Intelligent order distribution matching model training method and device

Abstract

The invention discloses a training method and device for an intelligent order distribution matching model. According to one specific embodiment, the method comprises the steps that a training sample set is acquired, and training samples in the training sample set comprise maintenance order feature data corresponding to a maintenance order, maintenance order portrait data and engineer portrait data corresponding to a maintenance engineer; and adopting a machine learning method, taking the maintenance order feature data as input, taking the maintenance order portrait data and the engineer portrait data corresponding to the input maintenance order feature data as expected output, and training to obtain an intelligent order distribution matching model. According to the method, the intelligent order distribution matching model capable of determining the engineer portrait data of the maintenance engineer corresponding to the maintenance order is obtained through training, the maintenance engineer corresponding to the maintenance order is determined according to the engineer portrait data, the dispatching operation is automatically executed, and the efficiency and accuracy of the maintenance order dispatching process are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of artificial intelligence, specifically to the field of deep learning, and in particular to a training method and device for an intelligent order splitting and matching model, an intelligent order splitting method and device, a computer-readable medium, and an electronic device. Background Art

[0002] After the items to be repaired enter the repair center, the dispatcher will pull the items to be repaired from the warehouse pallet to the dispatch area in batches at irregular intervals. Based on the size and brand of the items to be repaired, the items to be repaired will be distributed offline to the team leader of the maintenance engineer's group. The maintenance engineer team leader will distribute the items to be repaired within the group. Currently, the distribution process of items to be repaired is relatively cumbersome and heavily dependent on the dispatcher's experience. Summary of the invention

[0003] The embodiments of the present application propose a training method and device for an intelligent order splitting and matching model, an intelligent order splitting method and device, a computer-readable medium, and an electronic device.

[0004] In the first aspect, an embodiment of the present application provides a training method for an intelligent order matching model, including: obtaining a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data, and engineer portrait data corresponding to the maintenance engineer; using a machine learning method, with the maintenance order feature data as input, and the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected outputs, to train an intelligent order matching model.

[0005] In some examples, the above method also includes: obtaining maintenance order feature data of maintenance orders in the maintenance order set and engineer feature data of maintenance engineers in the engineer set; clustering the maintenance order feature data of maintenance orders in the maintenance order set to determine maintenance order portrait data corresponding to the maintenance orders; clustering the engineer feature data of maintenance engineers in the engineer set to determine engineer portrait data of maintenance engineers; generating an initial sample set based on the correspondence between maintenance orders in the maintenance order set and maintenance engineers in the engineer set, maintenance order portrait data and engineer portrait data, wherein each initial sample in the initial sample set includes maintenance order feature data corresponding to the maintenance order, maintenance order portrait data and engineer portrait data corresponding to the maintenance engineer; dividing the initial sample set according to a preset division ratio to obtain a training sample set, a verification sample set and a test sample set.

[0006] In some examples, the above method also includes: determining the verification result of the intelligent order matching model through a verification sample set; determining the test result of the intelligent order matching model through a test sample set; in response to determining that the verification result reaches a preset verification threshold and the test result reaches a preset test threshold, determining the intelligent order matching model as an intelligent order matching model that can be deployed for application.

[0007] In some examples, the above-mentioned clustering of the maintenance order feature data of the maintenance orders in the maintenance order set to determine the maintenance order portrait data corresponding to the maintenance orders includes: for the maintenance order feature data of the maintenance orders in the maintenance order set, clustering the maintenance orders from each feature dimension to obtain the maintenance order clustering data of the maintenance orders under each feature dimension; obtaining the maintenance order portrait data corresponding to the maintenance order based on the maintenance order clustering data of the maintenance orders under each feature dimension.

[0008] In some examples, the above-mentioned clustering of the engineer feature data of the maintenance engineers in the engineer set to determine the engineer portrait data of the maintenance engineers includes: for the engineer feature data of the maintenance engineers in the engineer set, clustering the maintenance engineers from each feature dimension to obtain the engineer clustering data of the maintenance engineers under each feature dimension; according to the engineer clustering data of the maintenance engineers under each feature dimension, obtaining the engineer portrait data corresponding to the maintenance engineers.

[0009] On the second aspect, an embodiment of the present application provides an intelligent order splitting method, including: obtaining the maintenance order feature data of the maintenance order to be assigned; inputting the maintenance order feature data into the trained intelligent order splitting matching model to determine the target engineer portrait data corresponding to the maintenance order; and dispatching the maintenance order to the maintenance engineer corresponding to the target engineer portrait data.

[0010] In some examples, the above-mentioned dispatching of the maintenance order to the maintenance engineer corresponding to the target engineer portrait data includes: determining whether the user to which the maintenance order belongs is a target type user; in response to determining that the user is a target type user, dispatching the maintenance order to the target maintenance engineer among multiple maintenance engineers corresponding to the target engineer portrait data.

