Entity object sorting method, device and equipment based on logistics fulfillment quality

Abstract

The present application provides a method and device for sorting entity objects based on logistics fulfillment quality, an electronic device and a storage medium, and relates to the field of data processing technology. The method can determine the N first parameters of the client's user corresponding to each entity object to be recommended based on the client's user characteristic data, logistics data and characteristic data of each entity object to be recommended; re-sort the list of recommended objects in combination with each first parameter to obtain a list of recommended entity objects; and provide the re-sorted list of recommended entity objects to the client for display. The sorting result of the embodiment of the present application comprehensively considers the user characteristic data, logistics data, and characteristic data of each entity object to be recommended, so as to guide the user to place an order, guide the user to orders with higher value and better logistics experience, and improve the quality and efficiency of the recommendation; while improving the user's logistics experience, it can have little or no impact on the overall order volume.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for sorting entity objects based on logistics fulfillment quality, an electronic device, and a storage medium. Background Art

[0002] With the continuous development of Internet technology and the improvement of people's living standards, the food delivery industry is becoming more and more developed, and more and more people choose to order food through food delivery. Food delivery can be delivered by logistics and delivery resources such as unmanned vehicles and delivery personnel carrying mobile smart devices. During peak meal times, the volume of food delivery orders is very large, and there may be situations where food delivery cannot be delivered on time. In the food delivery scenario, the quality and ability of logistics fulfillment have an impact on the overall order volume, user and entity merchant satisfaction, retention rate, value, etc. However, if we simply increase logistics and delivery resources, such as adding delivery personnel, etc., after the peak meal time, it will lead to a waste of logistics and delivery resources. Therefore, it is urgent to solve this technical problem. Summary of the invention

[0003] In view of the above problems, this application is proposed to provide a method and device for sorting physical objects based on logistics fulfillment quality, an electronic device and a storage medium that overcome the above problems or at least partially solve the above problems. The technical solution is as follows:

[0004] In a first aspect, a method for sorting entity objects based on logistics fulfillment quality is provided, comprising:

[0005] In response to a request from a client to obtain a list of entity objects, obtaining a list of proposed recommended objects for a user of the client, including N entity objects to be recommended, where N is greater than 1;

[0006] Based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, respectively determine N first parameters of each entity object to be recommended corresponding to the user of the client;

[0007] Reorder the list of recommended objects in combination with the first parameters to obtain a list of recommended entity objects;

[0008] The reordered list of recommended entity objects is provided to the client for display.

[0009] In a possible implementation, based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, respectively determining N first parameters of each entity object to be recommended corresponding to the user of the client, including:

[0010] Inputting the user characteristic data, logistics data, and characteristic data of each entity object to be recommended of the client into a multi-task logistics fulfillment quality prediction model;

[0011] The multi-task logistics fulfillment quality prediction model is used to predict the logistics fulfillment quality of each of the entity objects to be recommended corresponding to the user of the client, and N logistics fulfillment quality prediction values ​​corresponding to each of the entity objects to be recommended are obtained as N first parameters corresponding to each of the entity objects to be recommended corresponding to the user of the client;

[0012] Among them, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input.

[0013] In a possible implementation, the multi-task logistics fulfillment quality prediction model includes multiple task networks, each task network is used to predict the probability of occurrence of a logistics quality situation;

[0014] The multi-task logistics fulfillment quality prediction model is used to predict the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client, and N logistics fulfillment quality prediction values ​​corresponding to each entity object to be recommended corresponding to the user of the client are obtained, including:

[0015] Utilizing the multiple task networks of the multi-task logistics fulfillment quality prediction model, predicting the occurrence probabilities of various logistics quality situations of the client user corresponding to each of the entity objects to be recommended, and obtaining the predicted values ​​of the occurrence probabilities of various logistics quality situations of the client user corresponding to each of the entity objects to be recommended;

[0016] According to the predicted values ​​of the occurrence probabilities of various logistics quality situations of the user of the client corresponding to each of the entity objects to be recommended, N predicted values ​​of logistics fulfillment quality corresponding to each of the entity objects to be recommended by the user of the client are determined.

[0017] In a possible implementation, according to the predicted values ​​of the occurrence probabilities of multiple logistics quality situations of the user of the client corresponding to each of the entity objects to be recommended, determining N logistics fulfillment quality prediction values ​​corresponding to each of the entity objects to be recommended by the user of the client includes:

[0018] Obtaining respective influence weights of various logistics quality conditions of the client user corresponding to each of the entity objects to be recommended;

[0019] According to the influence weights of the various logistics quality conditions of each entity object to be recommended corresponding to the user of the client and the predicted probability values ​​of the occurrence of the various logistics quality conditions of each entity object to be recommended corresponding to the user of the client, the predicted value of the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client is determined.

[0020] In a possible implementation, the list of recommended objects is reordered in combination with the first parameters to obtain a list of recommended entity objects, including:

[0021] Obtaining N second parameters of the environment state of each of the entity objects to be recommended;

[0022] The list of recommended objects is reordered in combination with the first parameters and the second parameters to obtain a list of recommended entity objects.

[0023] In a possible implementation, the list of recommended objects is reordered in combination with the first parameters and the second parameters to obtain a list of recommended entity objects, including:

[0024] Pre-configuring a control reordering strategy, wherein factors of the control reordering strategy include the first parameters and the second parameters;

[0025] The proposed recommended object list is reordered using the regulation reordering strategy to obtain a recommended entity object list.

[0026] In a possible implementation, the list of objects to be recommended is reordered by using the regulation reordering strategy to obtain a list of recommended entity objects, including:

[0027] Determine the reordering value of each of the entity objects to be recommended by using the regulation reordering strategy;

[0028] The proposed recommended object list is reordered according to the reordering value of each of the to-be-recommended entity objects to obtain a recommended entity object list.

