Information generation method and device

Through the Bayesian network model, the historical data of the logistics distribution station was analyzed causally, and the problems of variable and correlation in complex scenarios were solved, and the accurate identification and optimization of factors affecting distribution efficiency was achieved, which improved the determination of distribution efficiency. sex.

CN119067523APending Publication Date: 2024-12-03BEIJING JINGDONG ZHENSHI INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310652685.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-06-02
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

In complex logistics and distribution scenarios, there are many factors that affect the efficiency of delivery personnel and are interrelated, which makes it difficult to accurately judge the existing causal analysis based on experience and cannot effectively improve the distribution efficiency.

Method used

By obtaining the historical data of the target distribution station, performing feature extraction and screening, using Bayesian network model for causal inference calculation, generating degree of impact level information affecting distribution efficiency, and determining the influencing factors and their importance.

Benefits of technology

Accurate attribution analysis and sorting of factors affecting distribution efficiency is achieved, the accuracy and efficiency of determining factors that affect distribution efficiency is improved, and manual participation is reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067523A_ABST
    Figure CN119067523A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides an information generation method and device. The information generation method comprises the steps of firstly obtaining historical distribution data corresponding to a target distribution station, and then performing feature extraction on the historical distribution data to obtain a plurality of candidate feature data used for influencing the distribution efficiency of the target distribution station and the distribution efficiency of the target distribution station, performing feature screening on the plurality of candidate feature data based on the delivery efficiency of the target delivery station, screening out a plurality of target feature data from the plurality of candidate feature data, and finally performing causal inference calculation on the plurality of target feature data by using a Bayesian network model. The influence degree grade information of the multiple pieces of target feature data for the distribution efficiency of the target distribution station is generated, and the historical distribution data of the target distribution station and the Bayesian network model are utilized to carry out causal inference, so that the factors influencing the distribution efficiency can be accurately and quickly determined. The accuracy and the efficiency of determining the influence factors of the distribution efficiency of the distribution station are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the fields of computer technology and Internet technology, and in particular, to an information generation method and device. Background Art

[0002] With the rapid development of the logistics industry and the increasing demands of consumers, logistics companies are facing more and more problems, such as improving the efficiency of each link and how to improve consumer satisfaction. The logistics links mainly include warehousing, picking, transportation, and distribution. Warehouse refers to the warehouse, which is mainly used to store goods. Picking refers to the sorting center, which divides the goods produced in the warehouse according to different flow directions. Transportation refers to transportation, which is responsible for vehicle arrangement in the whole process. Distribution refers to the distribution station, which is the logistics link closest to the user and is mainly responsible for delivering goods to customers. The efficiency of each link will affect the entire logistics, so improving the efficiency of each link can improve the efficiency of the entire logistics.

[0003] For some simple scenarios, such as when there are fewer influencing factors, attribution based on experience is simple, intuitive, and has a certain effect. For complex scenarios, where there are many influencing factors and the relationships between the factors are complex, judgment based on experience is rather one-sided, often resulting in inaccurate attribution, and thus failing to effectively solve the problem. For distribution stations, there are dozens of factors that affect the efficiency of delivery personnel, and they affect each other, so relying solely on business experience will not yield good results. Summary of the invention

[0004] Embodiments of the present disclosure provide an information generation method, device, electronic device, and computer-readable medium.

[0005] In a first aspect, an embodiment of the present disclosure provides an information generation method, the method comprising: acquiring historical distribution data corresponding to a target distribution station; performing feature extraction on the historical distribution data to obtain a plurality of candidate feature data for influencing the distribution efficiency of the target distribution station and the distribution efficiency of the target distribution station; performing feature screening on the plurality of candidate feature data based on the distribution efficiency of the target distribution station, and screening out a plurality of target feature data from the plurality of candidate feature data; performing causal inference calculations on the plurality of target feature data using a Bayesian network model, and generating information on the degree of influence of the plurality of target feature data on the distribution efficiency of the target distribution station.

[0006] In some embodiments, feature screening is performed on multiple candidate feature data based on the delivery efficiency of the target delivery station, and multiple target feature data are screened out from the multiple candidate feature data, including: based on the multiple candidate feature data and the delivery efficiency value of the target delivery station, regressing and fitting the multiple candidate feature data to obtain fitting data corresponding to the multiple candidate feature data; based on a first threshold and the fitting data corresponding to the multiple candidate feature data, selecting multiple first feature data from the multiple candidate feature data; calculating the importance level of the multiple first feature data to obtain the importance level information of the multiple first feature data; based on a second threshold and the importance level information of the multiple first feature data, screening out multiple target feature data from the multiple first feature data.