[0011] On the third aspect, an embodiment of the present application provides a training device for an intelligent order matching model, comprising: a first acquisition unit, configured to acquire a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data, and engineer portrait data corresponding to the maintenance engineer; a training unit, configured to adopt a machine learning method, with the maintenance order feature data as input, and the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected outputs, to train an intelligent order matching model.

[0012] In some examples, the above-mentioned device also includes: a second acquisition unit, configured to acquire maintenance order feature data of maintenance orders in the maintenance order set and engineer feature data of maintenance engineers in the engineer set; a first clustering unit, configured to cluster the maintenance order feature data of maintenance orders in the maintenance order set to determine maintenance order portrait data corresponding to the maintenance orders; a second clustering unit, configured to cluster the engineer feature data of maintenance engineers in the engineer set to determine engineer portrait data of maintenance engineers; a generation unit, configured to generate an initial sample set based on the correspondence between maintenance orders in the maintenance order set and maintenance engineers in the engineer set, maintenance order portrait data and engineer portrait data, wherein each initial sample in the initial sample set includes maintenance order feature data corresponding to the maintenance order, maintenance order portrait data and engineer portrait data corresponding to the maintenance engineer; a division unit, configured to divide the initial sample set according to a preset division ratio to obtain a training sample set, a verification sample set and a test sample set.

[0013] In some examples, the above-mentioned device also includes: a verification unit, configured to determine the verification result of the intelligent order matching model through a verification sample set; a testing unit, configured to determine the test result of the intelligent order matching model through a test sample set; and a deployment and application unit, configured to determine the intelligent order matching model as an intelligent order matching model that can be deployed for application in response to determining that the verification result reaches a preset verification threshold and the test result reaches a preset test threshold.

[0014] In some examples, the first clustering unit is further configured to: cluster the maintenance order feature data of the maintenance order in the maintenance order set from each feature dimension to obtain the maintenance order clustering data of the maintenance order under each feature dimension; and obtain the maintenance order portrait data corresponding to the maintenance order based on the maintenance order clustering data of the maintenance order under each feature dimension.

[0015] In some examples, the second clustering unit is further configured to: for the engineer feature data of the maintenance engineers in the engineer set, cluster the maintenance engineers from each feature dimension to obtain the engineer clustering data of the maintenance engineers under each feature dimension; and obtain the engineer portrait data corresponding to the maintenance engineers based on the engineer clustering data of the maintenance engineers under each feature dimension.

[0016] In the fourth aspect, an embodiment of the present application provides an intelligent order splitting device, including: a third acquisition unit, configured to acquire maintenance order feature data of a maintenance order to be assigned; a determination unit, configured to input the maintenance order feature data into a trained intelligent order splitting matching model to determine the target engineer portrait data corresponding to the maintenance order; and a dispatching unit, configured to dispatch the maintenance order to the maintenance engineer corresponding to the target engineer portrait data.

[0017] In some examples, the dispatching unit is further configured to: determine whether the user to which the maintenance order belongs is a target type user; in response to determining that the user is a target type user, dispatch the maintenance order to a target maintenance engineer among multiple maintenance engineers corresponding to the target engineer portrait data.

[0018] In a fifth aspect, an embodiment of the present application provides a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation of the first aspect and the second aspect is implemented.

[0019] In a sixth aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation manner of the first aspect and the second aspect.

[0020] The training method and device of the intelligent order matching model provided in the embodiment of the present application obtain a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data and engineer portrait data corresponding to the maintenance engineer; adopt a machine learning method, take the maintenance order feature data as input, and take the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected output, to train an intelligent order matching model, so as to train an intelligent order matching model capable of determining the engineer portrait data of the maintenance engineer corresponding to the maintenance order, so as to determine the maintenance engineer corresponding to the maintenance order according to the engineer portrait data, automatically perform the dispatch operation, and improve the efficiency and accuracy of the maintenance order dispatch process. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0022] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present application may be applied;

[0023] Figure 2 It is a flowchart of an embodiment of a training method for an intelligent order matching model according to the present application;

[0024] Figure 3 is a schematic diagram of an application scenario of the training method of the intelligent order matching model according to this embodiment;

[0025] Figure 4 is a flowchart of another embodiment of the training method of the intelligent order matching model according to the present application;

[0026] Figure 5 is a flow chart of an embodiment of the intelligent order splitting method according to the present application;

[0027] Figure 6 It is a structural diagram of an embodiment of a training device for an intelligent order matching model according to the present application;

[0028] Figure 7 is a structural diagram of an embodiment of an intelligent order splitting device according to the present application;

[0029] Figure 8 It is a structural diagram of a computer system suitable for implementing the embodiments of the present application. DETAILED DESCRIPTION

[0030] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0031] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0032] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0033] Figure 1 An exemplary architecture 100 is shown to which the training method and device of the intelligent order splitting and matching model, and the intelligent order splitting method and device of the present application can be applied.

