CN109829667B - Logistics node parcel prediction method and device - Google Patents

Abstract

The application discloses a method and a device for predicting the parcel volume of a logistics node. The method for predicting the parcel volume of the logistics node comprises the following steps: providing a plurality of logistics lines, wherein the logistics lines are respectively composed of a plurality of logistics nodes; determining a target logistics node and at least one target logistics line where the target logistics node is located; determining logistics related information of packages corresponding to target logistics nodes, wherein the logistics related information comprises transaction information and logistics information; and inputting the logistics related information into a parcel quantity prediction model, and determining the parcel quantity of the target logistics node on the at least one item of target streamline. The parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics line of the parcel according to the logistics related information of the parcel, such as transaction information and logistics information, so that the prediction time is advanced.

Description

Logistics node parcel prediction method and device

Technical Field

The application relates to the field of logistics, in particular to a method and a device for predicting the parcel volume of a logistics node.

Background

With the increase of the number of packages of electronic commerce and the continuous development of the express industry, the express company does not compete for the market by price combat, but pushes out logistics ageing service products by improving the service capacity of the express company, and makes ageing commitments to merchants. According to different lines, the aging service is divided into the next day promise and the alternate day promise.

But the resources such as operators and transport vehicles with fixed target logistics nodes of the express company are limited, the express industry is greatly influenced by the trade of the electronic commerce, the electronic commerce is greatly promoted, the fluctuation of the amount of the products is large, the shortage of the personnel or vehicle resources is easily caused, the packages reaching the target logistics nodes cannot be transported in time, and the backlog of the products is caused. This has two consequences, one is that the timeliness of the express mail is affected, the consumer experience is affected, and the scheduling of temporary personnel and temporary vehicles can greatly increase the operating cost of the target logistics node.

Under the background, the parcel quantity on each line of the target logistics node can be predicted in advance to assist the center in making a decision in advance, personnel and vehicles are arranged in advance, the timely delivery of the parcel is ensured, meanwhile, the vehicle dispatching cost can be reduced, and the operation efficiency of the vehicle is improved.

The traditional method is mainly based on a historical average value statistical method, and an operation manager of the center approximately estimates the quantity of packages on each line in the evening according to the quantity of packages in yesterday or the last week.

Specifically, the prior art adopts a prediction method based on historical average single quantity, and an operation manager at the center approximately estimates the quantity of packages on each route in the evening according to the quantity of packages in yesterday or the last cycle.

The method based on historical average single-quantity estimation has the following main defects:

1) Based on the average value mode, the variability of the single quantity along with time, seasons and areas is not considered, and the estimated daily single quantity accuracy is not high.

2) The express service is greatly influenced by the transaction of the electronic commerce, the electronic commerce is greatly promoted, the fluctuation of the quantity of the express service is large, and the quantity of the express service at the time of the great promotion cannot be estimated by the method based on the average quantity of the express service.

3) Because the single quantity of the line level fluctuates greatly, the average value-based mode can only estimate the whole departure bill quantity of the center, and the departure bill quantity on a specific line cannot be estimated.

4) Because the transfer route uncertainty is large and the information communication between the centers is insufficient, the single quantity of transfer from the center to the port can not be predicted by other centers.

5) And (3) correcting time lag, judging after the waiting express mail reaches the target logistics node based on whether the average single quantity is accurately predicted or not, and correcting the predicted result according to the fact that the package is shipped or received.

Disclosure of Invention

In view of the above problems, an embodiment of the present invention provides a method and an apparatus for predicting a parcel volume of a logistics node, so as to solve the problems in the prior art.

In order to solve the above problems, an embodiment of the present application discloses a method for predicting a parcel volume of a logistics node, including:

Providing a plurality of logistics lines, wherein the logistics lines are respectively composed of a plurality of logistics nodes;

determining a target logistics node and at least one target logistics line where the target logistics node is located;

determining logistics related information of packages corresponding to target logistics nodes, wherein the logistics related information comprises transaction information and logistics information;

and inputting the logistics related information into a parcel quantity prediction model, and determining the parcel quantity of the target logistics node on the at least one item of target streamline.

In order to solve the above-mentioned problem, an embodiment of the present application discloses a device for predicting a parcel volume of a logistics node, including:

the logistics line providing module is used for providing a plurality of logistics lines, and the logistics lines are respectively composed of a plurality of logistics nodes;

the target node determining module is used for determining a target logistics node and at least one target logistics line where the target logistics node is located;

the first associated information acquisition module is used for determining logistics associated information of the package corresponding to the target logistics node, wherein the logistics associated information comprises transaction information and logistics information;

and the prediction module is used for inputting the logistics related information into a parcel quantity prediction model and determining the parcel quantity of the target logistics node on the at least one item of target streamline.

In order to solve the above-mentioned problem, an embodiment of the present application further discloses an electronic device, which includes:

a memory for storing a computer readable program;

a processor, when the processor reads the computer readable program in the memory, the electronic device performs the operations of:

providing a plurality of logistics lines, wherein the logistics lines are respectively composed of a plurality of logistics nodes;

determining a target logistics node and at least one target logistics line where the target logistics node is located;

determining logistics related information of packages corresponding to target logistics nodes, wherein the logistics related information comprises transaction information and logistics information;

and inputting the logistics related information into a parcel quantity prediction model, and determining the parcel quantity of the target logistics node on the at least one item of target streamline.

One embodiment of the present application also discloses one or more machine-readable media having instructions stored thereon, which when executed by one or more processors, cause an electronic device to perform the above-described method.

From the foregoing, embodiments of the present application include the following advantages:

the parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics line of the parcel according to the logistics related information of the parcel, such as transaction information and logistics information, so that the prediction time is advanced. In addition, the method provided by the invention predicts the parcel volume according to the plurality of logistics lines respectively, refines the prediction dimension to the line level, and improves the accuracy of the prediction result.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of a method for predicting parcel volume of a logistics node according to a first embodiment of the present application.

Fig. 2 is a flowchart of a method for predicting parcel volume of a logistics node according to a first embodiment of the present application.

Fig. 3A to 3D are flowcharts of a method for predicting a parcel volume of a logistics node according to a second embodiment of the present application.

