CN112862541A

CN112862541A - Waybill creating method and device and electronic equipment

Info

Publication number: CN112862541A
Application number: CN202110252009.1A
Authority: CN
Inventors: 张鹏
Original assignee: Lazas Network Technology Shanghai Co Ltd
Current assignee: Lazas Network Technology Shanghai Co Ltd
Priority date: 2021-03-08
Filing date: 2021-03-08
Publication date: 2021-05-28

Abstract

One or more embodiments of the present specification provide a waybill creating method, an apparatus, and an electronic device, where the method includes: receiving a waybill establishing request corresponding to a target order; in response to the waybill creating request, extracting preset pricing features from order data in the waybill creating request, and respectively inputting the pricing features into a first prediction model corresponding to the distance features, a second prediction model corresponding to the weight features, a third prediction model corresponding to the passenger order features and a fourth prediction model corresponding to the natural attribute features; the prediction model is a machine learning model which is obtained by training a preset machine learning model corresponding to the target category based on historical waybill data and is trained; and creating the freight bill corresponding to the target order based on the freight bill price predicted by the first prediction model, the freight bill price predicted by the second prediction model, the freight bill price predicted by the third prediction model and the freight bill price predicted by the fourth prediction model.

Description

Waybill creating method and device and electronic equipment

Technical Field

One or more embodiments of the present disclosure relate to the technical field of computer applications, and in particular, to a waybill creation method, apparatus, and electronic device.

Background

Nowadays, with the development of internet technology, online shopping is more and more popular. The user can not only buy goods from merchants far away through the internet, but also can buy goods from merchants near, for example: the user can use the takeout APP (Application program) to purchase goods such as food from a restaurant near the user's location position, or purchase goods such as living goods from a supermarket near the user's location position, and logistics personnel (such as a takeout rider) distribute the goods purchased by the user to the user, so as to avoid the trouble that the user needs to go to an off-line physical shop to purchase goods by himself.

Generally, after a user purchases some goods on the network, an order corresponding to the goods is created for the user, and a freight bill corresponding to the order is further created, so as to allocate logistics personnel to the freight bill, so that the allocated logistics personnel can allocate the goods purchased by the user to the user according to the freight bill and obtain a reward with an amount equal to the allocation fee (i.e. freight bill price of the freight bill) agreed in the freight bill. Under the circumstance, how to reasonably price the freight note becomes a problem to be solved urgently.

Disclosure of Invention

The present specification proposes a waybill creation method, which includes:

receiving a waybill establishing request corresponding to a target order; wherein the waybill creation request comprises order data of the target order;

in response to the waybill creating request, extracting preset pricing features from order data in the waybill creating request, and inputting the pricing features into a first prediction model corresponding to a distance feature, a second prediction model corresponding to a weight feature, a third prediction model corresponding to a passenger order feature and a fourth prediction model corresponding to a natural attribute feature respectively, so that a first waybill price corresponding to the distance feature is predicted by the first prediction model, a second waybill price corresponding to the weight feature is predicted by the second prediction model, a third waybill price corresponding to the passenger order feature is predicted by the third prediction model, and a fourth waybill price corresponding to the natural attribute feature is predicted by the fourth prediction model; the prediction model is a machine learning model which is obtained by training a preset machine learning model corresponding to the target category based on historical waybill data and is trained;

and creating the freight bill corresponding to the target order based on the predicted first freight bill price, the second freight bill price, the third freight bill price and the fourth freight bill price.

Optionally, the method further comprises:

acquiring historical waybill data of a transaction moment in a preset time period; wherein the historical waybill data comprises the pricing characteristics and pricing details; wherein the price details include the first waybill price, the second waybill price, the third waybill price, and the fourth waybill price;

creating a first training sample set corresponding to the distance feature based on the pricing feature and the first waybill price in the historical waybill data; creating a second set of training samples corresponding to the weight features based on the pricing features and the second waybill price in the historical waybill data; creating a third training sample set corresponding to the customer order price feature based on the pricing feature and the third waybill price in the historical waybill data; creating a fourth training sample set corresponding to the natural attribute features based on the pricing features and the fourth waybill prices in the historical waybill data;

training a preset machine learning model corresponding to the distance features based on the first training sample set, and determining the trained machine learning model as the first prediction model; training a preset machine learning model corresponding to the weight features based on the second training sample set, and determining the machine learning model after training as the second prediction model; training a preset machine learning model corresponding to the passenger order features based on the third training sample set, and determining the trained machine learning model as the third prediction model; and training a preset machine learning model corresponding to the natural attribute characteristics based on the fourth training sample set, and determining the machine learning model after training as the fourth prediction model.

Optionally, the prediction model is an offline prediction model.