[0034] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The communication connection between the terminal devices 101, 102, 103 constitutes a topological network, and the network 104 is used to provide a medium for the communication link 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 optical fiber cables, etc.

[0035] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Terminal devices 101, 102, 103 can be hardware devices or software that support network connection for data interaction and data processing. When terminal devices 101, 102, 103 are hardware, they can be various electronic devices that support network connection, information acquisition, interaction, display, processing and other functions, including but not limited to smart phones, tablet computers, e-book readers, laptop portable computers and desktop computers, etc. When terminal devices 101, 102, 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules for providing distributed services, for example, or it can be implemented as a single software or software module. No specific limitation is made here.

[0036] Server 105 can be a server that provides various services, for example, a background processing server that receives training requests from terminal devices 101, 102, and 103, and uses machine learning methods to train an intelligent order matching model. For another example, a background processing server that receives repair order feature data from terminal devices 101, 102, and 103 and determines the maintenance engineer corresponding to the repair order through an intelligent order matching model. As an example, server 105 can be a cloud server.

[0037] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, software or software modules used to provide distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0038] It should also be noted that the training method of the intelligent order splitting matching model and the intelligent order splitting method provided in the embodiments of the present application can be executed by a server, or by a terminal device, or by the server and the terminal device in cooperation with each other. Accordingly, the various parts (such as various units) included in the training device of the intelligent order splitting matching model and the intelligent order splitting device can be all set in the server, or all set in the terminal device, or can be set in the server and the terminal device respectively.

[0039] It should be understood that Figure 1The number of terminal devices, networks and servers in the system is only illustrative. Depending on the implementation requirements, there may be any number of terminal devices, networks and servers. When the training method of the intelligent order matching model and the electronic device on which the intelligent order matching method runs do not need to transmit data with other electronic devices, the system architecture may only include the training method of the intelligent order matching model and the electronic device (such as a server or terminal device) on which the intelligent order matching method runs.

[0040] Continue to refer Figure 2 , a process 200 of an embodiment of a training method for an intelligent order matching model is shown, comprising the following steps:

[0041] Step 201: Obtain a training sample set.

[0042] In this embodiment, the execution subject of the training method of the intelligent order matching model (for example Figure 1 The terminal device or server in the training sample set obtains the training sample set remotely or locally through a wired network connection or a wireless network connection. The training samples in the training sample set include the maintenance order feature data corresponding to the maintenance order, the maintenance order portrait data, and the engineer portrait data corresponding to the maintenance engineer.

[0043] A repair order is a document that represents an item to be repaired. The document includes multi-dimensional repair order feature data such as the delivery address, brand, and category of the item to be repaired. The item to be repaired can be of various categories, such as home appliances, electronic products, etc.

[0044] For the characteristic data of each dimension in the maintenance order characteristic data, classification is performed based on a preset division method, and the sub-repair order portrait data corresponding to the characteristic data of this dimension can be determined; the maintenance order portrait data of the maintenance order can be determined by combining the sub-repair order portrait data corresponding to the characteristic data of each dimension.

[0045] As an example, the repair order feature data and the corresponding repair order portrait data of a repair order are shown in the following table:

[0046] Repair order number Delivery address brand Category Repair order image data 1 City A aBrand cell phone North China, brand a, mobile phone 2 City B b brand cell phone Northeast region, brand b, mobile phone 3 City C aBrand computer North China, brand a, computer

[0047] A maintenance engineer is a person who has the ability to repair the items to be repaired as represented by the maintenance order. The engineer portrait data of the maintenance engineer can be data reflecting the characteristics of the maintenance engineer from multiple dimensions such as the organization to which the maintenance engineer belongs, maintenance capability, processing time efficiency, and service quality.

[0048] As an example, the engineer characteristic data and corresponding engineer portrait data of a maintenance engineer are shown in the following table:

[0049]

[0050] In this embodiment, for each training sample in the training sample set, the maintenance order feature data and maintenance order portrait data in the training sample correspond to the same maintenance order, and the maintenance engineer corresponding to the engineer portrait data in the training sample is the maintenance engineer who has handled the maintenance order corresponding to the training sample.

[0051] In step 202, a machine learning method is used, with the maintenance order feature data as input, and the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected output, to train an intelligent order matching model.

[0052] In this embodiment, the above-mentioned execution entity can adopt a machine learning method, with the maintenance order feature data as input, and the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as the expected output, to train and obtain an intelligent order matching model.

[0053] As an example, the above-mentioned execution entity can iteratively perform the following training operations: first, select untrained training samples from the training sample set; then, input the maintenance order feature data in the selected training samples into the initial intelligent order matching model to obtain the output maintenance order portrait data and output engineer portrait data corresponding to the initial intelligent order matching model; then, determine the first loss between the output maintenance order portrait data and the maintenance order portrait data in the selected training sample, and the second loss between the output engineer portrait data and the engineer portrait data in the selected training sample; finally, determine the update gradient according to the first loss and the second loss, and use the stochastic gradient descent method to update the initial intelligent order matching model to obtain the initial intelligent order matching model corresponding to the next training operation.