Fig. 4 is a block diagram of a parcel predicting apparatus for a logistics node according to a third embodiment of the present application.

Fig. 5 is a block diagram of a parcel predicting apparatus for a logistics node according to a fourth embodiment of the present application.

Fig. 6 is a schematic diagram of a real-time data acquisition process according to an embodiment of the invention.

Fig. 7 schematically shows a block diagram of an electronic device for performing the method according to the invention.

Fig. 8 schematically shows a memory unit for holding or carrying program code for implementing the method according to the invention.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application are within the scope of the protection of the present application.

One of the core ideas of the present application is to provide a method and an apparatus for predicting the parcel volume of a logistics node, as shown in fig. 1, for different target logistics routes of a target logistics node, obtain logistics related information, where the logistics related information includes transaction information and logistics information, and predict the parcel volume passing through the target logistics node on the target logistics route according to the logistics related information.

First embodiment

The first embodiment of the invention provides a method for predicting the parcel volume of a logistics node. Fig. 2 is a flow chart showing steps of a method for predicting parcel volume at a logistics node according to a first embodiment of the present invention. As shown in fig. 2, the method for predicting the parcel volume of the logistics node according to the embodiment of the invention includes the following steps:

s101, providing a plurality of logistics lines, wherein the logistics lines are respectively composed of a plurality of logistics nodes;

In this step, a plurality of logistics routes stored in the history may be obtained, and each logistics route may include a plurality of logistics nodes, for example.

For example, a database may store a plurality of logistics lines, such as Beijing-Hangzhou-Guangzhou, beijing-Zhengzhou-Yunnan, and the like. These logistics routes are represented by logistics nodes, which may be transit centers, collection nodes, etc., and the present invention is not particularly limited.

The above-mentioned logistics route may be stored in a database, or may be generated according to the logistics route of the package in real time, which is not limited herein.

S102, determining a target logistics node and at least one item of target logistics line where the target logistics node is located;

in this step, the user designates the target logistics node. The target logistics node may be a logistics node requiring package amount prediction, through which the execution body may determine a target logistics line in which the target logistics node is located from the plurality of logistics lines provided in step S101, and determine at least one target logistics line for package amount prediction.

For example, when the target logistics node to be predicted is a Hangzhou relay center, several logistics routes associated with the Hangzhou relay center, such as Beijing-Hangzhou-Guangzhou, qingdao-Hangzhou-Xiamen, etc., may be selected from the plurality of routes provided in step S101. At least one of the logistics routes may be determined from Hangzhou-related logistics routes as a target logistics route for predicting the parcel volume of the target logistics.

S103, determining logistics related information of the package corresponding to the target logistics node, wherein the logistics related information comprises transaction information and logistics information;

in this step, the execution subject acquires, from the database, logistics related information of a first period of time, for example, a day on which the parcel volume needs to be predicted, including transaction information acquired from the transaction platforms, logistics information acquired from each logistics platform, and the like, and acquires information corresponding to the target logistics node from these pieces of information.

In the first time period, the order or the package which is successfully transacted has both the logistics related information related to the target logistics node and the irrelevant logistics related information. In this step, the logistics associated information related to the target logistics node can be screened out. After the logistics associated information is acquired, each logistics node through which the logistics associated information possibly passes can be predicted according to each piece of logistics associated information.

For example, if a piece of package logistics information is displayed from Beijing to Hangzhou, the passing logistics nodes can be primarily judged to be Beijing transit centers and Hangzhou transit centers. Therefore, the logistics information can be used as logistics related information corresponding to the target logistics node (Hangzhou). For another example, a transaction message indicates that the seller is in Hangzhou and the buyer is in Xiamen, and then the transfer center through which the seller will pass can be primarily determined to be a Hangzhou transfer center and a Xiamen transfer center; this transaction information may be used as logistic related information corresponding to the target logistic node (hangzhou).

In this step, the logistics related information related to the target logistics node in the plurality of logistics related information in the first time period is screened out for the subsequent prediction of the parcel volume.

S104, inputting the logistics related information into a parcel quantity prediction model, and determining the parcel quantity of the target logistics node on the at least one item of target streamline.

In this step, the above-mentioned logistics related information may be input into a parcel volume prediction model, and the parcel volume on each of the target object flow lines related to the target logistics node in the first period (for example, the current day) may be predicted by using the parcel volume prediction model. If the total wrapping amount of the target logistics node on the same day is required to be known, the wrapping amount on each item of target streamline is added to obtain the target logistics node.

The parcel volume prediction model may be a prediction model obtained by training according to logistics associated information corresponding to the target logistics line and parcel volume of the corresponding target logistics node in the historical time. For example, each target logistics line may be trained using a set of sample data associated with the target logistics line to obtain logistics associated information of the target logistics line; and the logistics associated information is combined with the latest logistics associated information and is input into a parcel volume prediction model together, so that the parcel volume of the logistics line on the current day of prediction of the target logistics node can be obtained.

For example, assuming that the Hangzhou transfer center needs to predict the amount of packages on the same day for two label logistics lines related to the Hangzhou transfer center, the Beijing transfer center and the Hangzhou transfer center, the Beijing transfer center inputs logistics related information on each label logistics line into a prediction model, and the amount of packages on each label logistics line of the target logistics node can be predicted through the model.

For example, by training the model in advance, parameter vectors (w 1, w2, …, wk) of the model are already calculated, and the parameter vectors are directly multiplied by feature vectors formed by the logistics associated information, so that the corresponding parcel num=Σwi x xi on each object streamline can be calculated.

From the above, the method for predicting the parcel volume of the logistics node according to the first embodiment of the present invention has at least the following technical effects:

the parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics line of the parcel according to the logistics related information of the parcel, such as transaction information and logistics information. Compared with the method for predicting the daily parcel volume by using historical data in the prior art, the method provided by the invention can predict the time in advance. In addition, the method provided by the invention predicts the parcel volume according to the plurality of logistics lines respectively, refines the prediction dimension to the line level, and improves the accuracy of the prediction result.