Optionally, the extracting, in response to the waybill creation request, preset pricing features from order data in the waybill creation request, and inputting the pricing features into a first prediction model corresponding to a distance feature, a second prediction model corresponding to a weight feature, a third prediction model corresponding to a passenger order feature, and a fourth prediction model corresponding to a natural attribute feature, respectively, includes:

responding to the freight note creating request, and pricing freight notes corresponding to the target orders according to a preset default pricing rule;

if the pricing fails, extracting preset pricing features from order data in the waybill creating request, and inputting the pricing features into an offline first prediction model corresponding to the distance features, an offline second prediction model corresponding to the weight features, an offline third prediction model corresponding to the passenger order features and an offline fourth prediction model corresponding to the natural attribute features respectively.

Optionally, the creating the waybill corresponding to the target order based on the predicted first waybill price, the predicted second waybill price, the predicted third waybill price, and the predicted fourth waybill price includes:

calculating the sum of the predicted first freight note price, the predicted second freight note price, the predicted third freight note price and the predicted fourth freight note price, and determining the calculated sum as the freight note total price of the freight note corresponding to the target order;

and creating the freight note corresponding to the target order based on the freight note total price.

Optionally, the natural attribute features include one or more of the following: a geographic region characteristic; a weather characteristic; a cargo level characteristic.

Optionally, the machine learning model is a linear model.

This specification also proposes a waybill creation device, the device comprising:

the receiving module is used for receiving the waybill establishing request corresponding to the target order; wherein the waybill creation request comprises the target order data;

a pricing module, configured to extract preset pricing features from order data in the waybill creating request in response to the waybill creating request, and input the pricing features into a first prediction model corresponding to a distance feature, a second prediction model corresponding to a weight feature, a third prediction model corresponding to a waybill price feature, and a fourth prediction model corresponding to a natural attribute feature, respectively, so that the first prediction model predicts a first waybill price corresponding to the distance feature, the second prediction model predicts a second waybill price corresponding to the weight feature, the third prediction model predicts a third waybill price corresponding to the waybill price feature, and the fourth prediction model predicts a fourth waybill price corresponding to the natural attribute feature; the prediction model is a machine learning model which is obtained by training a preset machine learning model corresponding to the target category based on historical waybill data and is trained;

a first creating module for creating the waybills corresponding to the target orders based on the predicted first waybill price, the second waybill price, the third waybill price, and the fourth waybill price.

Optionally, the apparatus further comprises:

the acquisition module is used for acquiring historical waybill data of a transaction moment in a preset time period; wherein the historical waybill data comprises pricing characteristics and pricing detail; wherein the price details include the first waybill price, the second waybill price, the third waybill price, and the fourth waybill price;

a second creating module, configured to create a first training sample set corresponding to the distance feature based on the pricing feature and the first waybill price in the historical waybill data; creating a second set of training samples corresponding to the weight features based on the pricing features and the second waybill price in the historical waybill data; creating a third training sample set corresponding to the customer order price feature based on the pricing feature and the third waybill price in the historical waybill data; creating a fourth training sample set corresponding to the natural attribute features based on the pricing features and the fourth waybill prices in the historical waybill data;

the training module is used for training a preset machine learning model corresponding to the distance feature based on the first training sample set and determining the trained machine learning model as the first prediction model; training a preset machine learning model corresponding to the weight features based on the second training sample set, and determining the machine learning model after training as the second prediction model; training a preset machine learning model corresponding to the passenger order features based on the third training sample set, and determining the trained machine learning model as the third prediction model; and training a preset machine learning model corresponding to the natural attribute characteristics based on the fourth training sample set, and determining the machine learning model after training as the fourth prediction model.

Optionally, the prediction model is an offline prediction model.

Optionally, the pricing module is specifically configured to:

Optionally, the first creating module is specifically configured to:

Optionally, the machine learning model is a linear model.

This specification also proposes an electronic device including:

a processor;

a memory for storing processor-executable instructions;

wherein the processor implements the steps of the method as described in any one of the above by executing the executable instructions.

The present specification also proposes a computer-readable storage medium having stored thereon computer instructions which, when executed by a processor, carry out the steps of the method according to any one of the preceding claims.