[0029] In a second aspect, a method for sorting entity objects based on logistics fulfillment quality is provided, including:

[0030] Sending a request to the server to obtain a list of entity objects, wherein the request carries user-related information of the client;

[0031] receiving a reordered list of recommended entity objects provided by the server, wherein the server determines N first parameters of each entity object to be recommended corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended; and reorders the list of recommended objects in combination with the first parameters to obtain a list of recommended entity objects;

[0032] Display the reordered list of recommended entity objects.

[0033] In a possible implementation, a request for obtaining a list of entity objects is sent to a server, including:

[0034] It is detected that the preset control of the display interface is triggered, and a request for obtaining a list of entity objects is sent to the server.

[0035] In a third aspect, a device for sorting entity objects based on logistics fulfillment quality is provided, comprising:

[0036] An acquisition module, configured to, in response to a request from a client to acquire a list of entity objects, acquire a list of proposed recommended objects for a user of the client, comprising N entity objects to be recommended, wherein N is greater than 1;

[0037] A determination module, configured to determine N first parameters of each of the entity objects to be recommended corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended;

[0038] A sorting module, used to re-sort the list of recommended objects in combination with the first parameters to obtain a list of recommended entity objects;

[0039] A module is provided, for providing the reordered list of recommended entity objects to the client for display.

[0040] In a possible implementation manner, the determining module is further configured to:

[0041] Inputting the user characteristic data, logistics data, and characteristic data of each entity object to be recommended of the client into a multi-task logistics fulfillment quality prediction model;

[0042] The multi-task logistics fulfillment quality prediction model is used to predict the logistics fulfillment quality of each of the entity objects to be recommended corresponding to the user of the client, and N logistics fulfillment quality prediction values ​​corresponding to each of the entity objects to be recommended are obtained as N first parameters corresponding to each of the entity objects to be recommended corresponding to the user of the client;

[0043] Among them, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input.

[0044] In a possible implementation, the multi-task logistics fulfillment quality prediction model includes multiple task networks, each task network is used to predict the probability of occurrence of a logistics quality situation;

[0045] The determining module is also used for:

[0046] Utilizing the multiple task networks of the multi-task logistics fulfillment quality prediction model, predicting the occurrence probabilities of various logistics quality situations of the client user corresponding to each of the entity objects to be recommended, and obtaining the predicted values ​​of the occurrence probabilities of various logistics quality situations of the client user corresponding to each of the entity objects to be recommended;

[0047] According to the predicted values ​​of the occurrence probabilities of various logistics quality situations of the user of the client corresponding to each of the entity objects to be recommended, N predicted values ​​of logistics fulfillment quality corresponding to each of the entity objects to be recommended by the user of the client are determined.

[0048] In a possible implementation manner, the determining module is further configured to:

[0049] Obtaining respective influence weights of various logistics quality conditions of the client user corresponding to each of the entity objects to be recommended;

[0050] According to the influence weights of the various logistics quality conditions of each entity object to be recommended corresponding to the user of the client and the predicted probability values ​​of the occurrence of the various logistics quality conditions of each entity object to be recommended corresponding to the user of the client, the predicted value of the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client is determined.

[0051] In a possible implementation, the sorting module is further used to:

[0052] Obtaining N second parameters of the environment state of each of the entity objects to be recommended;

[0053] The list of recommended objects is reordered in combination with the first parameters and the second parameters to obtain a list of recommended entity objects.

[0054] In a possible implementation, the sorting module is further used to:

[0055] Pre-configuring a control reordering strategy, wherein factors of the control reordering strategy include the first parameters and the second parameters;

[0056] The proposed recommended object list is reordered using the regulation reordering strategy to obtain a recommended entity object list.

[0057] In a possible implementation, the sorting module is further used to:

[0058] Determine the reordering value of each of the entity objects to be recommended by using the regulation reordering strategy;

[0059] The proposed recommended object list is reordered according to the reordering value of each of the to-be-recommended entity objects to obtain a recommended entity object list.

[0060] In a fourth aspect, a device for sorting entity objects based on logistics fulfillment quality is provided, comprising:

[0061] A sending module, used to send a request to the server to obtain a list of entity objects, wherein the request carries user-related information of the client;

[0062] A receiving module, configured to receive a reordered list of recommended entity objects provided by the server, wherein the server determines N first parameters of each entity object to be recommended corresponding to the user of the client based on the user characteristic data, logistics data, and characteristic data of each entity object to be recommended of the client; and reorders the list of recommended objects in combination with the first parameters to obtain a list of recommended entity objects;

[0063] The display module is used to display the re-ordered list of recommended entity objects.

[0064] In a possible implementation, the sending module is further configured to:

[0065] It is detected that the preset control of the display interface is triggered, and a request for obtaining a list of entity objects is sent to the server.

[0066] In a fifth aspect, an electronic device is provided, comprising a processor and a memory, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any of the above-mentioned entity object sorting methods based on logistics fulfillment quality applied to the server or the entity object sorting method based on logistics fulfillment quality applied to the client.

[0067] In a sixth aspect, a storage medium is provided, which stores a computer program, wherein the computer program is configured to execute any of the above-mentioned entity object sorting methods based on logistics fulfillment quality applied to the server or the entity object sorting method based on logistics fulfillment quality applied to the client at runtime.