Second embodiment

The second embodiment of the invention provides a method for predicting the parcel volume of a logistics node. Fig. 3A is a flowchart illustrating a step of a method for predicting a parcel volume of a logistics node according to a second embodiment of the present invention. As shown in fig. 3A, the method for predicting the parcel volume of the logistics node according to the embodiment of the present invention includes the following steps:

s201, providing a plurality of logistics lines, wherein the logistics lines are respectively composed of a plurality of logistics nodes;

s206, determining a target logistics node and at least one item of target logistics line where the target logistics node is located;

s207, determining logistics related information of the package corresponding to the target logistics node, wherein the logistics related information comprises transaction information and logistics information;

s208, inputting the logistics related information into a parcel quantity prediction model, and determining the parcel quantity of the target logistics node on the at least one item of target streamline.

The steps S201, S206, S207 to S208 are the same as or similar to the steps S101 to S104 of the previous embodiment, and will not be repeated here. This embodiment focuses on the differences from the previous embodiment.

In an alternative embodiment of the present invention, as shown in fig. 3B, step S207, that is, the step of determining the logistics associated information of the package corresponding to the target logistics node, may include, for example:

S2071, acquiring logistics associated information of packages in a first designated period;

in this step, the logistics associated information of the first designated period, for example, the day on which the package amount needs to be predicted, including the transaction information acquired from the transaction platform, the logistics information acquired from the physical distribution platform, and the like, may be obtained. For example, for a package to be received or a package to be sent corresponding to a completed transaction order, acquiring a shipper address and a receiver address in the order; and acquiring the generated logistics information aiming at the sent package.

The current-day logistics related information may be all logistics related information generated in the current day, or may be part of logistics related information, which is not limited herein.

In this embodiment, the server may obtain, from the transaction information of the current day, transaction information corresponding to each order generated in the current day, such as an order number, a receiver address, and a sender address; the package number, the receiving address, the sending address, the current state and the like of each logistics order from package single number generation to collection, circulation and dispatch can be obtained from the logistics information of the current day.

Fig. 6 is a schematic diagram of a real-time data acquisition process according to an embodiment of the invention. As shown in fig. 6, the server may calculate and store data streams generated in real time by the online shopping platform and the logistics platform based on a real-time streaming computing system. That is, the collected logistics information can come from an express service system, and the collected transaction information can come from a real-time service transaction system, such as each online shopping platform.

In this embodiment, since transaction information is collected, the method provided by the embodiment of the invention can advance the predicted time; by combining the logistics information, the parcel volume can be predicted more accurately.

S2072, determining a logistics node corresponding to the logistics associated information;

in this step, the logistics node corresponding to the logistics associated information can be obtained in a predictive manner. For example, the transaction order of the sender in the Hangzhou state at the Beijing receiver can predict the route to be the Beijing transportation center-Zhengzhou transportation center-Hangzhou transportation center according to the logistics related information; the corresponding logistics nodes are Beijing, zhengzhou and Hangzhou.

In addition, while illustrated herein as a diversion center, in a practical scenario, the logistics nodes may also include collection nodes, dispatch nodes, and the like, without limitation.

S2073, acquiring logistics related information related to the target logistics node from the plurality of logistics related information according to the corresponding logistics node;

the logistics associated information comprises transaction information and logistics information.

In this step, according to the logistics node obtained in the substep S2072, information related to the target logistics node in the logistics related information obtained in the first designated period may be screened out. For example, the logistics route of "Beijing Transit center-Zhengzhou Transit center-Hangzhou Transit center" obtained in step S2072 is related to the target logistics node Hangzhou; the logistics line of the Guangzhou transfer center-the Xiamen transfer center is irrelevant to the Hangzhou target logistics line.

In an optional embodiment of the present invention, step S208, that is, inputting the logistics related information into a parcel prediction model, and determining the parcel of the target logistics node on the at least one target logistics line, the method may further include:

s203, acquiring a plurality of logistics related information in the historical data aiming at least one target logistics line;

s204, obtaining the parcel volume of the at least one target logistics line in the operation period of the target logistics node;

s205, training a parcel quantity prediction model by utilizing a plurality of pieces of logistics related information of historical data and the parcel quantity in the operation time of the target logistics node aiming at the at least one target logistics line;

the above-mentioned history data may or may not include the first specified period, for example, the history data within 30 days including the predicted day, or the history data within 30 days including the predicted day.

The above sub-steps S203 to S205 describe the step of training the predictive model.

In sub-step S203, when the stream-related information includes transaction information and logistics information, an electronic face bill number-taking time, a shipping address, a receiving address, and the like may be collected for the transaction information. For logistics information, information and vehicle information of each entity node passing through in the logistics circulation process can be counted, the nodes comprise network points and target logistics nodes, and the information comprises node codes, node names, operation states, operation time, arrival node time and departure node time. For example, when a package is picked up by a website, there is a website code, website name, time to pick up, time to arrive at the website, time to leave the website. The target logistics node code, the target logistics node name, the target logistics node arrival time, the target logistics node departure time, the target logistics node boarding time, the target logistics node departure time and the like are collected when the package arrives at the target logistics node. For example, the collected information may be represented in table 1 as follows:

TABLE 1

In sub-step S204, for each target route, the amount of packages for a day corresponding to the target route may be obtained, and in sub-step S205, the model may be trained using the information on the physical distribution and the amount of packages in the history data.

For example, the logistics related information in the history data for the target logistics route, such as logistics related information of the past 30 days, is obtained in step S203, and the parcel volume of these days can be obtained in step S204. Dividing the logistics-related information into a plurality of time periods, such as each of 3/9/2017 to 4/9/2017, according to the acquired time information; the parcel volume obtained in step S204 is divided into parcel volumes of each day, a prediction model is input, and the logistics associated information of the past days and parcel volumes of a certain day are used as a sample to be input into the prediction model, and the model is trained in step S205.

The training model is the parameter vector (w 1, w2, …, wk) of the model determined according to the characteristic vector in the logistics associated information and the known parcel volume acquired from the historical data. As described above, the parameter vector is directly multiplied by the feature vector in the logistics related information, so that the corresponding parcel num=Σwi×xi on each object flow line can be calculated. The feature vector is the information of each aspect obtained from the physical stream related information, and can be set by a developer of the prediction model, for example, the feature vector comprises electronic face bill number taking time, delivery address and receiving address; through the information of each entity node and the vehicle information, the node comprises a network point and a target logistics node, and the information comprises a node code, a node name, an operation state, an operation time, an arrival node time, a departure node time and the like. The present invention is not particularly limited.