In the above technical solution, when receiving the waybill creating request, pricing features may be extracted from order data in the waybill creating request, and the extracted pricing features may be input to a first prediction model corresponding to a distance feature, a second prediction model corresponding to a weight feature, a third prediction model corresponding to a waybill price feature, and a fourth prediction model corresponding to a natural attribute feature, respectively, so that a first waybill price corresponding to the distance feature is predicted by the first prediction model, a second waybill price corresponding to the weight feature is predicted by the second prediction model, a third waybill price corresponding to the waybill price feature is predicted by the third prediction model, a fourth waybill price corresponding to the natural attribute feature is predicted by the fourth prediction model, and then, according to the predicted first waybill price, second waybill price, and third waybill price, and the fourth waybill price creates a waybill. By adopting the mode, the freight note prices corresponding to the four characteristics of the distance, the weight, the passenger unit price and the natural attribute are respectively predicted, so that the freight notes can be priced from a plurality of pricing dimensions, and the reasonable degree of freight note pricing can be ensured.

Drawings

FIG. 1 is a schematic diagram of a business system shown in an exemplary embodiment of the present description;

FIG. 2 is a schematic diagram of a waybill creation method shown in an exemplary embodiment of the present description;

FIG. 3 is a flow chart of another waybill creation method shown in an exemplary embodiment of the present description;

fig. 4 is a hardware structure diagram of an electronic device in which an waybill creating apparatus is located according to an exemplary embodiment of the present specification;

fig. 5 is a block diagram of an waybill creation device shown in an exemplary embodiment of the present specification.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of one or more embodiments of the specification, as detailed in the claims which follow.

It should be noted that: in other embodiments, the steps of the corresponding methods are not necessarily performed in the order shown and described herein. In some other embodiments, the method may include more or fewer steps than those described herein. Moreover, a single step described in this specification may be broken down into multiple steps for description in other embodiments; multiple steps described in this specification may be combined into a single step in other embodiments.

Referring to fig. 1, fig. 1 is a schematic diagram of a service system according to an exemplary embodiment of the present disclosure.

As shown in fig. 1, in the business system, a user client may be installed on an electronic device used by a user, and the user may shop online using a user interface provided by the user client. After the user completes the selection of the goods, the user client may create an order corresponding to the selected goods for the user and prompt the user to pay for the created order. After the user completes the payment, the user client may send the order to the server.

When the server receives the order sent by the user client, on one hand, the order can be sent to the merchant client, so that the merchant can prepare goods corresponding to the goods according to the goods purchased by the user in the order (for example, assuming that the goods purchased by the user are orange, the merchant can prepare orange as the goods corresponding to the goods). On the other hand, the server may further create an invoice corresponding to the order, where the created invoice may include information such as goods selected by the user, a delivery distance, a weight of the goods, a calculated price of the invoice (i.e., an amount of compensation obtained by the logistics worker after completing delivery according to the invoice), and the like; subsequently, the server side can send the waybill to the logistics client side, so that logistics personnel can check the waybill and decide whether to take the waybill or not based on the information in the waybill (namely, whether to distribute goods for the user according to the waybill and obtain payment or not).

Under the circumstance, if the freight note to be created cannot be priced normally, and the created freight note has no relevant information of the freight note price, logistics personnel cannot decide whether to take the freight note, so that logistics business cannot be executed normally, and a user cannot receive commodities purchased through online shopping, thereby seriously affecting user experience.

The present specification aims to provide a method for creating a waybill, comprising the steps of extracting pricing features from order data in a waybill creating request, inputting the extracted pricing features into a first prediction model corresponding to a distance feature, a second prediction model corresponding to a weight feature, a third prediction model corresponding to a waybill price feature, and a fourth prediction model corresponding to a natural attribute feature, predicting a first waybill price corresponding to the distance feature by the first prediction model, predicting a second waybill price corresponding to the weight feature by the second prediction model, predicting a third waybill price corresponding to the waybill price feature by the third prediction model, predicting a fourth waybill price corresponding to the natural attribute feature by the fourth prediction model, and then predicting the first waybill price, the second waybill price, and the third waybill price according to the predicted first waybill price, the second waybill price, and the predicted third waybill price, and the fourth waybill price creates the technical scheme of the waybill.

In particular implementations, the server may receive an order creation request corresponding to the target order. The waybill creating request comprises order data of a target order.

When the waybill creating request is received, preset pricing features can be extracted from order data in the waybill creating request in response to the waybill creating request. Then, the pricing characteristics may be calculated by inputting the pricing characteristics to a first prediction model corresponding to the distance characteristics, a second prediction model corresponding to the weight characteristics, a third prediction model corresponding to the passenger order characteristics, and a fourth prediction model corresponding to the natural attribute characteristics. In this case, the calculation result output by the first prediction model may be used as the first waybill price corresponding to the distance feature predicted by the first prediction model; taking the calculation result output by the second prediction model as a second waybill price predicted by the second prediction model and corresponding to the weight characteristic; taking a calculation result output by the third prediction model as a third freight bill price predicted by the third prediction model and corresponding to the passenger bill price characteristic; and taking the calculation result output by the fourth prediction model as a fourth waybill price corresponding to the natural attribute characteristic predicted by the fourth prediction model.