[0007] In some embodiments, a Bayesian network model is used to perform causal inference calculations on multiple target feature data to generate information on the degree of influence of multiple target feature data on the delivery efficiency of a target delivery station, including: discretizing multiple target feature data to obtain multiple discretized target feature data; using a Bayesian network model to perform causal inference calculations on multiple discretized target feature data to generate information on the degree of influence of multiple discretized target feature data on the delivery efficiency of a target delivery station.

[0008] In some embodiments, a Bayesian network model is used to perform causal inference calculations on multiple discrete target feature data to generate information on the degree of influence of multiple discrete target feature data on the delivery efficiency of a target delivery station, including: constructing a directed acyclic graph of multiple discrete target feature data based on fitting data and importance level information corresponding to the multiple target feature data; grouping the multiple discrete target feature data according to the directed acyclic graph of the multiple discrete target feature data to obtain multiple groups of target feature data; determining the data types corresponding to the multiple groups of target feature data based on the fitting data and importance level information corresponding to the multiple groups of target feature data; and performing causal inference calculations on the multiple groups of target feature data based on the data types corresponding to the multiple groups of target feature data to generate information on the degree of influence of multiple discrete target feature data on the delivery efficiency of a target delivery station.

[0009] In some embodiments, based on the data types corresponding to the multiple groups of target feature data, causal inference calculations are performed on the multiple groups of target feature data respectively to generate multiple levels of information on the degree of influence of the discrete target feature data on the delivery efficiency of the target delivery station, including: for each group of target feature data, in response to determining that the data type of the target feature data is a first preset type, obtaining the target feature data and the delivery efficiency of the target delivery station; based on the target feature data and the delivery efficiency of the target delivery station, performing causal inference calculations on the target feature data to generate the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0010] In some embodiments, based on the data types corresponding to multiple groups of target feature data, causal inference calculations are performed on the multiple groups of target feature data respectively to generate multiple levels of information on the degree of influence of discrete target feature data on the delivery efficiency of the target delivery station, including: for each group of target feature data, in response to determining that the data type of the target feature data is a second preset type, obtaining the target feature data, other feature data associated with the target feature data, and the delivery efficiency of the target delivery station; based on the target feature data, other feature data associated with the target feature data, and the delivery efficiency of the target delivery station, performing causal inference calculations on the target feature data to generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0011] In some embodiments, the method further includes: generating efficiency prompt information corresponding to the target delivery station based on the impact level information of the delivery efficiency of the target delivery station of the plurality of target characteristic data; and sending the efficiency prompt information to the target delivery station.

[0012] In a second aspect, an embodiment of the present disclosure provides an information generating device, which includes: an acquisition module, configured to acquire historical distribution data corresponding to a target distribution station; a feature extraction module, configured to perform feature extraction on the historical distribution data, and obtain a plurality of candidate feature data for influencing the distribution efficiency of the target distribution station and the distribution efficiency of the target distribution station; a feature screening module, configured to perform feature screening on a plurality of candidate feature data based on the distribution efficiency of the target distribution station, and screen out a plurality of target feature data from the plurality of candidate feature data; a generation module, configured to perform causal inference calculations on a plurality of target feature data using a Bayesian network model, and generate information on the degree of influence of the plurality of target feature data on the distribution efficiency of the target distribution station.

[0013] In some embodiments, the feature screening module is further configured to: perform regression fitting on multiple candidate feature data based on multiple candidate feature data and the delivery efficiency value of the target delivery station to obtain fitting data corresponding to the multiple candidate feature data; select multiple first feature data from the multiple candidate feature data based on a first threshold and the fitting data corresponding to the multiple candidate feature data; calculate the importance level of the multiple first feature data to obtain importance level information of the multiple first feature data; and screen out multiple target feature data from the multiple first feature data based on a second threshold and the importance level information of the multiple first feature data.

[0014] In some embodiments, the generation module includes: a discrete unit, configured to: discretize multiple target feature data to obtain multiple discretized target feature data; a generation unit, configured to: use a Bayesian network model to perform causal inference calculations on multiple discretized target feature data to generate multiple discrete target feature data on the distribution efficiency of the target distribution station. Impact level information.