In sub-step S205, for example, the parcel volume of each target line in the second designated period may be acquired according to the target line.

In an alternative embodiment of the present invention, as shown in fig. 3C, the step S205, that is, training, for the at least one target logistics line, a parcel prediction model using a plurality of logistics related information of historical data and a parcel in an operation period of the target logistics node, may include, for example:

s2051, providing a plurality of logistics features and weights to be determined corresponding to each logistics feature;

s2052, inputting the historical multiple logistics associated information and the parcel volume in the operation period of the target logistics node into a regression model to obtain the weight of each logistics feature;

s2053, inputting the parcel quantity before the first appointed period into a time sequence model to obtain estimated daily single quantity;

s2054, inputting the weight output by the regression model and the daily single quantity output by the time sequence model into a fusion model, and outputting the weight of each logistics characteristic after correction.

Substep S2051 through substep S2054 describe a method of determining the weight corresponding to each feature. In the substep S2051, the weight to be determined corresponding to each logistic feature may be set as x, and in the substep S2052, the logistic related information and the parcel volume are input into the regression model, and since the logistic related information already includes a plurality of logistic features, the weight corresponding to the logistic features may be obtained in this step according to the input logistic related information and parcel volume. In step S2053, another time series model may be used, and the periodicity of the historical parcel volume is analyzed by using the time series model, and the current parcel volume is predicted; in step S2054, the two are fused to correct the weight of the logistic feature.

The regression model, the time sequence model and the fusion model are respectively as follows:

regression models, aimed at scoring each input feature based on the input features. The regression model mainly considers a plurality of factors related to daily single quantity, and the relation between the factors and the daily single quantity is learned by constructing the regression model, so that factor information is better utilized, and more accurate single quantity estimation is obtained. Regression models used by the present module include, but are not limited to, linear regression, support vector regression, lifting regression trees, and the like.

Regression model training may result in a set of parameter vectors (w 1, w2, …, wk) for online prediction. Machine learning models used in the present invention include, but are not limited to, linear regression, support vector regression, and lifting regression trees.

The time sequence model is mainly based on a time sequence method, trends and periodicity of daily single-quantity changes are mined, and more accurate single-quantity estimation is provided.

First, calculate daily monologs (n 1, n2, …, n 30) for the line for a period of time (e.g., 30 days) based on the parcel data; then inputting a single quantity of the period of time into a time series model, including but not limited to, arma, arch, etc.; and finally outputting the model to obtain the estimated daily single quantity of the line.

For example, to predict daily inventory from Guangzhou target logistics nodes to Shanghai target logistics nodes, first, calculate the daily inventory of the line for about 30 days, such as (13013,14563, …, 12043), input the daily inventory into a time series model, and output the estimated daily inventory 13850 after training.

The model is fused, and the aim is to fuse the prediction results of the classification model and the regression model to obtain the prediction value of each line. Fusion model training may result in a set of parameter vectors (w 1, w2, …, wk) for online prediction. Fusion models used in the present invention include, but are not limited to, voting, weighted summation, linear regression.

In an alternative embodiment of the present invention, as shown in fig. 3D, step S201, that is, the step of providing a plurality of logistics lines may, for example, include:

s2011, predicting a first rotation center of the package;

s2012, predicting the final rotation center of the package;

s2013, acquiring the logistics line of the package by using the first rotation center and the last rotation center.

In sub-step S2011, the sender address may be partitioned according to the structured semantics; in this step, the sender address and the recipient address may be structured, for example, by extracting, from the structured labels, structured addresses of different administrative levels, such as province, city, district, etc. among the addresses. Then, the first rotation center prediction in the sub-step S2011 and the last rotation center prediction in the sub-step S2012 are performed, so as to obtain the target logistics line in the step S2013.

For the first transfer center, the first transfer center with the largest occurrence number can be used as the first transfer center of the package according to the occurrence number of the first transfer center of the shipping site structured address-the receiving site structured address in the past period of time.

And aiming at the final turning center, taking the final turning center with the largest occurrence number as the final turning center of the package according to the receiving address counted in the past period of time and the occurrence number of the final turning center.

In an alternative embodiment of the present invention, the step of predicting the first rotation center of the package in the substep S2011 may, for example, include:

s20111, predicting the first transfer center of the package according to the corresponding relation between the delivery address and the first transfer center in the transaction information of the logistics related information.

In an alternative embodiment of the present invention, the step S2011, i.e. the step of predicting the first rotation center of the package, may further comprise, for example:

s20112, correcting the first transfer center by utilizing the collecting network point information in the logistics information.

In an embodiment of the invention, predicting the first rotation center may include two substeps of predicting and correcting. In the substep S20111, the first-turn center with the largest occurrence number may be used as the first-turn center of the package according to the counted number of occurrences of the first-turn center from the shipping-place structured address to the receiving-place structured address counted in the past period of time. In substep S20122, the predicted first-turn center may be modified according to the collecting website information. Because the collecting network point can be used for predicting the first transfer center of the package better than the sending address, the first transfer center can be predicted more accurately by adding the step of collecting network point prediction after predicting according to the sending address.

For example, the number of occurrences of the collecting network point-first transfer center can be counted according to the orders in a period of time, and the first transfer center with the largest number of occurrences is taken out for each collecting network point as the first transfer center to be mapped in the administrative area.

For example: according to actual historical order statistics in a period of time, the number of times of occurrence of the Hangzhou Shore middle-Hangzhou target logistics node is the largest, and the first rotation center of the order package collected in the Hangzhou Shore middle is Beijing. As shown in table 2.

Collecting net point	First rotation center	Number of occurrences
			Middle of the Shangzhou Xiaoshan	Hangzhou target logistics node	38939
Middle of the Shangzhou Xiaoshan	Jiaxing target logistics node	1393
			Middle of the Shangzhou Xiaoshan	Shanghai target logistics node	500

TABLE 2

And correcting the first rotation center which is previously predicted according to the sender address by using the predicted first rotation center 'Beijing', and acquiring the corrected first rotation center.