When the four freight note prices are predicted by the four prediction models, respectively, freight notes corresponding to the target order may be created based on the predicted first freight note price, second freight note price, third freight note price, and fourth freight note price.

Referring to fig. 2, fig. 2 is a flowchart illustrating a waybill creation method according to an exemplary embodiment of the present disclosure.

In conjunction with the business system shown in fig. 1, the waybill creation method can be applied to a server in the business system; the waybill creating method can comprise the following steps:

step 201, receiving a waybill creating request corresponding to a target order; wherein the waybill creation request comprises order data of the target order;

step 202, in response to the waybill creating request, extracting preset pricing features from order data in the waybill creating request, and inputting the pricing features into a first prediction model corresponding to distance features, a second prediction model corresponding to weight features, a third prediction model corresponding to passenger order features, and a fourth prediction model corresponding to natural attribute features respectively, so that the first prediction model predicts a first waybill price corresponding to the distance features, the second prediction model predicts a second waybill price corresponding to the weight features, the third prediction model predicts a third waybill price corresponding to the passenger order features, and the fourth prediction model predicts a fourth waybill price corresponding to the natural attribute features; the prediction model is a machine learning model which is obtained by training a preset machine learning model corresponding to the target category based on historical waybill data and is trained;

step 203, creating the waybill corresponding to the target order based on the predicted first waybill price, the predicted second waybill price, the predicted third waybill price, and the predicted fourth waybill price.

In this embodiment, for a certain created order (called a target order), a waybill corresponding to the target order needs to be further created, so that the logistics worker can perform cargo delivery for the target order according to the created waybill. In this case, the server may generate an waybill creation request corresponding to the target order by itself when receiving the target order sent by the user client; or, the client may directly send an waybill creation request corresponding to the target order to the server, where the sent waybill creation request includes the target order.

It should be noted that the waybill creation request includes order data of the target order. The order data may include, among other things: the account name of the ordering user; the time when the user places an order; the name of the item purchased by the user; a total amount of the goods purchased by the user; the total weight of the goods purchased by the user; a value rating of the goods purchased by the user; a shipping address of a merchant selling goods purchased by the user; the shipping address of the user; the distance between the shipping address and the receiving address; a geographic region of the shipping address; the merchant's current customer price (i.e., the average amount of items purchased by each user at the merchant); weather at the current time; and so on.

In practical application, the server may also obtain data of the current customer unit price, the current weather at the current time, and the like of the merchant corresponding to the target order as order data of the target order when receiving the waybill creation request, which is not limited in this specification.

In this embodiment, when the waybill creation request is received, a preset pricing feature may be extracted from order data in the waybill creation request in response to the waybill creation request.

It should be noted that the pricing characteristics are characteristic data that can affect the price of the waybill to be created, such as: distance (i.e., the distance between the shipping address and the receiving address, typically the farther the distance, the higher the price of the invoice); weight (i.e., the total weight of the goods purchased by the ordering user, typically the heavier the goods the higher the price of the invoice); the price of the passenger (generally, the higher the price of the passenger, the higher the price of the freight note); a geographic region (specifically, the geographic region where the receiving address of the ordering user is located, such as a business circle, and generally, the more prosperous the business circle, the lower the freight rate price); weather (often the price of the waybill is high in severe weather such as rain and snow); the goods grade (specifically, the value grade of the goods can be mentioned, and generally, the higher the value grade of the goods is, the higher the price of the freight note is); and so on.

In practical application, because the four characteristics of distance, weight, passenger order price and natural attributes have a large influence on the price of the waybill, the waybill price corresponding to the distance characteristic, the waybill price corresponding to the weight characteristic, the waybill price corresponding to the passenger order price and the waybill price corresponding to the natural attributes can be respectively determined for the waybill to be created, and the waybill price of the waybill to be created is determined based on the four waybill prices, so that the waybill can be priced from multiple pricing dimensions, and the reasonable degree of the waybill pricing is ensured.

In one illustrated embodiment, the natural attribute features described above may include one or more of the following: a geographic region characteristic; a weather characteristic; a cargo level characteristic.