[0054] In the iterative training process, in response to reaching a preset end condition, a trained intelligent order matching model is obtained, wherein the preset training condition may be, for example, that the number of training iterations exceeds a preset number threshold, that the training time operation exceeds a preset time threshold, or that the training loss tends to converge.

[0055] Among them, the initial intelligent order matching model can adopt various matching models. As an example, the initial intelligent order matching model can be a model that adopts a traditional matching method based on latent space, for example, a model that adopts a BM (Best Matching) algorithm, a model that adopts a PLS (Partial Least Squares) method, and a RMLS (Regularized Mapping to Latent Space) model; the initial intelligent order matching model can also be a deep matching model based on representation learning, for example, a DSSM (Deep Structured Semantic Models) model, a CNTN (Convolutional Neural Tensor Network) model, and a LSTM (Longshort-term memory)-RNN (Recurrent Neural Network) model.

[0056] Continue to see Figure 3 , Figure 3 FIG. 3 is a schematic diagram 300 of an application scenario of the training method of the intelligent order matching model according to this embodiment. Figure 3 In the application scenario, the server 301 first obtains the training sample set 303 from the database 302. The training samples in the training sample set include the maintenance order feature data corresponding to the maintenance order, the maintenance order portrait data and the engineer portrait data corresponding to the maintenance engineer. Then, the server 301 adopts a machine learning method, takes the maintenance order feature data as the input of the initial intelligent order matching model 304, obtains the output maintenance order portrait data and the output engineer portrait data, and then determines the first loss between the output maintenance order portrait data and the maintenance order portrait data corresponding to the input maintenance order feature data, and the second loss between the output engineer portrait data and the engineer portrait data corresponding to the input maintenance order feature data; then, the initial intelligent order matching model is updated according to the first loss and the second loss, and finally the intelligent order matching model 305 is trained.

[0057] The method provided by the above-mentioned embodiment of the present application obtains a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data and engineer portrait data corresponding to the maintenance engineer; adopts a machine learning method, takes the maintenance order feature data as input, and takes the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected output, and trains to obtain an intelligent order matching model, so as to train to obtain an intelligent order matching model capable of determining the engineer portrait data of the maintenance engineer corresponding to the maintenance order, so as to determine the maintenance engineer corresponding to the maintenance order according to the engineer portrait data, automatically perform the dispatch operation, and improve the efficiency and accuracy of the maintenance order dispatch process.

[0058] In some optional implementations of this embodiment, the above execution subject may also perform the following operations:

[0059] First, the maintenance order feature data of the maintenance orders in the maintenance order set and the engineer feature data of the maintenance engineers in the engineer set are obtained.

[0060] The maintenance orders in the maintenance order set are maintenance orders processed by the maintenance engineers in the engineer set, that is, the maintenance orders in the maintenance order set correspond to the maintenance engineers in the engineer set.

[0061] Second, cluster the maintenance order feature data of the maintenance orders in the maintenance order set to determine the maintenance order portrait data corresponding to the maintenance orders.

[0062] Third, the engineer characteristic data of the maintenance engineers in the engineer set are clustered to determine the engineer portrait data of the maintenance engineers.

[0063] As an example, whether it is the maintenance order feature data of the maintenance order in the maintenance order set, or the engineer feature data of the maintenance engineer in the engineer set, the above-mentioned execution entity can use the Kmeans clustering method to perform feature clustering to determine the corresponding maintenance order portrait data and engineer portrait data.

[0064] Fourth, an initial sample set is generated based on the correspondence between the maintenance orders in the maintenance order set and the maintenance engineers in the engineer set, the maintenance order portrait data, and the engineer portrait data.

[0065] Each initial sample in the initial sample set includes maintenance order feature data corresponding to the maintenance order, maintenance order portrait data, and engineer portrait data corresponding to the maintenance engineer. The data structure in the initial samples in the initial sample set is the same as the data structure in the training samples in the training sample set.

[0066] As an example, for each maintenance order in the maintenance order set, the maintenance order feature data, maintenance order portrait data, and engineer portrait data of the maintenance engineer who processed the maintenance order are combined to obtain the initial sample corresponding to the maintenance order; the initial samples corresponding to each maintenance order in the maintenance order set are combined to obtain the initial sample set.

[0067] Fifth, the initial sample set is divided according to a preset division ratio to obtain a training sample set, a verification sample set, and a test sample set.

[0068] The preset division ratio can be specifically set according to the actual situation. For example, if the amount of data in the initial sample set is small, the preset division ratio of the training sample set, the validation sample set and the test sample set is 6:2:2; if the amount of data in the initial sample set is large, the preset division ratio of the training sample set, the validation sample set and the test sample set is 98:1:1.