In an optional embodiment of the present invention, S20111, that is, the step of predicting the first-turn center of the package according to the correspondence between the shipping address and the first-turn center in the transaction information of the logistics related information may, for example, include:

s20111a, word segmentation is carried out on the shipping address according to structural semantics;

s20111b, labeling the segmented shipping address with a structural semantic tag, and obtaining a multi-level structural address;

S20111c, complementing missing contents in the shipping address according to a preset structural semantic label template;

s20111d, establishing a first transfer center corresponding to each level of structured address by using the complemented shipping address;

and S20111e, taking the first rotation center with the largest frequency corresponding to each stage of structured address as the first rotation center of the predicted package.

The steps S20111a to S20111e describe that the sender address and the receiver address are segmented to obtain a multi-level structured address, after the structured address is obtained, the labeled structured semantic tag is compared with a preset structured semantic tag template in step S20111c, and if a certain level is found to be absent, the missing content is complemented; if a stage is found to be faulty, error correction can be performed.

For example, the original destination address string is "good road 111 in Hangzhou city, zhejiang province". The address is structured into provinces by text string segmentation based on address information: "Zhejiang province", city: "Hangzhou city", street: "good road". The 'good road' of Hangzhou city is found in the 'Hangzhou area' through the address database, and the administrative area of the full street is the Hangzhou area.

In an alternative embodiment of the present invention, S2012, the step of predicting the last turn center of the package may, for example, include:

and predicting the final turning center of the package by utilizing the corresponding relation between the receiving address of the package and the final turning center.

In this step, regarding the last turn center, the last turn center with the largest occurrence number is taken as the last turn center of the package according to the number of times of occurrence of the receiving address-last turn center counted in the past period of time.

In an optional embodiment of the present invention, step S201, that is, providing a plurality of logistics lines, where the plurality of logistics lines are respectively composed of a plurality of logistics nodes, the method may further include:

s202, dividing the plurality of logistics related information into collection logistics related information and transfer logistics related information according to logistics node information in the logistics related information;

and aiming at the receiving logistics related information and the transit logistics related information, respectively executing the following operations:

acquiring the parcel quantity of the at least one target logistics line in the operation period of the target logistics node;

training a parcel quantity prediction model by utilizing a plurality of pieces of logistics related information of historical data and the parcel quantity within the operation time of the target logistics node aiming at the at least one target logistics line;

Wherein the second designated time period comprises an operational period of the target logistics node.

In step S202, predictions may be made for modeling of the collected stream related information and the transit stream related information, respectively. The package collection refers to the package quantity sent to the next center after the subordinate website of the center collects the package from the merchant; the transfer package refers to the package quantity sent by other centers to the center and sent to the next center through the center. The method can learn through the data, and divide the package into the collection port and the transfer port, so that the package quantity of the center can be predicted more accurately, and the method can be used for modeling and training models respectively aiming at the collection logistics related information and the transfer logistics related information.

In an optional embodiment of the present invention, after step S208, that is, the step of determining the amount of the package of the target logistics node on the target logistics line, the method may further include:

s209, obtaining the total wrapping amount of the target logistics node in the first period according to the wrapping amount of each item of the target logistics node on the object flow line.

In this step, the parcel volume for each target flow route for the target flow node may be summed to obtain a total parcel volume.

In summary, the method for predicting the parcel volume of the logistics node provided by the embodiment has at least the following advantages:

The parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics line of the parcel according to the logistics related information of the parcel, such as transaction information and logistics information, so that the prediction time is advanced. In addition, the method provided by the invention predicts the parcel volume according to the plurality of logistics lines respectively, refines the prediction dimension to the line level, and improves the accuracy of the prediction result.

In addition, the method for predicting the parcel volume of the logistics node provided by the embodiment at least further comprises the following advantages:

the parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics lines of the parcel according to the circulation information of the parcel, respectively acquire the circulation characteristics of the parcel corresponding to the target logistics lines by utilizing the circulation information, predict the parcel volume of the target logistics node by the circulation characteristics of the parcel, and predict the parcel volume of the target logistics node in advance from the moment of order establishment by acquiring logistics related information comprising transaction information and logistics information, so that the prediction time is advanced; and simultaneously, the prediction result can be continuously corrected according to the real-time logistics information, so that the accuracy of the prediction result is improved. Through the design of the prediction model, the invention counts the shipping characteristics, the receiving characteristics and the net point emission characteristics which are strongly related to the prediction result, so that the prediction is more accurate.

Third embodiment

A third embodiment of the present invention provides a device for predicting the parcel volume of a logistics node, as shown in FIG. 4, the device includes:

a logistics line providing module 301, configured to provide a plurality of logistics lines, where the plurality of logistics lines are respectively composed of a plurality of logistics nodes;

a target node determining module 302, configured to determine a target logistics node and at least one target logistics line where the target logistics node is located;

the first associated information obtaining module 303 is configured to determine logistics associated information of a package corresponding to a target logistics node, where the logistics associated information includes transaction information and logistics information;

theprediction module 304 is configured to input the logistics related information into a parcel prediction model, and determine a parcel of the target logistics node on the at least one target streamline.

In summary, the device for predicting the parcel volume of the logistics node provided in the embodiment has at least the following advantages:

the parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics line of the parcel according to the logistics related information of the parcel, such as transaction information and logistics information, so that the prediction time is advanced. In addition, the method provided by the invention predicts the parcel volume according to the plurality of logistics lines respectively, refines the prediction dimension to the line level, and improves the accuracy of the prediction result.

Fourth embodiment

A fourth embodiment of the present invention provides a device for predicting a parcel volume of a logistics node, as shown in FIG. 5, the device includes:

a logistics line providing module 401, configured to provide a plurality of logistics lines, where the plurality of logistics lines are respectively composed of a plurality of logistics nodes;

a target node determining module 402, configured to determine a target logistics node and at least one target logistics line where the target logistics node is located;

a first associated information obtaining module 403, configured to determine logistics associated information of a package corresponding to a target logistics node, where the logistics associated information includes transaction information and logistics information;

and the prediction module 404 is configured to input the logistics related information into a parcel prediction model, and determine the parcel of the target logistics node on the at least one target streamline.