In the present embodiment, after the pricing feature is extracted, the pricing feature may be input to a prediction model corresponding to the distance feature (hereinafter referred to as a first prediction model), a prediction model corresponding to the weight feature (hereinafter referred to as a second prediction model), a prediction model corresponding to the passenger-unit feature (hereinafter referred to as a third prediction model), and a prediction model corresponding to the natural attribute feature (hereinafter referred to as a fourth prediction model), respectively, and calculated. In this case, the calculation result output by the first prediction model may be used as the waybill price (hereinafter referred to as a first waybill price) corresponding to the distance feature predicted by the first prediction model; using the calculation result output by the second prediction model as the price of the waybill (hereinafter referred to as the second waybill price) corresponding to the weight characteristic predicted by the second prediction model; taking the calculation result output by the third prediction model as the freight bill price (hereinafter referred to as third freight bill price) corresponding to the passenger bill price characteristic predicted by the third prediction model; the calculation result output by the fourth prediction model is used as the freight note price (hereinafter referred to as the fourth freight note price) predicted by the fourth prediction model and corresponding to the natural attribute feature.

In practical applications, a pricing feature corresponding to the distance feature, a pricing feature corresponding to the weight feature, a pricing feature corresponding to the customer order feature, and a pricing feature corresponding to the natural attribute feature may be extracted from the order data in the waybill creation request. In this case, the pricing feature corresponding to the distance feature may be input to the first prediction model, and the first freight-order price may be predicted by the first prediction model based on the pricing feature corresponding to the distance feature; inputting the pricing feature corresponding to the weight feature to the second prediction model, and predicting the second waybill price by the second prediction model based on the pricing feature corresponding to the weight feature; inputting the pricing feature corresponding to the passenger order feature to the third prediction model, and predicting the third freight order price by the third prediction model based on the pricing feature corresponding to the passenger order feature; the pricing feature corresponding to the natural attribute feature is input to the fourth prediction model, and the fourth freight rate price is predicted by the fourth prediction model based on the pricing feature corresponding to the natural attribute feature.

In this embodiment, when the four freight note prices are predicted by the four prediction models, respectively, the freight note corresponding to the target order may be created based on the predicted first freight note price, the predicted second freight note price, the predicted third freight note price, and the predicted fourth freight note price.

In one embodiment, the sum of the predicted first, second, third and fourth consignments prices may be calculated, and the calculated sum may be used as the total consignment price of the consignment corresponding to the target order. Subsequently, an invoice corresponding to the target order may be created based on the total price of the invoice.

Specifically, when the first waybill price, the second waybill price, the third waybill price, and the fourth waybill price are predicted, the total waybill price may be calculated as the first waybill price + the second waybill price + the third waybill price + the fourth waybill price. Subsequently, an invoice may be created based on the invoice total price.

In practical application, after the waybill is created, the created waybill can be sent to the logistics client, so that the logistics client outputs the waybill to logistics personnel, the logistics personnel can check the waybill, and whether to take the waybill or not is determined based on information in the waybill. It should be noted that, the total price of the waybill and the prices of the four waybill can be displayed simultaneously in the waybill; it is also possible to display only the total price of the waybill; this is not limited by the present description.

In one embodiment shown, the waybill pricing creation process can be used as an alternative to avoid the problem that the waybill to be created cannot be priced.

Specifically, when the waybill creating request is received, pricing may be performed on the waybill to be created corresponding to the target order based on the pricing features in the waybill creating request according to a preset default pricing rule (specifically, preset by a technician according to actual needs).

If the pricing is successful, a waybill corresponding to the target order can be created directly based on the pricing.

However, if the pricing fails, the pricing features may be extracted from the order data in the waybill creation request, and the extracted pricing features are respectively input into the four prediction models, so that the four waybill prices are respectively predicted by the four prediction models, and then the waybill is created according to the predicted four waybill prices; that is, the steps 102 and 103 may be continuously adopted to price the waybill corresponding to the target order to be created, so as to ensure the execution stability of the waybill creation, and avoid the problem that the waybill to be created cannot be priced, so that the subsequent logistics personnel cannot receive the waybill.

It should be noted that the prediction model may be an offline prediction model; that is, the prediction model may be a machine learning model trained in advance, and may not be retrained based on the prediction result in the course of performing the waybill price prediction.

To obtain the above prediction model, please refer to fig. 3, and fig. 3 is a flowchart illustrating another waybill creation method according to an exemplary embodiment of the present disclosure.

In conjunction with the business system shown in fig. 1, the waybill creation method can also be applied to a server in the business system; the waybill creating method can comprise the following steps:

step 301, acquiring historical waybill data of a transaction moment in a preset time period; wherein the historical waybill data comprises the pricing characteristics and pricing details; wherein the price details include the first waybill price, the second waybill price, the third waybill price, and the fourth waybill price;

step 302, based on the pricing features and the first waybill price in the historical waybill data, creating a first training sample set corresponding to the distance features; creating a second set of training samples corresponding to the weight features based on the pricing features and the second waybill price in the historical waybill data; creating a third training sample set corresponding to the customer order price feature based on the pricing feature and the third waybill price in the historical waybill data; creating a fourth training sample set corresponding to the natural attribute features based on the pricing features and the fourth waybill prices in the historical waybill data;

step 303, training a preset machine learning model corresponding to the distance feature based on the first training sample set, and determining the machine learning model after training as the first prediction model; training a preset machine learning model corresponding to the weight features based on the second training sample set, and determining the machine learning model after training as the second prediction model; training a preset machine learning model corresponding to the passenger order features based on the third training sample set, and determining the trained machine learning model as the third prediction model; and training a preset machine learning model corresponding to the natural attribute characteristics based on the fourth training sample set, and determining the machine learning model after training as the fourth prediction model.