[0015] In some embodiments, the generating unit is further configured to: construct a directed acyclic graph of multiple discretized target feature data based on the fitting data and importance level information corresponding to the multiple target feature data; group the multiple discretized target feature data according to the directed acyclic graph of the multiple discretized target feature data to obtain multiple groups of target feature data; determine the data types corresponding to the multiple groups of target feature data based on the fitting data and importance level information corresponding to the multiple groups of target feature data; perform causal inference calculations on the multiple groups of target feature data based on the data types corresponding to the multiple groups of target feature data, and generate influence level information of the multiple discretized target feature data on the delivery efficiency of the target delivery station.

[0016] In some embodiments, the generation unit is further configured to: for each set of target feature data, in response to determining that the data type of the target feature data is a first preset type, obtain the target feature data and the delivery efficiency of the target delivery station; based on the target feature data and the delivery efficiency of the target delivery station, perform causal inference calculations on the target feature data, and generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0017] In some embodiments, the generation unit is further configured to: for each set of target feature data, in response to determining that the data type of the target feature data is a second preset type, obtain the target feature data, other feature data associated with the target feature data, and the delivery efficiency of the target delivery station; based on the target feature data, other feature data associated with the target feature data, and the delivery efficiency of the target delivery station, perform causal inference calculations on the target feature data, and generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0018] In some embodiments, the device also includes a sending module; the generating unit is further configured to: generate efficiency prompt information corresponding to the target delivery station based on the impact level information of multiple target feature data on the delivery efficiency of the target delivery station; the sending module is configured to: send the efficiency prompt information to the target delivery station.

[0019] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by one or more processors, the one or more processors implement the information generation method described in any embodiment of the first aspect.

[0020] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the information generating method as described in any embodiment of the first aspect.

[0021] The information generation method provided by the embodiment of the present disclosure is as follows: the execution subject first obtains the historical distribution data corresponding to the target distribution station, then performs feature extraction on the historical distribution data, obtains multiple candidate feature data for affecting the distribution efficiency of the target distribution station and the distribution efficiency of the target distribution station, then performs feature screening on the multiple candidate feature data based on the distribution efficiency of the target distribution station, screens out multiple target feature data from the multiple candidate feature data, and finally uses the Bayesian network model to perform causal inference calculation on the multiple target feature data to generate multiple target feature data on the distribution efficiency of the target distribution station. Causal inference is performed using the historical distribution data of the target distribution station and the Bayesian network model, and attribution analysis is performed on the factors affecting the distribution efficiency, which can accurately and quickly determine the factors affecting the distribution efficiency, and further determine the influence level information of each factor affecting the distribution efficiency, so that users can better understand the important factors in the influencing factors, and realize the causal inference of the factors affecting the performance of the distribution station using the Bayesian network model at the terminal, without the need for human participation, thereby improving the accuracy and efficiency of determining the factors affecting the distribution efficiency of the distribution station. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Other features, objects and advantages of the present disclosure will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0023] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present disclosure may be applied;

[0024] Figure 2 is a flow chart of an embodiment of the information generation method according to the present disclosure;

[0025] Figure 3A is a flow chart of one embodiment of generating impact level information according to the present disclosure;

[0026] Figure 3B is a schematic diagram of a directed acyclic graph of multiple discretized target feature data according to the present disclosure;

[0027] Figure 4 is a flow chart of an embodiment of generating impact level information based on data types corresponding to multiple groups of target feature data according to the present disclosure;

[0028] Figure 5 is a schematic structural diagram of an embodiment of an information generating device according to the present disclosure;

[0029] Figure 6 It is a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. DETAILED DESCRIPTION

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

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

[0032] Figure 1 An exemplary system architecture 100 to which embodiments of the information generating method or information generating apparatus of the present application can be applied is shown.

[0033] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0034] The user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. The terminal devices 101, 102, 103 can be user terminal devices, on which various client applications can be installed, such as e-commerce platform applications, image applications, video applications, search applications, financial applications, etc.

[0035] The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting receiving server messages, including but not limited to smart phones, tablet computers, e-book readers, electronic players, laptop computers, desktop computers, and the like.