[0069] Among them, the training sample set is used to train the intelligent order matching model, the verification sample set is used to determine the network structure of the intelligent order matching model and adjust the model's hyperparameters; the test sample set is used to test the generalization ability of the intelligent order matching model.

[0070] In this implementation, a specific implementation method for generating a training sample set, a verification sample set, and a test sample set is provided, thereby improving the efficiency of generating the sample set.

[0071] In some optional implementations of the present embodiment, the above-mentioned execution entity may perform the above-mentioned second step in the following manner: first, for the maintenance order feature data of the maintenance orders in the maintenance order set, the maintenance orders are clustered from each feature dimension to obtain the maintenance order clustering data of the maintenance orders under each feature dimension; then, based on the maintenance order clustering data of the maintenance orders under each feature dimension, the maintenance order portrait data corresponding to the maintenance orders is obtained.

[0072] As an example, for each maintenance order in the maintenance order set, the three feature dimensions of delivery address, brand, and category in the maintenance order feature data of the maintenance order are clustered using the Kmeans clustering method to obtain the maintenance order clustering data under each feature dimension; then, the maintenance order clustering data under each feature dimension is combined to obtain the maintenance order portrait data corresponding to the maintenance order.

[0073] In this implementation, feature clustering is performed on the maintenance order from various feature dimensions, and finally maintenance order portrait data is generated, thereby improving the accuracy of the obtained maintenance order portrait data.

[0074] In some optional implementations of the present embodiment, the execution entity may perform the third step as follows: first, for the engineer feature data of the maintenance engineers in the engineer set, cluster the maintenance engineers from each feature dimension to obtain the engineer clustering data of the maintenance engineers under each feature dimension; then, based on the engineer clustering data of the maintenance engineers under each feature dimension, obtain the engineer portrait data corresponding to the maintenance engineers.

[0075] As an example, for each engineer in the engineer set, the Kmeans clustering method is used to perform feature clustering on the four feature dimensions of the maintenance engineer's engineer feature data, namely, the affiliated organization, maintenance capability, processing time, and service quality, to obtain the engineer clustering data under each feature dimension; then, the engineer clustering data of the maintenance engineer under each feature dimension is combined to obtain the engineer portrait data corresponding to the engineer.

[0076] In this implementation, feature clustering is performed on maintenance engineers from various feature dimensions, and finally engineer portrait data is generated, thereby improving the accuracy of the obtained engineer portrait data.

[0077] In some optional implementations of this embodiment, the above-mentioned execution entity may also perform the following operations: first, determine the verification result of the intelligent order matching model by verifying the sample set; then, determine the test result of the intelligent order matching model by testing the sample set; finally, in response to determining that the verification result reaches a preset verification threshold and the test result reaches a preset test threshold, determine the intelligent order matching model as an intelligent order matching model that can be deployed for application.

[0078] In this implementation, the preset verification threshold and the preset test threshold can be specifically set according to the actual situation and are not limited here. After obtaining the deployable intelligent order matching model, it can be deployed to be applied to the intelligent dispatching process for the maintenance order.

[0079] In this implementation, the intelligent order splitting and matching model is verified by verifying the sample set and tested by testing the sample set, thereby further improving the accuracy of the intelligent order splitting and matching model.

[0080] Continue to refer Figure 4 , shows a schematic process 400 of an embodiment of a training method for an intelligent order matching model according to the present application, including the following steps:

[0081] Step 401: Acquire maintenance order feature data of maintenance orders in a maintenance order set and engineer feature data of maintenance engineers in an engineer set.

[0082] Step 402: cluster the maintenance orders from each characteristic dimension for the maintenance order characteristic data of the maintenance orders in the maintenance order set, and obtain maintenance order clustering data under each characteristic dimension.

[0083] Step 403, obtaining maintenance order portrait data corresponding to the maintenance order based on the maintenance order clustering data under each feature dimension.

[0084] Step 404 , for the engineer characteristic data of the maintenance engineers in the engineer set, cluster the maintenance engineers from each characteristic dimension to obtain the engineer clustering data of the maintenance engineers under each characteristic dimension.

[0085] Step 405 , obtaining engineer portrait data corresponding to the maintenance engineer based on the engineer clustering data of the maintenance engineer in each feature dimension.

[0086] Step 406, generating an initial sample set according to the correspondence between the maintenance orders in the maintenance order set and the maintenance engineers in the engineer set, the maintenance order portrait data and the engineer portrait data.

[0087] Among them, each initial sample in the initial sample set includes maintenance order feature data corresponding to the maintenance order, maintenance order portrait data, and engineer portrait data corresponding to the maintenance engineer.

[0088] Step 407, dividing the initial sample set according to a preset division ratio to obtain a training sample set, a verification sample set, and a test sample set.

[0089] Step 408, adopting a machine learning method, taking the maintenance order feature data in the training samples in the training sample set as input, and taking the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected output, to train and obtain an intelligent order matching model.