In this embodiment, historical waybill data of a deal time within a preset time period may be obtained first. Wherein, the historical waybill data can comprise the pricing characteristics and the price details; the price details may include the first waybill price corresponding to the distance feature, the second waybill price corresponding to the weight feature, the third waybill price corresponding to the passenger-fare feature, and the fourth waybill price corresponding to the natural-attribute feature.

It should be noted that the preset time period may be preset by a technician according to actual requirements, for example: a time period within 30 minutes before the current time may be used as the preset time period (assuming that the current time is 13:00, a time period of 12:30 to 13:00 may be used as the preset time period).

In practical application, for a certain historical waybill created, when a logistics worker performs a waybill receiving operation on the waybill, the waybill can be regarded as a committed waybill; that is, the time when the logistics personnel perform order receiving operation on the waybill can be used as the transaction time of the historical waybill.

In this embodiment, when the historical waybill data is obtained, the pricing feature and the first waybill price may be extracted from the historical waybill data to create a training sample set (hereinafter referred to as a first training sample set) corresponding to the distance feature; extracting the pricing characteristic and the second waybill price from the historical waybill data to create a training sample set (hereinafter referred to as a second training sample set) corresponding to the weight characteristic; extracting the pricing characteristic and the third waybill price from the historical waybill data to create a training sample set (hereinafter referred to as a third training sample set) corresponding to the passenger waybill price characteristic; the pricing feature and the fourth waybill price are extracted from the historical waybill data to create a training sample set (hereinafter referred to as a fourth training sample set) corresponding to the natural attribute feature.

It should be noted that, if only the total price of the historical waybill is available in a certain historical waybill, but there are no waybill prices corresponding to the above four categories, the technician may label the waybill with the waybill prices corresponding to the four categories according to the total price of the historical waybill.

In practical applications, on one hand, a pricing feature corresponding to the distance feature, a pricing feature corresponding to the weight feature, a pricing feature corresponding to the customer order feature, and a pricing feature corresponding to the natural attribute feature may be extracted from the historical waybill data; on the other hand, the first waybill price, the second waybill price, the third waybill price, and the fourth waybill price are extracted, respectively. In this case, the first training sample set may be created based on the pricing feature corresponding to the distance feature and the first waybill price; the second training sample set is created based on the pricing feature corresponding to the weight feature and the second waybill price, the third training sample set is created based on the pricing feature corresponding to the passenger waybill price feature and the third waybill price, and the fourth training sample set is created based on the pricing feature corresponding to the natural attribute feature and the fourth waybill price.

The training sample set a corresponding to the distance feature may be as shown in table 1 below:

training sample	Pricing features	Price of waybill
			Training sample A1	Pricing feature A1	Freight order price A1
Training sample A2	Pricing feature A2	Freight order price A2
			……	……	……

TABLE 1

Specifically, a training sample a1 may be generated based on the pricing feature a1 corresponding to the distance feature and the waybill price a1 corresponding to the distance feature extracted from the created historical waybill 1; generating a training sample A2 based on the pricing characteristic A2 corresponding to the distance characteristic and the freight note price A2 corresponding to the distance characteristic extracted from the created historical freight notes 2; and so on.

The training sample set B corresponding to the above-mentioned weight features may be as shown in table 2 below:

training sample	Pricing features	Price of waybill
			Training sample B1	Pricing feature B1	Freight order price B1
Training sample B2	Pricing feature B2	Freight order price B2
			……	……	……

TABLE 2

Specifically, a training sample B1 may be generated based on the pricing feature B1 corresponding to the weight feature and the waybill price B1 corresponding to the weight feature extracted from the created historical waybill 1; generating a training sample B2 based on the pricing characteristic B2 corresponding to the weight characteristic and the waybill price B2 corresponding to the weight characteristic extracted from the created historical waybill 2; and so on.

