elements that are not explicitly listed, or elements that are inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article, or device that includes the element.

[0167] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the device, electronic device, computer-readable storage medium, and computer program product, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0168] The above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are all included within the protection scope of the present invention.

Claims

1. A method for determining a service mode, characterized in that, the method includes: obtaining object information of an object to be served; wherein, the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served; performing a mapping process on the object information of the object to be served to obtain a feature vector of the object to be served as the first feature vector; inputting the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model; wherein, the level prediction model is trained based on sample object information and sample service levels of sample objects; determining, in a preset correspondence between service levels and service modes, the service mode corresponding to the service level of the object to be served; wherein, the service mode includes: the period of providing elderly care services to the object to be served; the level prediction model includes: a linear module, a non-linear module, a first multi-layer perceptron, a second multi-layer perceptron, and an output module; the step of inputting the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model includes: performing a linear transformation on the first feature vector through the linear module to obtain a second feature vector; performing a non-linear transformation on the first feature vector through the non-linear module to obtain a third feature vector; performing a linear transformation on the third feature vector through the first multi-layer perceptron to obtain a fourth feature vector; performing feature fusion on the second feature vector and the fourth feature vector to obtain a fifth feature vector; performing a linear transformation on the fifth feature vector through the second multi-layer perceptron to obtain a first value; performing a normalization process on the first value through the output module to obtain a second value for representing the service level of the object to be served.

2. The method according to claim 1, characterized in that, the personal basic information of the object to be served includes at least one of the following: the age, gender, marital status, and number of children of the object to be served; the physical health information of the object to be served includes at least one of the following: the height, weight, and current disease information of the object to be served; the living environment information of the object to be served includes at least one of the following: the number of dining places and the number of activity centers within a preset geographical range of the residence of the object to be served; the social activity information of the object to be served includes at least one of the following: the types and frequencies of community activities participated by the object to be served within a historical time period.

3. The method according to claim 1, characterized in that, performing a mapping process on the object information of the object to be served to obtain a feature vector of the object to be served as the first feature vector includes: determining the numerical information and non-numerical information in the object information of the object to be served; wherein, the numerical information is information represented by numerical values; the non-numerical information is other information except the numerical information; Encode the non-numerical information to obtain a sixth feature vector; Normalize the numerical information and perform a linear transformation on the normalized numerical information to obtain a seventh feature vector; Concatenate the sixth feature vector and the seventh feature vector to obtain a feature vector of the object to be served, which is used as the first feature vector.

4. The method according to claim 1, wherein, the training process of the level prediction model includes the following steps: Obtain the sample object information of the sample object and the sample value representing the sample service level of the sample object; Perform a mapping process on the sample object information to obtain a feature vector of the sample object, which is used as a sample feature vector; Use the sample feature vector as the input data of the level prediction model with an initial structure, and use the sample value representing the sample service level of the sample object as the output data of the level prediction model with the initial structure, and adjust the model parameters of the level prediction model with the initial structure until the level prediction model with the initial structure reaches a preset convergence condition to obtain a trained level prediction model.

5. A service mode determination device, wherein, the device includes: An acquisition module, configured to acquire the object information of the object to be served; wherein, the object information includes at least one of the personal basic information, physical health information, living environment information, and social activity information of the object to be served; A mapping module, configured to perform a mapping process on the object information of the object to be served to obtain a feature vector of the object to be served, which is used as the first feature vector; A prediction module, configured to input the first feature vector into a pre-trained level prediction model to obtain the service level of the object to be served output by the level prediction model; wherein, the level prediction model is trained based on the sample object information and the sample service level of the sample object; A determination module, configured to determine the service mode corresponding to the service level of the object to be served in a preset correspondence between the service level and the service mode; wherein, the service mode includes: the period of providing elderly care services to the object to be served; The level prediction model includes: a linear module, a non-linear module, a first multi-layer perceptron, a second multi-layer perceptron, and an output module; The prediction module is specifically configured to perform a linear transformation on the first feature vector through the linear module to obtain a second feature vector; Perform a non-linear transformation on the first feature vector through the non-linear module to obtain a third feature vector; Perform a linear transformation on the third feature vector through the first multi-layer perceptron to obtain a fourth feature vector; Perform feature fusion on the second feature vector and the fourth feature vector to obtain a fifth feature vector; Perform a linear transformation on the fifth feature vector through the second multi-layer perceptron to obtain a first value; Perform a normalization process on the first value through the output module to obtain a second value representing the service level of the object to be served.

6. The device according to claim 5, wherein, The personal basic information of the object to be served includes at least one of the following: the age, gender, marital status, and number of children of the object to be served; The physical health information of the object to be served includes at least one of the following: the height, weight, and current disease information of the object to be served; The living environment information of the object to be served includes at least one of the following: the number of dining places and the number of activity centers within a preset geographical range of the residence of the object to be served; The social activity information of the object to be served includes at least one of the following: the types and frequencies of community activities participated by the object to be served within a historical time period.

7. The device according to claim 5, wherein, The mapping module is specifically configured to determine the numerical information and non-numerical information in the object information of the object to be served; wherein, the numerical information is information represented by a numerical value; the non-numerical information is other information except the numerical information; Perform encoding processing on the non-numerical information to obtain a sixth feature vector; Perform normalization processing on the numerical information, and perform linear transformation on the normalized numerical information to obtain a seventh feature vector; Concatenate the sixth feature vector and the seventh feature vector to obtain a feature vector of the object to be served as the first feature vector.

8. The device according to claim 5, wherein, The device further includes a training module, configured to obtain the sample object information of the sample object and the sample value representing the sample service level of the sample object; Perform mapping processing on the sample object information to obtain a feature vector of the sample object as the sample feature vector; Use the sample feature vector as the input data of the level prediction model with the initial structure, and use the sample value representing the sample service level of the sample object as the output data of the level prediction model with the initial structure, and adjust the model parameters of the level prediction model with the initial structure until the level prediction model with the initial structure reaches a preset convergence condition to obtain a trained level prediction model.

9. An electronic device, wherein, It includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus; The memory is used to store a computer program; The processor is configured to implement the method steps described in any one of claims 1-4 when executing the program stored on the memory.

10. A computer-readable storage medium, wherein, The computer-readable storage medium stores a computer program, and the computer program realizes the method steps described in any one of claims 1-4 when executed by a processor.

Images