[0090] Step 409, determining the verification result of the intelligent order matching model by verifying the sample set.

[0091] Step 410, determining the test result of the intelligent order matching model through the test sample set.

[0092] Step 411, in response to determining that the verification result reaches a preset verification threshold and the test result reaches a preset test threshold, the intelligent order matching model is determined as an intelligent order matching model that can be deployed for application.

[0093] It can be seen from this embodiment that Figure 2Compared with the corresponding embodiments, process 400 of the training method of the intelligent order matching model in this embodiment specifically illustrates the generation process of the maintenance order portrait data and the engineer portrait data, the generation process of the sample set, the verification process and the testing process of the intelligent order matching model, further improving the accuracy of the determined intelligent order matching model.

[0094] Continue to refer Figure 5 , shows a process 500 of an embodiment of an intelligent order splitting method, including the following steps:

[0095] Step 501, obtaining maintenance order characteristic data of the maintenance order to be dispatched.

[0096] In this embodiment, the execution subject of the training method of the intelligent order matching model (for example Figure 1 The terminal device or server in the system) obtains the maintenance order characteristic data of the maintenance order to be assigned remotely or locally through a wired network connection or a wireless network connection.

[0097] A repair order is a document that represents the item to be repaired. The repair order feature data includes multi-dimensional feature data such as the delivery address, brand, category, etc. of the item to be repaired.

[0098] Step 502: Input the maintenance order feature data into the trained intelligent order matching model to determine the target engineer portrait data corresponding to the maintenance order.

[0099] In this embodiment, the above-mentioned execution entity can input the maintenance order feature data into the trained intelligent order matching model to determine the target engineer portrait data corresponding to the maintenance order.

[0100] Among them, the intelligent order matching model is used to characterize the correspondence between the maintenance order feature data and the target engineer portrait data, and is obtained through training in the above-mentioned embodiments 200 and 400.

[0101] Step 503: dispatch the maintenance order to the maintenance engineer corresponding to the target engineer portrait data.

[0102] In this embodiment, the above-mentioned execution entity can assign the maintenance order to the maintenance engineer corresponding to the target engineer portrait data.

[0103] As an example, based on the target engineer portrait data, at least one maintenance engineer matching the target engineer portrait data can be determined; when multiple maintenance engineers are determined, the maintenance order to be assigned can be randomly assigned to one of the multiple maintenance engineers.

[0104] As another example, the execution entity may, based on the determined business volumes of the multiple maintenance engineers and in accordance with the principle of business volume balance, assign the maintenance order to be assigned to a maintenance engineer with a smaller business volume among the multiple maintenance engineers.

[0105] In some optional implementations of this embodiment, the execution subject may perform step 503 in the following manner:

[0106] First, determine whether the user to whom the maintenance order belongs is a target type user.

[0107] Among them, the target type user represents the user that the above-mentioned execution subject needs to pay special attention to. As an example, the target type user is a member user, a risky customer, or a user with high purchasing power. Among them, the risky user represents a user who is more sensitive to the maintenance process and has a higher frequency of complaints and negative reviews.

[0108] Second, in response to determining that the user is a target type user, the maintenance order is assigned to a target maintenance engineer among a plurality of maintenance engineers corresponding to the target engineer portrait data.

[0109] As an example, the above-mentioned execution entity may determine, among multiple maintenance engineers corresponding to the target engineer portrait data, a maintenance engineer with stronger maintenance capability, shorter processing time, and better service quality as the target maintenance engineer.

[0110] In this implementation, the types of users are distinguished, and targeted processing is performed on the maintenance orders of target type users, which further improves the accuracy and rationality of the maintenance order dispatching process.

[0111] Continue to refer Figure 6 As an implementation of the methods shown in the above figures, the present application provides an embodiment of a training device for an intelligent order matching model. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0112] like Figure 6 As shown, the training device of the intelligent order matching model includes: a first acquisition unit 601, configured to acquire a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data and engineer portrait data corresponding to the maintenance engineer; a training unit 602, configured to adopt a machine learning method, with the maintenance order feature data as input, and the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected outputs, to train and obtain the intelligent order matching model.

[0113] In some optional implementations of the present embodiment, the above-mentioned device also includes: a second acquisition unit (not shown in the figure), configured to acquire maintenance order feature data of maintenance orders in the maintenance order set and engineer feature data of maintenance engineers in the engineer set; a first clustering unit (not shown in the figure), configured to cluster the maintenance order feature data of maintenance orders in the maintenance order set to determine maintenance order portrait data corresponding to the maintenance orders; a second clustering unit (not shown in the figure), configured to cluster the engineer feature data of maintenance engineers in the engineer set to determine engineer portrait data of maintenance engineers; a generation unit (not shown in the figure), configured to generate an initial sample set according to the correspondence between maintenance orders in the maintenance order set and maintenance engineers in the engineer set, maintenance order portrait data and engineer portrait data, wherein each initial sample in the initial sample set includes maintenance order feature data corresponding to the maintenance order, maintenance order portrait data and engineer portrait data corresponding to the maintenance engineer; a division unit (not shown in the figure), configured to divide the initial sample set according to a preset division ratio to obtain a training sample set, a verification sample set and a test sample set.