[0068] By means of the above technical scheme, the entity object sorting method and device based on logistics fulfillment quality, electronic device and storage medium provided in the embodiment of the present application can obtain a list of proposed recommended objects for the user of the client, including N entity objects to be recommended, in response to a request from the client to obtain a list of entity objects; based on the user characteristic data, logistics data and characteristic data of each entity object to be recommended of the client, respectively determine the N first parameters of each entity object to be recommended corresponding to the user of the client; re-sort the list of proposed recommended objects in combination with each first parameter to obtain a list of recommended entity objects; and provide the re-sorted list of recommended entity objects to the client for display. It can be seen that the embodiment of the present application can comprehensively consider the user characteristic data, logistics data and characteristic data of each entity object to be recommended of the client under the condition of certain logistics distribution capacity, and apply them to the sorting of the merchant list of entity objects; through refined regulation, the sorting result takes into account the fulfillment experience and logistics factors, thereby guiding the user to place an order, and guiding the user to an order with higher value and better logistics experience; while improving the user's logistics experience, it can have little or no impact on the overall order volume. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0070] Figure 1a A flow chart of a method for sorting entity objects based on logistics fulfillment quality applied to a service end provided in an embodiment of the present application is shown;

[0071] Figure 1b A schematic diagram of a control reordering strategy provided by an embodiment of the present application is shown;

[0072] Figure 2 A flow chart of a method for sorting entity objects based on logistics fulfillment quality applied to a client provided in an embodiment of the present application is shown;

[0073] Figure 3 A flowchart of a method for sorting entity objects based on logistics fulfillment quality in combination with a client and a server provided in an embodiment of the present application is shown;

[0074] Figure 4 A structural diagram of a physical object sorting device based on logistics fulfillment quality applied to a service end provided in an embodiment of the present application is shown;

[0075] Figure 5 A structural diagram of a physical object sorting device based on logistics fulfillment quality and applied to a client provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0076] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0077] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such use is interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the term "including" and its variants are to be interpreted as open-ended terms meaning "including but not limited to".

[0078] As mentioned above, in the takeaway delivery scenario, the quality and ability of logistics fulfillment have an impact on the overall order volume, user and entity merchant satisfaction, retention rate, value, etc. If you simply increase logistics and distribution resources, such as adding delivery personnel, etc., it will lead to a waste of logistics and distribution resources after the peak dining period. In addition, the existing solutions control the flow of traffic by considering the logistics pressure of the region, but since the quality of fulfillment is a comprehensive result, it not only involves logistics pressure, but also includes factors such as the physical merchant's meal delivery and the difficulty of delivery to the user's residence; in addition, regional granularity flow control will have a relatively obvious loss on the order volume, and the control is not precise.

[0079] To solve the above technical problems, the present application embodiment provides a method for sorting entity objects based on logistics fulfillment quality applied to a server, such as Figure 1a As shown, the entity object sorting method based on logistics fulfillment quality applied to the service end may include the following steps S101 to S104:

[0080] Step S101, in response to a request from a client to obtain an entity object list, obtaining a list of proposed recommended objects for a user of the client, which includes N entity objects to be recommended, where N is greater than 1;

[0081] Step S102, based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, respectively determine N first parameters of each entity object to be recommended corresponding to the user of the client;

[0082] Step S103, re-sorting the list of recommended objects in combination with each first parameter to obtain a list of recommended entity objects;

[0083] Step S104: providing the reordered list of recommended entity objects to the client for display.

[0084] The embodiment of the present application can respond to the client's request to obtain the list of entity objects, obtain a list of proposed recommended objects for the client's user, which includes N entity objects to be recommended; based on the client's user feature data, logistics data, and feature data of each entity object to be recommended, respectively determine the N first parameters of each entity object to be recommended corresponding to the client's user; re-sort the list of proposed recommended objects in combination with each first parameter to obtain a list of recommended entity objects; and provide the re-sorted list of recommended entity objects to the client for display. It can be seen that the embodiment of the present application can, under the condition of a certain logistics distribution capacity, comprehensively consider the client's user feature data, logistics data, and feature data of each entity object to be recommended, and apply them to the sorting of the merchant list of entity objects; through refined regulation, the sorting result takes into account the fulfillment experience and logistics factors, thereby guiding the user to place orders and guiding the user to orders with higher value and better logistics experience; while improving the user's logistics experience, it can have little or no impact on the overall order volume.

[0085] A possible implementation method is provided in the embodiment of the present application. In the takeaway delivery scenario, the entity object mentioned in step S101 above can be a merchant. The merchant registers a store on the takeaway platform. The user can log in to the takeaway platform through the client to browse the merchant's store, and then place an order in the merchant's store. The merchant can request delivery resources for takeaway delivery. In addition, the user-related information of the client can be obtained, which can specifically be the user's geographical location, the traffic conditions at the user's location, the weather conditions at the user's location, the property management conditions at the user's location, etc., and the embodiment of the present application does not limit this. Furthermore, the user feature data of the client can be extracted based on the user-related information of the client. In addition, logistics data and feature data of N entity objects to be recommended can be extracted, which can specifically be the geographical location of the entity object to be recommended, the business district where it is located, the logistics capacity of the business district where it is located, the traffic conditions at the location, the weather conditions at the location, etc., and the embodiment of the present application does not limit this.

[0086] The present application provides a possible implementation method. In the above step S102, based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, N first parameters of each entity object to be recommended corresponding to the user of the client are determined respectively. Specifically, the following steps A1 and A2 may be included:

[0087] Step A1, inputting the user feature data, logistics data, and feature data of each entity object to be recommended of the client into the multi-task logistics fulfillment quality prediction model;

[0088] Step A2, using the multi-task logistics fulfillment quality prediction model to predict the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client, and obtaining N logistics fulfillment quality prediction values ​​of each entity object to be recommended corresponding to the user of the client, as N first parameters of each entity object to be recommended corresponding to the user of the client;

[0089] Among them, the multi-task logistics fulfillment quality prediction model takes sample user feature data, sample logistics data and feature data of multiple sample entity objects as input, and takes sample logistics fulfillment quality value as output, and is obtained by training the multi-task learning MMOE (Multi-gate MixtureOfExperts) model.

[0090] In the above embodiment, the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client can be predicted in real time, so that the N logistics fulfillment quality prediction values ​​of each entity object to be recommended corresponding to the user of the client can be applied to the sorting of the merchant list of the entity object. For example, according to the order of the N logistics fulfillment quality prediction values ​​of each entity object to be recommended from large to small, the list of recommended objects is re-sorted to obtain a list of recommended entity objects; through refined regulation, the sorting results seen by the user appropriately consider the fulfillment experience and logistics conditions, thereby guiding the user to place orders and guiding the user to orders with higher value and better logistics experience; while improving the user's logistics experience, the overall order volume can be less affected or not affected.