The training sample set C corresponding to the above passenger order features may be as shown in table 3 below:

training sample	Pricing features	Price of waybill
			Training sample C1	Pricing feature C1	Freight order price C1
Training sample C2	Pricing feature C2	Freight order price C2
			……	……	……

TABLE 3

Specifically, a training sample C1 may be generated based on the pricing feature C1 corresponding to the customer price feature and the waybill price C1 corresponding to the customer price feature extracted from the created history waybill 1; generating a training sample C2 based on a pricing characteristic C2 corresponding to the passenger order price characteristic and a freight order price C2 corresponding to the passenger order price characteristic, which are extracted from the created historical freight notes 2; and so on.

The training sample set D corresponding to the above natural attribute features may be as shown in table 4 below:

TABLE 4

Specifically, a training sample D1 may be generated based on the pricing feature D1 corresponding to the natural attribute feature and the waybill price D1 corresponding to the natural attribute feature extracted from the created historical waybill 1; generating a training sample D2 based on the pricing feature D2 corresponding to the natural attribute feature and the waybill price D2 corresponding to the natural attribute feature extracted from the created historical waybill 2; and so on.

In this embodiment, when the first training sample set, the second training sample set, the third training sample set, and the fourth training sample set are created, the first training sample set may be input to a preset machine learning model (which may be preset by a technician according to actual needs) corresponding to the distance feature, and model parameters of the machine learning model may be adjusted according to a calculation result output by the machine learning model, so as to implement supervised training on the machine learning model by using pricing features in the first training sample set as features and using waybill prices in the first training sample set as labels; subsequently, the trained machine learning model may be determined as the first prediction model corresponding to the distance feature. Similarly, a second training sample set may be input to a preset machine learning model corresponding to the weight feature, so as to implement supervised training of the machine learning model based on the second training sample set, and determine the machine learning model after training as the second prediction model corresponding to the weight feature; inputting a third training sample set into a preset machine learning model corresponding to the passenger order characteristic to realize supervised training of the machine learning model based on the third training sample set, and determining the trained machine learning model as the third prediction model corresponding to the passenger order characteristic; inputting a fourth training sample set to a preset machine learning model corresponding to the natural attribute features to realize supervised training of the machine learning model based on the fourth training sample set, and determining the machine learning model after training as the fourth prediction model corresponding to the natural attribute features.

For example, four machine learning models can be preset, which are: a machine learning model A corresponding to the weight characteristics; a machine learning model B corresponding to the weight characteristics; a machine learning model C corresponding to the guest unit price feature; and a machine learning model D corresponding to the natural attribute features. In this case, a machine learning model a may be trained based on the training sample set a, and the trained machine learning model a may be determined as the first prediction model; training a machine learning model B based on the training sample set B, and determining the trained machine learning model B as the second prediction model; training a machine learning model C based on the training sample set C, and determining the trained machine learning model C as the third prediction model; and training a machine learning model D based on the training sample set D, and determining the trained machine learning model D as the fourth prediction model.

In one illustrated embodiment, the machine learning model may be a linear model.

Continuing with the above example as an example, the machine learning model a may specifically be a linear model as shown below:

wherein, distance is the delivery distance in the waybill, DFright is the waybill price corresponding to the delivery distance, DThreshold is the shortest added price distance, and DPerFreeght is the model parameter of the machine learning model A.

The machine learning model B may be a linear model as follows:

wherein weight is the weight of goods in the waybill, WFright is the waybill price corresponding to the weight of goods, WTHhreshold is the least added weight, and WPerFreight is the model parameter of the machine learning model B.

The machine learning model C may be a linear model as shown below:

wherein, amout is the price of the passenger in the freight note, afright is the price of the freight note corresponding to the price of the passenger, AThreshold is the lowest price of the passenger, and aperfeight is the model parameter of the machine learning model C.

The machine learning model D may be a linear model as shown below:

OtherFreight＝Grid×GWeight+Weather×WWeight+ProductLevel×PLWeight

wherein, OtherFreight is the price of the waybill corresponding to the natural attribute, Grid is the grade corresponding to the geographical region in the waybill (for example, the grade of the geographical region is higher as the shopping mall is more busy), Weather is the grade corresponding to the Weather in the waybill (for example, the grade of the Weather is higher as the Weather is worse), ProductLevel is the grade of the goods in the waybill (for example, the grade of the goods is higher as the value of the goods is higher), GWeight, WWeight and PLWeight are the weights of the geographical region, the Weather and the grade of the goods respectively, and are the model parameters of the machine learning model D.

Corresponding to the embodiment of the waybill creating method, the specification also provides an embodiment of a waybill creating device.

The embodiment of the waybill creating device can be applied to electronic equipment. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. Taking a software implementation as an example, as a logical device, the device is formed by reading, by a processor of the electronic device where the device is located, a corresponding computer program instruction in the nonvolatile memory into the memory for operation. From a hardware aspect, as shown in fig. 4, the hardware structure diagram of the electronic device where the waybill creating apparatus is located in this specification is shown, except for the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 4, the electronic device where the apparatus is located in the embodiment may also include other hardware according to the actual function that the waybill is created, which is not described again.