[0114] In some optional implementations of the present embodiment, the above-mentioned device also includes: a verification unit (not shown in the figure), configured to determine the verification result of the intelligent order matching model through a verification sample set; a testing unit (not shown in the figure), configured to determine the test result of the intelligent order matching model through a test sample set; and a deployment and application unit (not shown in the figure), configured to determine the intelligent order matching model as an intelligent order matching model that can be deployed for application in response to determining that the verification result reaches a preset verification threshold and the test result reaches a preset test threshold.

[0115] In some optional implementations of the present embodiment, the above-mentioned first clustering unit is further configured to: for the maintenance order feature data of the maintenance order in the maintenance order set, cluster the maintenance order from each feature dimension to obtain the maintenance order clustering data of the maintenance order under each feature dimension; based on the maintenance order clustering data of the maintenance order under each feature dimension, obtain the maintenance order portrait data corresponding to the maintenance order.

[0116] In some optional implementations of the present embodiment, the above-mentioned second clustering unit is further configured to: for the engineer feature data of the maintenance engineers in the engineer set, cluster the maintenance engineers from each feature dimension to obtain the engineer clustering data of the maintenance engineers under each feature dimension; based on the engineer clustering data of the maintenance engineers under each feature dimension, obtain the engineer portrait data corresponding to the maintenance engineers.

[0117] In this embodiment, the first acquisition unit in the training device of the intelligent order matching model acquires a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data, and engineer portrait data corresponding to the maintenance engineer; the training unit adopts a machine learning method, with the maintenance order feature data as input, and the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected output, to train an intelligent order matching model, thereby training an intelligent order matching model that can determine the engineer portrait data of the maintenance engineer corresponding to the maintenance order, so as to determine the maintenance engineer corresponding to the maintenance order according to the engineer portrait data, automatically perform the dispatch operation, and improve the efficiency and accuracy of the maintenance order dispatch process.

[0118] Continue to refer Figure 7 As an implementation of the methods shown in the above figures, the present application provides an embodiment of an intelligent order splitting device, and the device embodiment is similar to Figure 5 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0119] like Figure 7 As shown, the intelligent order distribution device includes: a third acquisition unit 701, configured to obtain the maintenance order feature data of the maintenance order to be distributed; a determination unit 702, configured to input the maintenance order feature data into the trained intelligent order distribution matching model to determine the target engineer portrait data corresponding to the maintenance order; and a dispatching unit 703, configured to dispatch the maintenance order to the maintenance engineer corresponding to the target engineer portrait data.

[0120] In some optional implementations of this embodiment, the above-mentioned dispatching unit 703 is further configured to: determine whether the user to whom the maintenance order belongs is a target type user; in response to determining that the user is a target type user, dispatch the maintenance order to a target maintenance engineer among multiple maintenance engineers corresponding to the target engineer portrait data.

[0121] In this embodiment, the third acquisition unit in the intelligent order distribution device acquires the maintenance order characteristic data of the maintenance order to be distributed; the determination unit inputs the maintenance order characteristic data into the trained intelligent order distribution matching model to determine the target engineer portrait data corresponding to the maintenance order; the dispatching unit dispatches the maintenance order to the maintenance engineer corresponding to the target engineer portrait data, thereby improving the efficiency and accuracy of the maintenance order distribution process through the intelligent order distribution matching model.

[0122] Reference below Figure 8 , which shows a device suitable for implementing the embodiments of the present application (eg Figure 1 Schematic diagram of the structure of a computer system 800 of the devices 101, 102, 103, 105 shown. Figure 8The device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0123] like Figure 8 As shown, the computer system 800 includes a processor (e.g., CPU, central processing unit) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage part 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the system 800 are also stored. The processor 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0124] The following components are connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed, so that a computer program read therefrom is installed into the storage section 808 as needed.

[0125] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part 809, and / or installed from a removable medium 811. When the computer program is executed by the processor 801, the above-mentioned functions defined in the method of the present application are executed.

[0126] It should be noted that the computer-readable medium of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0127] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the client computer, partially on the client computer, as a separate software package, partially on the client computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the client computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0128] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the device, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0129] The units involved in the embodiments described in the present application may be implemented by software or by hardware. The described units may also be provided in a processor. For example, they may be described as: a processor including a first acquisition unit and a training unit; for another example, they may be described as: a processor including a third acquisition unit, a determination unit, and a dispatching unit. Among them, the names of these units do not constitute limitations on the units themselves under certain circumstances. For example, the training unit may also be described as "a unit that uses a machine learning method, takes maintenance order feature data as input, and takes maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected outputs, to train an intelligent order matching model".