[0091] In addition, through the multi-task logistics fulfillment quality prediction model, based on the client's user feature data, logistics data, and feature data of each entity object to be recommended, the whole process comprehensively considers multiple real-time factors such as logistics, merchants and users of entity objects, and predicts the logistics fulfillment quality of each entity object to be recommended among the multiple entity objects to be recommended where the client's user is located in real time. The logistics fulfillment quality prediction value of each entity object to be recommended where the client's user is located is applied to the sorting of the merchant list of the entity object, and the prediction and sorting efficiency and accuracy are high.

[0092] In an embodiment of the present application, a possible implementation method is provided. The multi-task logistics fulfillment quality prediction model mentioned in the above step A1 may include multiple task networks, and each task network may be used to predict the probability of occurrence of a logistics quality situation. Then, in step A2, the multi-task logistics fulfillment quality prediction model is used to predict the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client, and N logistics fulfillment quality prediction values ​​of each entity object to be recommended corresponding to the user of the client are obtained. Specifically, it may include the following steps A21 and A22:

[0093] Step A21, using multiple task networks of the multi-task logistics fulfillment quality prediction model, predicting the occurrence probabilities of multiple logistics quality situations corresponding to each entity object to be recommended for the user of the client, and obtaining the predicted values ​​of the occurrence probabilities of multiple logistics quality situations corresponding to each entity object to be recommended for the user of the client;

[0094] Step A22, determining N logistics fulfillment quality prediction values ​​corresponding to each entity object to be recommended by the user of the client according to the occurrence probability prediction values ​​of various logistics quality situations corresponding to each entity object to be recommended by the user of the client.

[0095] For example, the multiple task networks may be task network 1, task network 2, task network 3, and task network 4. Task network 1 is used to predict the probability of a logistics quality condition being "timeout", task network 2 is used to predict the probability of a logistics quality condition being "timeout for 15 minutes", task network 3 is used to predict the probability of a logistics quality condition being "timeout for 50 minutes", and task network 4 is used to predict the probability of a logistics quality condition being "the merchant or logistics of the entity object cancels the delivery". Thus, by using the multiple task networks of the multi-task logistics fulfillment quality prediction model, the probability of occurrence of multiple logistics quality conditions corresponding to each of the N entity objects to be recommended corresponding to the user of the client can be predicted, and the probability prediction value of the multiple logistics quality conditions corresponding to each entity object to be recommended can be obtained; further, according to the probability prediction value of the multiple logistics quality conditions corresponding to each entity object to be recommended by the user of the client, the N logistics fulfillment quality prediction values ​​corresponding to each entity object to be recommended by the user of the client are determined.

[0096] The embodiment of the present application uses multiple task networks, each of which can be used to predict the probability of occurrence of a logistics quality situation, and then determines the logistics fulfillment quality prediction value corresponding to each entity object to be recommended by the client user based on the predicted value of the occurrence probability of multiple logistics quality situations of each entity object to be recommended among N entity objects to be recommended corresponding to the client user, and comprehensively considers factors such as logistics pressure, merchant pressure and user location, so the prediction accuracy is high; under the condition of given logistics capacity, the user's fulfillment experience is optimized with very fine granularity, so as to achieve the goal of reducing the poor logistics quality of users while having little or no impact on the order volume.

[0097] In an embodiment of the present application, a possible implementation method is provided. In the above step A22, according to the predicted values ​​of the occurrence probabilities of multiple logistics quality situations corresponding to each entity object to be recommended by the user of the client, N logistics fulfillment quality prediction values ​​corresponding to each entity object to be recommended by the user of the client are determined, which may specifically include the following steps A221 and A222:

[0098] Step A221, obtaining the influence weights of various logistics quality conditions of each entity object to be recommended corresponding to the client user;

[0099] Step A222, determines the predicted value of the logistics fulfillment quality of each entity object to be recommended corresponding to the client user based on the influence weights of the various logistics quality conditions of each entity object to be recommended corresponding to the client user and the predicted value of the occurrence probability of the various logistics quality conditions of each entity object to be recommended corresponding to the client user.

[0100] It can be seen that the embodiment of the present application can flexibly adjust the influence weights of various logistics quality conditions of each entity object to be recommended where the client user is located according to actual conditions, so that the logistics fulfillment quality prediction value of each entity object to be recommended where the client user is located can meet the needs of actual conditions, improve the accuracy of logistics fulfillment quality, and enhance the user's fulfillment experience.

[0101] The present application provides a possible implementation method. In the above step S103, the list of recommended objects is reordered in combination with each first parameter to obtain a list of recommended entity objects. Specifically, the following steps B1 and B2 may be included:

[0102] Step B1, obtaining N second parameters about the environment state of each entity object to be recommended;

[0103] Step B2: re-sort the list of recommended objects by combining the first parameters and the second parameters to obtain a list of recommended entity objects.

[0104] In the above embodiment, the data about the environmental status of each entity object to be recommended can be weather, time period, whether it is a holiday, logistics pressure, order quantity, visit rate, etc. Based on these data, N second parameters about the environmental status of each entity object to be recommended can be obtained. In this way, under the condition of a certain logistics distribution capacity, the logistics fulfillment quality signal parameter (i.e., the first parameter) and the environmental status signal parameter (i.e., the second parameter) corresponding to the user of the client can be comprehensively considered and applied to the sorting of the merchant list of the entity object; through refined regulation, the sorting result takes into account the fulfillment experience and environmental factors, thereby guiding the user to place orders, and guiding the user to orders with higher value and better logistics experience; while improving the user's logistics experience, it can have little or no impact on the overall order volume.