[0036] Terminal devices 101, 102, 103 can be hardware or software. When terminal devices 101, 102, 103 are hardware, they can be various electronic devices. When terminal devices 101, 102, 103 are software, they can be installed in the electronic devices listed above. They can be implemented as multiple software or software modules (for example, multiple software modules for providing distributed services), or they can be implemented as a single software or software module. No specific limitation is made here.

[0037] Terminal devices 101, 102, and 103 can obtain historical distribution data corresponding to the target distribution station, and then perform feature extraction on the historical distribution data to obtain multiple candidate feature data and the distribution efficiency of the target distribution station that are used to affect the distribution efficiency of the target distribution station. Thereafter, feature screening is performed on the multiple candidate feature data based on the distribution efficiency of the target distribution station, and multiple target feature data are screened out from the multiple candidate feature data. Finally, a Bayesian network model is used to perform causal inference calculations on the multiple target feature data to generate information on the degree of influence of the multiple target feature data on the distribution efficiency of the target distribution station.

[0038] The server 105 may be a server providing various services, such as a backend server that receives a request sent by a terminal device that establishes a communication connection with the server 105. The backend server may receive and analyze the request sent by the terminal device and generate a processing result.

[0039] It should be noted that the server can be hardware or software. When the server is hardware, it can be various electronic devices that provide various services to the terminal device. When the server is software, it can be implemented as multiple software or software modules that provide various services to the terminal device, or it can be implemented as a single software or software module that provides various services to the terminal device. No specific limitation is made here.

[0040] It should be noted that the information generating method provided in the embodiments of the present disclosure may be executed by the terminal devices 101 , 102 , 103 , and accordingly, the information generating apparatus may be disposed in the terminal devices 101 , 102 , 103 .

[0041] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0042] refer to Figure 2 , shows a flowchart 200 of an embodiment of an information generation method according to the present disclosure. The information generation method comprises the following steps:

[0043] Step 210, obtaining historical delivery data corresponding to the target delivery station.

[0044] In this step, the execution subject (e.g. Figure 1 The terminal devices 101, 102, and 103 in the diagram can determine the factors that affect the delivery efficiency for the target delivery station, and the influencing factors include the delivery station information, the delivery personnel information, the delivery equipment information, and the upstream of the delivery station. The delivery station information can refer to the characteristics of the delivery station itself, such as the coverage area, the number of road areas, the type of road areas, the city to which it belongs, etc.; the delivery personnel information can refer to the characteristics of the delivery personnel at the delivery station, such as work proficiency, gender, age, etc.; the delivery equipment information can refer to the characteristics of the equipment used by the delivery station and the delivery personnel, including whether there are conveyor belts, delivery vehicles, etc.; the upstream of the delivery station can refer to some characteristics of the handover between the picking and the delivery station, such as the arrival time of the vehicle, etc.

[0045] The above-mentioned execution entity can obtain the corresponding historical delivery data from the above-mentioned influencing factors from the target delivery station. The historical delivery data can represent the basic data that can affect the delivery efficiency of the target delivery station, and may include historical delivery waybill information in the target delivery station, delivery station information, road area information associated with the delivery station, delivery personnel information in the delivery station, delivery equipment information, etc.

[0046] Among them, the above-mentioned historical delivery waybill information may include the basic information of the historical waybill, such as the date, waybill operation type, operation completion time, the road area where the waybill is located, etc. This application does not make specific limitations on this, and may include basic information related to the waybill that is well known in the art.

[0047] The above-mentioned distribution station information may include basic information of the distribution station, such as the name of the province where the distribution station is located, the distribution area, the number of distribution road areas, the available capacity of the distribution station, etc. This application does not make specific limitations on this, and may include basic information related to the distribution station that is well known in the art.

[0048] The road area information associated with the above-mentioned distribution station may include relevant information of the road area, such as the number of road areas, the type of road areas, etc., which is not specifically limited in the present application and may include basic information related to the road area information associated with the distribution station that is well known in the art.

[0049] The delivery personnel information in the above-mentioned delivery station may include basic information of the delivery personnel, for example, may include basic situation information related to the delivery personnel.

[0050] The above-mentioned delivery equipment information may include basic information of the delivery equipment used by the delivery personnel, such as the equipment number and equipment type of the delivery equipment. This application does not make any specific limitation on this, and may include basic information related to the delivery equipment that is well known in the art.

[0051] Step 220 , extracting features from the historical delivery data to obtain a plurality of candidate feature data affecting the delivery efficiency of the target delivery station and the delivery efficiency of the target delivery station.