Referring to fig. 5, fig. 5 is a block diagram of a waybill creation device according to an exemplary embodiment of the present disclosure. The waybill creating device 50 can be applied to the electronic apparatus shown in fig. 4; the waybill creating device 50 may include:

a receiving module 501, configured to receive an waybill creation request corresponding to a target order; wherein the waybill creation request comprises the target order data;

a pricing module 502, configured to extract preset pricing features from order data in the waybill creating request in response to the waybill creating request, and input the pricing features into a first prediction model corresponding to a distance feature, a second prediction model corresponding to a weight feature, a third prediction model corresponding to a waybill price feature, and a fourth prediction model corresponding to a natural attribute feature, respectively, so that the first prediction model predicts a first waybill price corresponding to the distance feature, the second prediction model predicts a second waybill price corresponding to the weight feature, the third prediction model predicts a third waybill price corresponding to the waybill price feature, and the fourth prediction model predicts a fourth waybill price corresponding to the natural attribute feature; the prediction model is a machine learning model which is obtained by training a preset machine learning model corresponding to the target category based on historical waybill data and is trained;

a first creating module 503, configured to create the waybills corresponding to the target orders based on the predicted first waybill price, the second waybill price, the third waybill price, and the fourth waybill price.

In this embodiment, the apparatus 50 further comprises:

an obtaining module 504, configured to obtain historical waybill data of a deal time within a preset time period; wherein the historical waybill data comprises pricing characteristics and pricing detail; wherein the price details include the first waybill price, the second waybill price, the third waybill price, and the fourth waybill price;

a second creating module 505, configured to create a first training sample set corresponding to the distance feature based on the pricing feature and the first waybill price in the historical waybill data; creating a second set of training samples corresponding to the weight features based on the pricing features and the second waybill price in the historical waybill data; creating a third training sample set corresponding to the customer order price feature based on the pricing feature and the third waybill price in the historical waybill data; creating a fourth training sample set corresponding to the natural attribute features based on the pricing features and the fourth waybill prices in the historical waybill data;

a training module 506, configured to train a preset machine learning model corresponding to the distance feature based on the first training sample set, and determine the machine learning model after training as the first prediction model; training a preset machine learning model corresponding to the weight features based on the second training sample set, and determining the machine learning model after training as the second prediction model; training a preset machine learning model corresponding to the passenger order features based on the third training sample set, and determining the trained machine learning model as the third prediction model; and training a preset machine learning model corresponding to the natural attribute characteristics based on the fourth training sample set, and determining the machine learning model after training as the fourth prediction model.

In this embodiment, the prediction model is an offline prediction model.

In this embodiment, the pricing module 502 is specifically configured to:

In this embodiment, the first creating module 503 is specifically configured to:

In this embodiment, the natural attribute features include one or more of the following: a geographic region characteristic; a weather characteristic; a cargo level characteristic.

In this embodiment, the machine learning model is a linear model.

The implementation process of the functions and actions of each module in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.

For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution in the specification. One of ordinary skill in the art can understand and implement it without inventive effort.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.

In a typical configuration, a computer includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used in one or more embodiments of the present description to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of one or more embodiments herein. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

The above description is only for the purpose of illustrating the preferred embodiments of the one or more embodiments of the present disclosure, and is not intended to limit the scope of the one or more embodiments of the present disclosure, and any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the one or more embodiments of the present disclosure should be included in the scope of the one or more embodiments of the present disclosure.

Claims

1. A waybill creation method, the method comprising:

2. The method of claim 1, further comprising:

3. The method of claim 1, the predictive model being an offline predictive model.

4. The method according to claim 3, wherein the step of extracting preset pricing features from order data in the waybill creation request in response to the waybill creation request, and inputting the pricing features into a first prediction model corresponding to a distance feature, a second prediction model corresponding to a weight feature, a third prediction model corresponding to a customer price feature, and a fourth prediction model corresponding to a natural attribute feature comprises:

5. The method of claim 1, the creating the waybill corresponding to the target order based on the predicted first waybill price, the second waybill price, the third waybill price, and the fourth waybill price, comprising:

6. The method of claim 1, the natural attribute features comprising one or more of the following: a geographic region characteristic; a weather characteristic; a cargo level characteristic.

7. The method of claim 1, the machine learning model being a linear model.

8. An waybill creation apparatus, the apparatus comprising:

9. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor implements the method of any one of claims 1 to 7 by executing the executable instructions.

10. A computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the method of any one of claims 1 to 7.