[0105] A possible implementation method is provided in the embodiment of the present application. In the above step B2, the list of recommended objects is reordered in combination with each first parameter and each second parameter to obtain a list of recommended entity objects. Specifically, the following steps B21 and B22 may be included:

[0106] Step B21, pre-configuring a control reordering strategy, the factors of which include various first parameters and various second parameters;

[0107] Step B22: reorder the list of recommended objects using the regulation and reordering strategy to obtain a list of recommended entity objects.

[0108] In this embodiment, the control reordering strategy may have characteristics such as capacity separation, special delivery, express delivery, and self-delivery separation strategies; hierarchical nonlinear control, based on the relationship between logistics fulfillment quality signals, environmental status signals, and the recommended sorting signal itself, etc., as a nonlinear function; adaptive dynamic adjustment, automatic adjustment and adaptation for weekend and holiday suppression, etc. The embodiment of the present application does not limit this, and can be configured according to actual demand goals, and can comprehensively consider logistics fulfillment quality signals, environmental status signals, and the recommended sorting signal itself, such as Figure 1b The regulation and reordering strategy shown uses the regulation and reordering strategy to reorder multiple entity objects to be recommended to obtain a reordered list of recommended entity objects, which is more precise, accurate and efficient.

[0109] A possible implementation method is provided in an embodiment of the present application. In step B22 above, the list of objects to be recommended is reordered using the control reordering strategy to obtain a list of recommended entity objects. Specifically, the control reordering strategy may be used to determine the reordering value of each entity object to be recommended; and then the list of objects to be recommended is reordered according to the reordering value of each entity object to be recommended to obtain a list of recommended entity objects. The control reordering in this embodiment comprehensively considers the logistics fulfillment quality signal, the environmental status signal, and the signal of the recommendation sorting itself, and comprehensively outputs a reordering score to adjust the sorting. The entire process comprehensively considers real-time factors such as logistics, merchants, and users; the final user and merchant granularity logistics fulfillment experience is predicted in real time, and multiple factors are comprehensively considered in real-time recommendation sorting for fine traffic regulation.

[0110] Accordingly, the embodiment of the present application provides a method for sorting entity objects based on logistics fulfillment quality applied to a client, such as Figure 2 As shown, the entity object sorting method based on logistics fulfillment quality applied to the client may include the following steps S201 to S203:

[0111] Step S201, sending a request for obtaining a list of entity objects to a server, wherein the request carries user-related information of the client;

[0112] Step S202, receiving a reordered list of recommended entity objects provided by the server, wherein the server determines N first parameters of each entity object to be recommended corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended; reordering the list of recommended objects in combination with each first parameter to obtain a list of recommended entity objects, wherein N is greater than 1;

[0113] Step S203: display the re-ordered list of recommended entity objects.

[0114] In the embodiment of the present application, when the logistics distribution capacity is certain, the server can comprehensively consider the user feature data, logistics data, and feature data of each entity object to be recommended of the client, and apply them to the sorting of the merchant list of the entity object; through refined regulation, the sorting result takes into account the fulfillment experience and logistics factors, thereby guiding the user to place orders with higher value and better logistics experience; while improving the user's logistics experience, the overall order volume can be less affected or not affected.

[0115] In the embodiment of the present application, a possible implementation method is provided. In the above step S201, a request for obtaining a list of entity objects is sent to the server. Specifically, it can be detected that a preset control of the display interface is triggered, and a request for obtaining a list of entity objects is sent to the server. This is convenient for user operation, and a real-time list of entity objects can be obtained in a timely and accurate manner.

[0116] The above introduces Figure 1a and Figure 2 There are multiple implementation methods for each link in the illustrated embodiment. The entity object sorting method based on logistics fulfillment quality provided in the embodiment of the present application is further explained below through specific embodiments.

[0117] Figure 3 A flowchart of a method for sorting entity objects based on logistics fulfillment quality in combination with a client and a server provided in an embodiment of the present application is shown. Figure 3 In the method for sorting entity objects based on logistics fulfillment quality, the method may include the following steps S301 to S309.

[0118] Step S301: the client detects that a preset control on the display interface is triggered, and sends a request to the server to obtain a list of entity objects.

[0119] Step S302, the server responds to the client's request for a list of entity objects, obtains the client's user-related information and multiple entity objects to be recommended in real time, and obtains the client's user feature data, logistics data, and feature data of multiple entity objects to be recommended.

[0120] In step S303, the server inputs the client's user feature data, logistics data and feature data of multiple entity objects to be recommended into the multi-task logistics fulfillment quality prediction model, and uses the multiple task networks of the multi-task logistics fulfillment quality prediction model to predict the probability of occurrence of multiple logistics quality conditions for each of the multiple entity objects to be recommended corresponding to the client's user, and obtains the predicted value of the occurrence probability of multiple logistics quality conditions for each entity object to be recommended corresponding to the client's user.

[0121] In this step, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input. In addition, the multi-task logistics fulfillment quality prediction model can include multiple task networks, each of which can be used to predict the probability of occurrence of a logistics quality situation.

[0122] Step S304, the server determines the predicted value of logistics fulfillment quality corresponding to each entity object to be recommended to the user of the client according to the predicted values ​​of occurrence probabilities of various logistics quality situations corresponding to each entity object to be recommended to the user of the client.

[0123] In this step, the influence weights of the various logistics quality conditions of each entity object to be recommended corresponding to the client user can be obtained, and then the logistics fulfillment quality prediction value of each entity object to be recommended corresponding to the client user can be determined based on the influence weights of the various logistics quality conditions of each entity object to be recommended corresponding to the client user and the predicted probability value of the various logistics quality conditions of each entity object to be recommended corresponding to the client user.

[0124] In step S305, the server determines whether the predicted value of the logistics fulfillment quality of each entity object to be recommended to the user of the client exceeds a preset threshold. If so, execute step S306; if not, execute step S307.

[0125] Step S306, the service end downgrades the score of the entity object to be recommended whose predicted value of logistics fulfillment quality exceeds the preset threshold, and executes step S307.