In an alternative embodiment of the present invention, the first association information obtaining module 403 includes:

the associated information acquisition sub-module is used for acquiring logistics associated information of the package in a first appointed period;

the logistics node determining submodule is used for determining logistics nodes corresponding to the logistics associated information;

the screening sub-module is used for acquiring logistics related information related to the target logistics node from the plurality of logistics related information according to the corresponding logistics node;

The logistics associated information comprises transaction information and logistics information.

In an alternative embodiment of the present invention, the apparatus further comprises:

a second associated information obtaining module 405, configured to obtain, for at least one target logistics line, a plurality of logistics associated information in the historical data;

a parcel obtaining module 406, configured to obtain a parcel of the at least one target logistics line in an operation period of the target logistics node;

the model training module 407 is configured to train, for the at least one target logistics line, a parcel prediction model by using the plurality of logistics related information of the historical data and the parcel in the operation time of the target logistics node;

wherein the second designated time period comprises an operational period of the target logistics node.

In an alternative embodiment of the present invention, the model training module 407 includes:

the sub-module is used for providing a plurality of logistics features and weights to be determined corresponding to the logistics features;

the weight acquisition sub-module is used for inputting the plurality of logistics associated information in the second appointed time period and the parcel volume in the operation time period of the target logistics node into a regression model to acquire the weight aiming at each logistics characteristic;

The estimated daily single quantity determination submodule is used for inputting the parcel quantity before the first appointed period into a time sequence model and determining the estimated daily single quantity;

the weight determination submodule is used for inputting the weight output by the regression model and the daily single quantity output by the time sequence model into the fusion model and outputting the weight of each logistics characteristic after correction.

In an alternative embodiment of the present invention, the logistics circuit providing module 401 includes:

the first rotation center prediction sub-module is used for predicting the first rotation center of the package;

the final rotation center prediction sub-module is used for predicting the final rotation center of the package;

and the logistics line determining sub-module is used for acquiring the logistics line of the package by utilizing the first rotation center and the last rotation center.

In an alternative embodiment of the present invention, the first rotation center prediction submodule includes:

and the prediction unit is used for predicting the first transfer center of the package according to the corresponding relation between the delivery address and the first transfer center in the transaction information of the logistics associated information.

In an optional embodiment of the invention, the first rotation center prediction submodule further includes:

and the correction unit is used for correcting the first transfer center by utilizing the collecting network point information in the logistics information.

In an alternative embodiment of the present invention, the prediction unit includes:

the word segmentation subunit is used for segmenting the delivery address according to the structural semantics;

the structured address acquisition subunit is used for labeling the segmented shipping address with a structured semantic tag to acquire a multi-level structured address;

a complementing subunit, configured to complement missing contents in the shipping address according to a preset structural semantic label template;

the mapping establishing subunit is used for establishing a first transfer center corresponding to each level of structured address by utilizing the complemented shipping address;

and the prediction subunit is used for taking the first-turn center with the largest frequency corresponding to each stage of structured address as the first-turn center of the predicted package.

In an alternative embodiment of the present invention, the last-rotation central prediction submodule is configured to:

and predicting the final turning center of the package by utilizing the corresponding relation between the receiving address of the package and the final turning center.

In an alternative embodiment of the present invention, the apparatus further comprises:

the dividing module 408 divides the plurality of logistics related information into collection logistics related information and transfer logistics related information according to the logistics node information in the logistics related information.

In an alternative embodiment of the present invention, the apparatus further comprises:

and the accumulation module 409 is configured to obtain, according to the parcel on each item of the target logistics node, the total parcel of the target logistics node in the first period.

In summary, the device for predicting the parcel volume of the logistics node provided in the embodiment has at least the following advantages:

the parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics line of the parcel according to the logistics related information of the parcel, such as transaction information and logistics information, so that the prediction time is advanced. In addition, the method provided by the invention predicts the parcel volume according to the plurality of logistics lines respectively, refines the prediction dimension to the line level, and improves the accuracy of the prediction result.

In addition, the predicting device for the parcel volume of the logistics node provided by the embodiment at least further comprises the following advantages:

the parcel volume prediction method and the parcel volume prediction device for the logistics node provided by the embodiment of the invention can predict the target logistics lines of the parcel according to the circulation information of the parcel, respectively acquire the circulation characteristics of the parcel corresponding to the target logistics lines by utilizing the circulation information, predict the parcel volume of the target logistics node by the circulation characteristics of the parcel, and predict the parcel volume of the target logistics node in advance from the moment of order establishment by acquiring logistics related information comprising transaction information and logistics information, so that the prediction time is advanced; and simultaneously, the prediction result can be continuously corrected according to the real-time logistics information, so that the accuracy of the prediction result is improved. Through the design of the prediction model, the invention counts the shipping characteristics, the receiving characteristics and the net point emission characteristics which are strongly related to the prediction result, so that the prediction is more accurate.

For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.

Fig. 7 is a schematic hardware structure of an electronic device according to an embodiment of the disclosure. As shown in fig. 7, the electronic apparatus may include aninput device 90, a processor 91, anoutput device 92, amemory 93, and at least onecommunication bus 94. Thecommunication bus 94 is used to enable communication connections between the elements. Thememory 93 may comprise a high-speed RAM memory or may further comprise a non-volatile memory NVM, such as at least one magnetic disk memory, in which various programs may be stored for performing various processing functions and implementing the method steps of the present embodiment.

Alternatively, the processor 91 may be implemented as, for example, a central processing unit (Central Processing Unit, abbreviated as CPU), an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic components, and the processor 91 is coupled to theinput device 90 and theoutput device 92 through wired or wireless connection.

Alternatively, theinput device 90 may include a variety of input devices, for example, may include at least one of a user-oriented user interface, a device-oriented device interface, a programmable interface to software, a camera, and a sensor. Optionally, the device interface facing the device may be a wired interface for data transmission between devices, or may be a hardware insertion interface (such as a USB interface, a serial port, etc.) for data transmission between devices; alternatively, the user-oriented user interface may be, for example, a user-oriented control key, a voice input device for receiving voice input, and a touch-sensitive device (e.g., a touch screen, a touch pad, etc. having touch-sensitive functionality) for receiving user touch input by a user; optionally, the programmable interface of the software may be, for example, an entry for a user to edit or modify a program, for example, an input pin interface or an input interface of a chip, etc.; optionally, the transceiver may be a radio frequency transceiver chip, a baseband processing chip, a transceiver antenna, etc. with a communication function. An audio input device such as a microphone may receive voice data. Theoutput device 92 may include a display, audio, etc.