[0130] As another aspect, the present application also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently and not be assembled into the device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by the device, the computer device: obtains a training sample set, wherein the training samples in the training sample set include maintenance order feature data, maintenance order portrait data and engineer portrait data corresponding to the maintenance engineer corresponding to the maintenance order; adopts a machine learning method, takes the maintenance order feature data as input, and takes the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as the expected output, and trains to obtain an intelligent order matching model. The computer device can also: obtain the maintenance order feature data of the maintenance order to be assigned; input the maintenance order feature data into the trained intelligent order matching model to determine the target engineer portrait data corresponding to the maintenance order; and assign the maintenance order to the maintenance engineer corresponding to the target engineer portrait data.

[0131] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A training method for an intelligent order matching model, comprising: Acquire a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data, and engineer portrait data corresponding to the maintenance engineer; A machine learning method is adopted, with the repair order feature data as input and the repair order portrait data and engineer portrait data corresponding to the input repair order feature data as the expected output, to train an intelligent order matching model.

2. The method according to claim 1, wherein: Also includes: Obtain the maintenance order feature data of the maintenance order in the maintenance order set and the engineer feature data of the maintenance engineer in the engineer set; Clustering the maintenance order feature data of the maintenance orders in the maintenance order set to determine the maintenance order portrait data corresponding to the maintenance orders; Clustering the engineer feature data of the maintenance engineers in the engineer set to determine the engineer portrait data of the maintenance engineers; Generate an initial sample set according to the correspondence between the maintenance orders in the maintenance order set and the maintenance engineers in the engineer set, the maintenance order portrait data and the engineer portrait data, wherein each initial sample in the initial sample set includes the maintenance order feature data corresponding to the maintenance order, the maintenance order portrait data and the engineer portrait data corresponding to the maintenance engineer; The initial sample set is divided according to a preset division ratio to obtain the training sample set, the verification sample set and the test sample set.

3. The method according to claim 2, wherein: Also includes: Determine the verification result of the intelligent order matching model through the verification sample set; Determine the test result of the intelligent order matching model through the test sample set; In response to determining that the verification result reaches a preset verification threshold and the test result reaches a preset test threshold, the intelligent order matching model is determined as an intelligent order matching model that can be deployed for application.

4. The method according to claim 2, wherein: The clustering of the maintenance order feature data of the maintenance orders in the maintenance order set to determine the maintenance order portrait data corresponding to the maintenance orders includes: For the maintenance order feature data of the maintenance orders in the maintenance order set, clustering the maintenance orders from each feature dimension to obtain maintenance order clustering data of the maintenance orders under each feature dimension; According to the maintenance order clustering data of the maintenance order under each characteristic dimension, the maintenance order portrait data corresponding to the maintenance order is obtained.

5. The method according to claim 2, wherein: The clustering of the engineer characteristic data of the maintenance engineers in the engineer set to determine the engineer portrait data of the maintenance engineers includes: For the engineer characteristic data of the maintenance engineers in the engineer set, clustering the maintenance engineers from each characteristic dimension to obtain the engineer clustering data of the maintenance engineers under each characteristic dimension; According to the engineer clustering data of the maintenance engineer under each characteristic dimension, the engineer portrait data corresponding to the maintenance engineer is obtained.

6. An intelligent order splitting method, comprising: Obtain maintenance order characteristic data of the maintenance order to be dispatched; Input the maintenance order feature data into the trained intelligent order matching model to determine the target engineer portrait data corresponding to the maintenance order; The maintenance order is assigned to the maintenance engineer corresponding to the target engineer portrait data.

7. The method according to claim 6, wherein: The step of dispatching the maintenance order to the maintenance engineer corresponding to the target engineer portrait data includes: Determine whether the user to which the maintenance order belongs is a target type user; In response to determining that the user is a target type user, the maintenance order is assigned to a target maintenance engineer among a plurality of maintenance engineers corresponding to the target engineer portrait data.

8. A training device for an intelligent order matching model, comprising: A first acquisition unit is configured to acquire a training sample set, wherein the training samples in the training sample set include maintenance order feature data corresponding to the maintenance order, maintenance order portrait data, and engineer portrait data corresponding to the maintenance engineer; The training unit is configured to adopt a machine learning method, with the maintenance order feature data as input, and the maintenance order portrait data and engineer portrait data corresponding to the input maintenance order feature data as expected output, to train an intelligent order matching model.

9. An intelligent order splitting device, comprising: a third acquisition unit, configured to acquire maintenance order characteristic data of the maintenance order to be dispatched; A determination unit is configured to input the maintenance order feature data into the trained intelligent order matching model to determine the target engineer portrait data corresponding to the maintenance order; The dispatching unit is configured to dispatch the maintenance order to the maintenance engineer corresponding to the target engineer portrait data.

10. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

11. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

Images