[0126] Step S307, the server sorts multiple entity objects to be recommended according to the logistics fulfillment quality prediction value corresponding to each entity object to be recommended by the client user and one or more of the logistics resource parameters, environmental resource parameters, and preset sorting parameters of each entity object to be recommended, and obtains a sorted recommended entity object list.

[0127] Step S308: The server provides the sorted list of recommended entity objects to the client for display.

[0128] Step S309: the client receives the sorted recommended entity object list provided by the server, and displays it for browsing by the user of the client.

[0129] In the embodiment of the present application, the odcost strategy for predicting the quality of logistics fulfillment is to downgrade the ranking scores of the merchant stores before sorting, so as to achieve the purpose of ranking the stores with low probability of logistics fulfillment. Specifically, the server algorithm side will give each store a ranking score through fine sorting, and control the sorting of the store through this score. For example, the first preset number of stores of the fine sorting candidates request the logistics fulfillment quality signal odcost in real time to evaluate the probability of fulfillment; for each store logistics line to be sorted, such as special delivery, express delivery, self-delivery, etc., the odcost prediction value, whether it is a working day, etc., make a combined threshold judgment, and implement a downgrade strategy for stores with low probability of fulfillment; for stores that are judged to need weighting, the ranking score is downgraded in this way, and the final effect is that as the probability of failure to fulfill the contract increases, the intensity of re-sorting is accelerated. Afterwards, the ranking score is truncated, and the second preset number of stores at the front are selected for re-ranking, and finally pushed to the user of the client. Here, the first preset number or the second preset number can be set according to actual needs. It should be noted that the enumeration here is only illustrative and does not limit the embodiment of the present application.

[0130] In an optional embodiment, the list of merchants that the user is about to see is output after being processed by logistics fulfillment quality prediction and sorting control strategy. The specific process may be that multiple entity objects to be recommended are processed by logistics fulfillment quality prediction and sorting control strategy and then a sorted list of recommended entity objects is output.

[0131] First, the logistics fulfillment quality prediction can directly predict the possibility that the user will encounter poor logistics quality if he currently places an order at a certain merchant, that is, the probability of occurrence of various logistics quality situations. Specifically, the feature data of the client's user-related information, logistics data, and feature data of multiple entity objects to be recommended can be input into a multi-task logistics fulfillment quality prediction model, which is trained on a multi-task learning MMOE model using feature data of sample user-related information, sample logistics data, and feature data of multiple sample entity objects as input; the deep MMOE model is used to predict the possibility of logistics quality, including whether it is timed out, T15 timed out, T50 timed out, and cancelled, and the final score is the weighted average of these values. The feature data here may include data such as logistics capacity, business district, merchant entity address, cross-features of merchant entity and user location, environmental parameters, and user location, which are not limited in this embodiment of the present application.

[0132] Next, the control reordering comprehensively considers the logistics fulfillment quality prediction score, as well as the environmental status signal, the signal of the recommended sorting itself, etc., and comprehensively outputs a reordering score to adjust the sorting. In addition, the control reordering strategy has the following characteristics: separation of logistics capacity, separation of special delivery, express delivery, and self-delivery; hierarchical nonlinear control, nonlinear function based on the relationship between signals and logistics; adaptive dynamic adjustment, automatic adjustment and adaptation of reordering on weekends and holidays. The entire process comprehensively considers real-time factors such as logistics, merchants, and users; real-time prediction of the final user and merchant granularity of logistics fulfillment experience, and comprehensive consideration of multiple factors for fine traffic control in real-time recommendation sorting; while reducing the poor user logistics experience, it has little or no impact on the order volume.

[0133] It should be noted that, in practical applications, all possible implementation methods described above can be arbitrarily combined to form possible embodiments of the present application, which will not be described one by one here.

[0134] Based on the entity object sorting method based on logistics fulfillment quality provided in the above embodiments, based on the same inventive concept, the embodiment of the present application also provides a entity object sorting device based on logistics fulfillment quality.

[0135] Figure 4 The structure diagram of the physical object sorting device based on logistics performance quality applied to the service end provided by the embodiment of the present application is shown. Figure 4 As shown, the entity object sorting device based on logistics fulfillment quality applied to the service end may specifically include an acquisition module 410, a determination module 420, a sorting module 430 and a providing module 440.

[0136] The acquisition module 410 is used to obtain a list of proposed recommended objects for a user of the client, including N entity objects to be recommended, in response to a request from the client to obtain a list of entity objects, where N is greater than 1;

[0137] A determination module 420, configured to determine N first parameters of each entity object to be recommended corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended;

[0138] A sorting module 430 is used to re-sort the list of recommended objects in combination with each first parameter to obtain a list of recommended entity objects;

[0139] A module 440 is provided, which is used to provide the reordered list of recommended entity objects to the client for display.

[0140] A possible implementation method is provided in the embodiments of the present application. Figure 4 The illustrated determination module 420 is also used to:

[0141] Input the client's user feature data, logistics data, and feature data of each entity object to be recommended into the multi-task logistics fulfillment quality prediction model;

[0142] The multi-task logistics fulfillment quality prediction model is used to predict the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client, and N logistics fulfillment quality prediction values ​​corresponding to each entity object to be recommended are obtained as the N first parameters of each entity object to be recommended corresponding to the user of the client;

[0143] Among them, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input.

[0144] A possible implementation method is provided in an embodiment of the present application. The multi-task logistics fulfillment quality prediction model includes multiple task networks, each of which is used to predict the probability of occurrence of a logistics quality situation; Figure 4 The illustrated determination module 420 is also used to:

[0145] Using multiple task networks of the multi-task logistics fulfillment quality prediction model, the occurrence probabilities of various logistics quality situations corresponding to each entity object to be recommended for the client user are predicted, and the occurrence probability prediction values ​​of various logistics quality situations corresponding to each entity object to be recommended for the client user are obtained;

[0146] According to the predicted values ​​of occurrence probabilities of various logistics quality situations corresponding to each entity object to be recommended by the user of the client, N predicted values ​​of logistics fulfillment quality corresponding to each entity object to be recommended by the user of the client are determined.