In this embodiment, the processor of the electronic device may include functions for executing each module of the data processing device in each apparatus, and specific functions and technical effects may be referred to the above embodiments and are not described herein again.

Fig. 8 is a schematic hardware structure of an electronic device according to another embodiment of the present application. Fig. 8 is a diagram of one particular embodiment of the implementation of fig. 7. As shown in fig. 8, the electronic device of the present embodiment includes aprocessor 101 and amemory 102.

Theprocessor 101 executes the computer program code stored in thememory 102 to implement the method for predicting the parcel volume of the logistics node in the embodiment shown in fig. 2, 3A to 3D.

Thememory 102 is configured to store various types of data to support operations at the electronic device. Examples of such data include instructions, such as messages, pictures, videos, etc., for any application or method operating on the electronic device. Thememory 102 may include a random access memory (random access memory, simply referred to as RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

Optionally, aprocessor 101 is provided in theprocessing assembly 100. The electronic device may further include: acommunication component 103, apower supply component 104, a multimedia component 105, anaudio component 106, an input/output interface 107 and/or asensor component 108. The components and the like included in the electronic device are set according to actual requirements, and the embodiment is not limited thereto.

Theprocessing assembly 100 generally controls the overall operation of the electronic device. Theprocessing assembly 100 may include one ormore processors 101 to execute instructions to perform all or part of the steps of the methods of fig. 2, 3A-3D described above. Further, theprocessing component 100 may include one or more modules that facilitate interactions between theprocessing component 100 and other components. For example, theprocessing component 100 may include a multimedia module to facilitate interaction between the multimedia component 105 and theprocessing component 100.

Thepower supply assembly 104 provides power to the various components of the electronic device. Thepower components 104 can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for electronic devices.

The multimedia component 105 includes a display screen between the electronic device and the user that provides an output interface. In some embodiments, the display screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the display screen includes a touch panel, the display screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or slide action, but also the duration and pressure associated with the touch or slide operation.

Theaudio component 106 is configured to output and/or input audio signals. For example, theaudio component 106 includes a Microphone (MIC) configured to receive external audio signals when the electronic device is in an operational mode, such as a speech recognition mode. The received audio signals may be further stored in thememory 102 or transmitted via thecommunication component 103. In some embodiments, theaudio component 106 further comprises a speaker for outputting audio signals.

The input/output interface 107 provides an interface between theprocessing assembly 100 and peripheral interface modules, which may be click wheels, buttons, etc. These buttons may include, but are not limited to: volume button, start button and lock button.

Thesensor assembly 108 includes one or more sensors for providing status assessment of various aspects of the electronic device. For example, thesensor assembly 108 may detect an on/off state of the electronic device, a relative positioning of the assembly, and the presence or absence of user contact with the electronic device. Thesensor assembly 108 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact, including detecting the distance between the user and the electronic device. In some embodiments, thesensor assembly 108 may also include a camera or the like.

Thecommunication component 103 is configured to facilitate communication between the electronic apparatus and other devices in a wired or wireless manner. The electronic device may access a wireless network based on a communication standard, such as WiFi,2G or 3G, or a combination thereof. In one embodiment, the electronic device may include a SIM card slot, where the SIM card slot is used to insert a SIM card, so that the electronic device may log onto a GPRS network, and establish communication with a server through the internet.

From the above, thecommunication component 103, theaudio component 106, the input/output interface 107, and thesensor component 108 in the embodiment of fig. 8 can be implemented as the input device in the embodiment of fig. 7.

The embodiment of the application provides an electronic device, which comprises: one or more processors; and one or more machine-readable media having instructions stored thereon, which when executed by the one or more processors, cause the electronic device to perform a method of generating a video summary as described in one or more of the embodiments of the application.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.

While preferred embodiments of the present embodiments have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the present application.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or electronic device 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 electronic device. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or electronic device that comprises the element.

The above description is provided in detail of a method and a device for predicting the parcel volume of a logistics node, and specific examples are applied to illustrate the principle and the implementation of the present application, and the above description of the examples is only used for helping to understand the method and the core idea of the present application; meanwhile, as those skilled in the art will have modifications in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims (20)

1. The method for predicting the parcel volume of the logistics node is characterized by comprising the following steps:

providing a plurality of logistics lines, wherein the logistics lines are respectively composed of a plurality of logistics nodes;

determining a target logistics node and at least one target logistics line where the target logistics node is located;

determining logistics related information of packages corresponding to target logistics nodes, wherein the logistics related information comprises transaction information and logistics information;

acquiring a plurality of logistics related information in historical data aiming at least one item of target logistics line;

acquiring the parcel quantity of the at least one target logistics line in the operation period of the target logistics node;

Providing a plurality of logistics features and weights to be determined corresponding to each logistics feature;

inputting a plurality of logistics associated information of historical data and the parcel volume in the operation period of the target logistics node into a regression model to obtain the weight aiming at each logistics characteristic;

inputting the parcel quantity before the first appointed period into a time sequence model to obtain estimated daily single quantity;

inputting the weight output by the regression model and the daily single quantity output by the time sequence model into a fusion model, and outputting the weight of each logistics characteristic after correction;

and inputting the logistics related information into a parcel quantity prediction model, and determining the parcel quantity of the target logistics node on the at least one item of target streamline.

2. The method of claim 1, wherein the step of determining the logistics associated information for the package corresponding to the target logistics node comprises:

acquiring logistics associated information of packages in a first designated period;

determining a logistics node corresponding to the logistics associated information;

acquiring logistics related information related to the target logistics node from the plurality of logistics related information according to the corresponding logistics node; the logistics associated information comprises transaction information and logistics information.