[0147] A possible implementation method is provided in the embodiments of the present application. Figure 4 The illustrated determination module 420 is also used to:

[0148] Obtain the influence weights of various logistics quality conditions of each entity object to be recommended corresponding to the client user;

[0149] According to the influence weights of the various logistics quality conditions of the client's users corresponding to each entity object to be recommended and the predicted probability values ​​of the various logistics quality conditions of the client's users corresponding to each entity object to be recommended, the predicted value of the logistics fulfillment quality of each entity object to be recommended corresponding to the client's users is determined.

[0150] A possible implementation method is provided in the embodiments of the present application. Figure 4 The illustrated sorting module 430 is also used to:

[0151] Obtaining N second parameters of each entity object to be recommended regarding the environmental state;

[0152] The list of recommended objects is reordered in combination with each first parameter and each second parameter to obtain a list of recommended entity objects.

[0153] A possible implementation method is provided in the embodiments of the present application. Figure 4 The illustrated sorting module 430 is also used to:

[0154] Pre-configuring a control reordering strategy, wherein factors of the control reordering strategy include various first parameters and various second parameters;

[0155] The list of recommended objects is reordered using the regulation reordering strategy to obtain a list of recommended entity objects.

[0156] A possible implementation method is provided in the embodiments of the present application. Figure 4 The illustrated sorting module 430 is also used to:

[0157] Determine the re-ranking value of each entity object to be recommended by using the regulation re-ranking strategy;

[0158] The list of recommended objects is reordered according to the reordering value of each entity object to be recommended to obtain a list of recommended entity objects.

[0159] Figure 5 The structure diagram of the entity object sorting device based on logistics fulfillment quality applied to the client provided by the embodiment of the present application is shown. Figure 5 As shown, the physical object sorting device based on logistics fulfillment quality applied to the client may specifically include a sending module 510, a receiving module 520 and a display module 530.

[0160] A sending module 510 is used to send a request for obtaining a list of entity objects to a server, wherein the request carries user-related information of the client;

[0161] The receiving module 520 is used to receive the re-ordered recommended entity object list provided by the server, wherein the server determines N first parameters of each to-be-recommended entity object corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each to-be-recommended entity object; and re-orders the to-be-recommended object list in combination with each first parameter to obtain the recommended entity object list;

[0162] The display module 530 is used to display the re-ordered recommended entity object list.

[0163] A possible implementation method is provided in the embodiments of the present application. Figure 5 The sending module 510 is also shown to be used for:

[0164] It is detected that the preset control of the display interface is triggered, and a request for obtaining a list of entity objects is sent to the server.

[0165] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a processor and a memory, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any one of the above-mentioned embodiments of the entity object sorting method based on logistics fulfillment quality applied to the server or the entity object sorting method based on logistics fulfillment quality applied to the client.

[0166] Based on the same inventive concept, an embodiment of the present application also provides a storage medium, which stores a computer program, wherein the computer program is configured to execute any one of the above-mentioned embodiments of the entity object sorting method based on logistics fulfillment quality applied to the server or the entity object sorting method based on logistics fulfillment quality applied to the client when running.

[0167] Those skilled in the art can clearly understand that the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the aforementioned method embodiments, and for the sake of brevity, they are not further described here.

[0168] Those skilled in the art can understand that the technical solution of the present application can be essentially or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, which includes a number of program instructions, so that an electronic device (such as a personal computer, a server, or a network device, etc.) executes all or part of the steps of the method described in each embodiment of the present application when running the program instructions. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0169] Alternatively, all or part of the steps of implementing the aforementioned method embodiments may be accomplished by hardware associated with program instructions (such as electronic devices such as personal computers, servers, or network devices), and the program instructions may be stored in a computer-readable storage medium. When the program instructions are executed by a processor of an electronic device, the electronic device executes all or part of the steps of the methods described in the embodiments of the present application.

[0170] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that within the spirit and principles of the present application, they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate from the protection scope of the present application.

Claims

1. A method for sorting entity objects based on logistics fulfillment quality, characterized in that: include: In response to a request from a client to obtain a list of entity objects, obtaining a list of proposed recommended objects for a user of the client, comprising N entity objects to be recommended, where N is greater than 1; Based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, respectively determining N first parameters of each entity object to be recommended corresponding to the user of the client; Reordering the list of recommended objects based on the first parameters to obtain a list of recommended entity objects; Providing the reordered list of recommended entity objects to the client for display; Wherein, based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, respectively determining N first parameters of each entity object to be recommended corresponding to the user of the client includes: Inputting the user feature data, logistics data, and feature data of each entity object to be recommended of the client into a multi-task logistics fulfillment quality prediction model; Using the multi-task logistics fulfillment quality prediction model, predict the logistics fulfillment quality of each of the entity objects to be recommended corresponding to the user of the client, and obtain N logistics fulfillment quality prediction values ​​of each of the entity objects to be recommended corresponding to the user of the client, as N first parameters of each of the entity objects to be recommended corresponding to the user of the client; Among them, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input.

2. The method according to claim 1, characterized in that The multi-task logistics fulfillment quality prediction model includes multiple task networks, each task network is used to predict the probability of occurrence of a logistics quality situation; The multi-task logistics fulfillment quality prediction model is used to predict the logistics fulfillment quality of each entity object to be recommended corresponding to the user of the client, and N logistics fulfillment quality prediction values ​​corresponding to each entity object to be recommended corresponding to the user of the client are obtained, including: Utilizing the multiple task networks of the multi-task logistics fulfillment quality prediction model, predicting the occurrence probabilities of various logistics quality situations corresponding to each of the entity objects to be recommended for the user of the client, and obtaining predicted values ​​of the occurrence probabilities of various logistics quality situations corresponding to each of the entity objects to be recommended for the user of the client; According to the predicted values ​​of the occurrence probabilities of multiple logistics quality situations corresponding to each of the entity objects to be recommended by the user of the client, N logistics fulfillment quality prediction values ​​corresponding to each of the entity objects to be recommended by the user of the client are determined.

3. The method according to claim 2, characterized in that Determining N logistics fulfillment quality prediction values ​​corresponding to each of the entity objects to be recommended by the user of the client according to the predicted values ​​of occurrence probabilities of multiple logistics quality situations corresponding to each of the entity objects to be recommended by the user of the client includes: Obtaining respective influence weights of multiple logistics quality conditions of the client user corresponding to each of the entity objects to be recommended; Based on the influence weights of the various logistics quality conditions of each entity object to be recommended corresponding to the user of the client and the predicted probability values ​​of the occurrence of the various logistics quality conditions of each entity object to be recommended corresponding to the user of the client, the logistics fulfillment quality prediction value of each entity object to be recommended corresponding to the user of the client is determined.

4. The method according to any one of claims 1 to 3, characterized in that The list of recommended objects is reordered based on the first parameters to obtain a list of recommended entity objects, including: Obtaining N second parameters related to the environmental state of each of the entity objects to be recommended; The list of recommended objects is reordered in combination with the first parameters and the second parameters to obtain a list of recommended entity objects.

5. The method according to claim 4, characterized in that The list of recommended objects is reordered in combination with the first parameters and the second parameters to obtain a list of recommended entity objects, including: Pre-configuring a control reordering strategy, wherein factors of the control reordering strategy include the first parameters and the second parameters; The list of objects to be recommended is reordered using the regulation and reordering strategy to obtain a list of recommended entity objects.

6. The method according to claim 5, characterized in that The list of recommended objects is reordered using the control reordering strategy to obtain a list of recommended entity objects, including: Determining the reordering value of each of the entity objects to be recommended by using the regulation and reordering strategy; The proposed recommended object list is reordered according to the reordering value of each of the to-be-recommended entity objects to obtain a recommended entity object list.

7. A method for sorting entity objects based on logistics fulfillment quality, characterized in that: include: Send a request to the server to obtain a list of entity objects, wherein the request carries user-related information of the client; receiving a reordered list of recommended entity objects provided by the server, wherein the server determines, based on user characteristic data of the client, logistics data, and characteristic data of each entity object to be recommended, N first parameters corresponding to each entity object to be recommended for the user of the client; and reordering the list of recommended objects based on the first parameters to obtain a list of recommended entity objects; Display the re-sorted list of recommended entity objects; The server determines N first parameters of each entity object to be recommended corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, including: Inputting the user feature data, logistics data, and feature data of each entity object to be recommended of the client into a multi-task logistics fulfillment quality prediction model; Using the multi-task logistics fulfillment quality prediction model, predict the logistics fulfillment quality of each of the entity objects to be recommended corresponding to the user of the client, and obtain N logistics fulfillment quality prediction values ​​of each of the entity objects to be recommended corresponding to the user of the client, as N first parameters of each of the entity objects to be recommended corresponding to the user of the client; Among them, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input.

8. A physical object sorting device based on logistics performance quality, characterized in that: include: an acquisition module, configured to, in response to a request from a client for an entity object list, acquire a list of proposed recommended objects for a user of the client, comprising N entity objects to be recommended, where N is greater than 1; a determination module, configured to determine N first parameters of each entity object to be recommended corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended; a sorting module, configured to re-sort the list of proposed recommended objects based on the first parameters to obtain a list of recommended entity objects; Providing a module for providing the reordered list of recommended entity objects to the client for display; Wherein, the determining module is further used for: Inputting the user feature data, logistics data, and feature data of each entity object to be recommended of the client into a multi-task logistics fulfillment quality prediction model; Using the multi-task logistics fulfillment quality prediction model, predict the logistics fulfillment quality of each of the entity objects to be recommended corresponding to the user of the client, and obtain N logistics fulfillment quality prediction values ​​of each of the entity objects to be recommended corresponding to the user of the client, as N first parameters of each of the entity objects to be recommended corresponding to the user of the client; Among them, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input.

9. A physical object sorting device based on logistics performance quality, characterized in that: include: A sending module, configured to send a request to the server for obtaining a list of entity objects, wherein the request carries user-related information of the client; a receiving module configured to receive a reordered list of recommended entity objects provided by the server, wherein the server determines, based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, N first parameters corresponding to each entity object to be recommended for the user of the client; and reorders the list of recommended objects based on the first parameters to obtain a list of recommended entity objects; The display module is used to display the re-ordered list of recommended entity objects; The server determines N first parameters of each entity object to be recommended corresponding to the user of the client based on the user characteristic data of the client, the logistics data, and the characteristic data of each entity object to be recommended, including: Inputting the user feature data, logistics data, and feature data of each entity object to be recommended of the client into a multi-task logistics fulfillment quality prediction model; Using the multi-task logistics fulfillment quality prediction model, predict the logistics fulfillment quality of each of the entity objects to be recommended corresponding to the user of the client, and obtain N logistics fulfillment quality prediction values ​​of each of the entity objects to be recommended corresponding to the user of the client, as N first parameters of each of the entity objects to be recommended corresponding to the user of the client; Among them, the multi-task logistics fulfillment quality prediction model is obtained by training the multi-task learning MMOE model with sample user feature data, sample logistics data and feature data of multiple sample entity objects as input.

10. An electronic device, characterized in that: It includes a processor and a memory, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the entity object sorting method based on logistics fulfillment quality described in any one of claims 1 to 6 or the entity object sorting method based on logistics fulfillment quality described in claim 7.

11. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the entity object sorting method based on logistics fulfillment quality described in any one of claims 1 to 6 or the entity object sorting method based on logistics fulfillment quality described in claim 7 during runtime.

Images