3. The method of claim 1, wherein the step of providing a plurality of logistics lines comprises:

predicting a first rotation center of the package;

predicting the final center of rotation of the package;

and acquiring the logistics line of the package by using the first rotation center and the last rotation center.

4. A method according to claim 3, wherein the step of predicting a first turn center of the package comprises:

and predicting the first transfer center of the package according to the corresponding relation between the delivery address and the first transfer center in the transaction information of the logistics associated information.

5. The method of claim 4, wherein predicting a first turn center of the package further comprises:

and correcting the first transfer center by utilizing the collecting network point information in the logistics information.

6. The method of claim 4, wherein predicting the first turn center of the package based on the correspondence of the shipping address and the first turn center in the transaction information of the logistics associated information comprises:

the delivery address is segmented according to the structural semantics;

labeling the classified delivery address with a structural semantic label to obtain a multi-level structural address;

According to a preset structural semantic label template, the missing content in the shipping address is complemented;

establishing a first transfer center corresponding to each level of structured address by utilizing the complemented shipping address;

and taking the first rotation center with the largest frequency corresponding to each stage of structured address as the first rotation center of the predicted package.

7. A method according to claim 3, wherein the step of predicting the last turn center of the package comprises:

and predicting the final turning center of the package by utilizing the corresponding relation between the receiving address of the package and the final turning center.

8. The method of claim 1, wherein the step of inputting the logistics associated information into a parcel prediction model to determine the parcel of the target logistics node on the at least one target streamline, the method further comprising:

dividing the plurality of logistics related information into collection logistics related information and transfer logistics related information according to logistics node information in the logistics related information;

and aiming at the receiving logistics related information and the transit logistics related information, respectively executing the following operations:

acquiring the parcel quantity of the at least one target logistics line in the operation period of the target logistics node;

And training a parcel quantity prediction model by utilizing a plurality of pieces of logistics related information of historical data and the parcel quantity within the operation time of the target logistics node aiming at the at least one target logistics line.

9. The method of claim 1, wherein after the step of determining the amount of the target logistics node's parcel on the target logistics line, the method further comprises:

and acquiring the total wrapping quantity of the target logistics node in the first period according to the wrapping quantity of each item of the target logistics node on the object flow line.

10. The utility model provides a predicting device of commodity circulation node parcel volume which characterized in that includes:

the logistics line providing module is used for providing a plurality of logistics lines, and the logistics lines are respectively composed of a plurality of logistics nodes;

the target node determining module is used for determining a target logistics node and at least one target logistics line where the target logistics node is located;

the first associated information acquisition module is used for determining logistics associated information of the package corresponding to the target logistics node, wherein the logistics associated information comprises transaction information and logistics information;

the second associated information acquisition module is used for acquiring a plurality of logistics associated information in the historical data aiming at least one target logistics line;

The parcel volume acquisition module is used for acquiring parcel volume of the at least one target logistics line in the operation period of the target logistics node;

the model training module is used for providing a plurality of logistics features and weights to be determined corresponding to each logistics feature; inputting a plurality of logistics associated information of historical data and the parcel volume in the operation period of the target logistics node into a regression model to obtain the weight aiming at each logistics characteristic; inputting the parcel quantity before the first appointed period into a time sequence model, and determining a predicted daily single quantity; inputting the weight output by the regression model and the daily single quantity output by the time sequence model into a fusion model, and outputting the weight of each logistics characteristic after correction;

and the prediction module is used for inputting the logistics related information into a parcel quantity prediction model and determining the parcel quantity of the target logistics node on the at least one item of target streamline.

11. The apparatus of claim 10, wherein the first association information acquisition module comprises:

the associated information acquisition sub-module is used for acquiring logistics associated information of the package in a first appointed period;

the logistics node determining submodule is used for determining logistics nodes corresponding to the logistics associated information;

The screening sub-module is used for acquiring logistics related information related to the target logistics node from the plurality of logistics related information according to the corresponding logistics node;

the logistics associated information comprises transaction information and logistics information.

12. The apparatus of claim 10, wherein the logistics circuit providing module comprises:

the first rotation center prediction sub-module is used for predicting the first rotation center of the package;

the final rotation center prediction sub-module is used for predicting the final rotation center of the package;

and the logistics line determining sub-module is used for acquiring the logistics line of the package by utilizing the first rotation center and the last rotation center.

13. The apparatus of claim 12, wherein the first rotation center prediction submodule comprises:

and the prediction unit is used for predicting the first transfer center of the package according to the corresponding relation between the delivery address and the first transfer center in the transaction information of the logistics associated information.

14. The apparatus of claim 13, wherein the first rotation center prediction submodule further comprises:

and the correction unit is used for correcting the first transfer center by utilizing the collecting network point information in the logistics information.

15. The apparatus of claim 13, wherein the prediction unit comprises:

the word segmentation subunit is used for segmenting the delivery address according to the structural semantics;

the structured address acquisition subunit is used for labeling the segmented shipping address with a structured semantic tag to acquire a multi-level structured address;

a complementing subunit, configured to complement missing contents in the shipping address according to a preset structural semantic label template;

the mapping establishing subunit is used for establishing a first transfer center corresponding to each level of structured address by utilizing the complemented shipping address;

and the prediction subunit is used for taking the first-turn center with the largest frequency corresponding to each stage of structured address as the first-turn center of the predicted package.

16. The apparatus of claim 12, wherein the last-rotation-center prediction submodule is configured to:

and predicting the final turning center of the package by utilizing the corresponding relation between the receiving address of the package and the final turning center.

17. The apparatus of claim 10, wherein the apparatus further comprises:

the division module divides the plurality of logistics related information into collection logistics related information and transfer logistics related information according to logistics node information in the logistics related information.

18. The apparatus of claim 10, wherein the apparatus further comprises:

and the accumulation module is used for acquiring the total wrapping quantity of the target logistics node in the first period according to the wrapping quantity of each item of the target logistics node on the flow line.

19. An electronic device, comprising:

one or more processors; and

one or more machine readable media having instructions stored thereon, which when executed by the one or more processors, cause the electronic device to perform the method of one or more of claims 1-9.

20. One or more machine readable media having instructions stored thereon that, when executed by one or more processors, cause an electronic device to perform the method of one or more of claims 1-9.

Images