[0052] In this step, after the above-mentioned execution entity obtains the historical distribution data corresponding to the target distribution station, it can perform feature analysis on the historical distribution data to determine the possible factors affecting the efficiency of the distribution station, which may include the distribution station site, distribution station-related personnel, distribution equipment, distribution goods, etc. The distribution station site mainly refers to the coverage area of ​​the distribution station, the characteristics of the station itself, etc. The distribution station-related personnel are mainly delivery personnel, management personnel, etc. The distribution equipment is mainly value-added delivery vehicles, conveyor belts and other factors. The distribution goods refer to the characteristics of the goods.

[0053] The execution subject extracts features from the historical delivery data corresponding to the target delivery station according to the result of feature analysis, and obtains multiple candidate feature data and the delivery efficiency of the target delivery station that are used to affect the delivery efficiency of the target delivery station. Among them, the multiple candidate feature data and the delivery efficiency of the target delivery station may include parcel waybill ratio, city level, delivery timeliness rate, number of road areas, collection and dispatch ratio, waybill density, delivery efficiency, etc. This application does not make specific restrictions on this, and may include any feature information that can affect the delivery efficiency of the target delivery station.

[0054] Step 230 , performing feature screening on a plurality of candidate feature data based on the delivery efficiency of the target delivery station, and screening out a plurality of target feature data from the plurality of candidate feature data.

[0055] In this step, after the above-mentioned execution entity obtains multiple candidate feature data and the delivery efficiency of the target delivery station through feature extraction, it can use the delivery efficiency of the target delivery station to analyze and judge the multiple candidate feature data to determine whether each candidate feature data can affect the delivery efficiency of the target delivery station and whether it is feature data that requires attribution judgment.

[0056] Specifically, the above-mentioned execution entity can analyze, judge and screen the characteristics of each candidate feature data according to the delivery efficiency of the target delivery station, determine whether each candidate feature data can affect the delivery efficiency of the target delivery station, delete the candidate feature data with little or no impact, and screen out multiple target feature data from multiple candidate feature data, and the multiple target feature data are all feature data that can affect the delivery efficiency of the target delivery station.

[0057] As an optional implementation, the above step 230, screening multiple candidate feature data based on the delivery efficiency of the target delivery station, and screening multiple target feature data from the multiple candidate feature data, may include the following steps:

[0058] In the first step, based on multiple candidate feature data and the distribution efficiency value of the target distribution station, multiple candidate feature data are regressed and fitted to obtain fitting data corresponding to the multiple candidate feature data.

[0059] Specifically, after the above-mentioned execution entity obtains multiple candidate feature data and the distribution efficiency of the target distribution station through feature extraction, it can use the distribution efficiency of the target distribution station to perform regression fitting on the multiple candidate feature data, and regress and fit the relationship between the multiple candidate feature data and the distribution efficiency through the regression model lasso (Least absolute shrinkage and selection operator), so as to obtain the fitting coefficient of each candidate feature data for the distribution efficiency, thereby obtaining the fitting data corresponding to the multiple candidate feature data.

[0060] In the second step, based on the first threshold and the fitting data corresponding to the plurality of candidate feature data, a plurality of first feature data are selected from the plurality of candidate feature data.

[0061] Specifically, after the execution subject obtains the fitting data corresponding to the plurality of candidate feature data, the fitting data corresponding to each candidate feature data is compared with the first threshold value, and it is determined whether the fitting data corresponding to each candidate feature data is greater than the first threshold value, and the candidate feature data corresponding to the fitting data greater than the first threshold value is screened out from the plurality of candidate feature data, thereby obtaining a plurality of first feature data. The first threshold value may be a value pre-set according to the fitting relationship, and may be 0, etc., and the present disclosure does not specifically limit this.

[0062] The third step is to calculate the importance levels of the plurality of first feature data to obtain importance level information of the plurality of first feature data.

[0063] Specifically, after the above-mentioned execution entity obtains multiple first feature data, it can use the delivery efficiency of the target delivery station to calculate the importance levels of the multiple first feature data, determine the importance of each first feature data for the delivery efficiency, and perform importance analysis on each first feature data through the catboost model to calculate the importance level information of each first feature data for the delivery efficiency.

[0064] In the fourth step, based on the second threshold and the importance level information of the plurality of first feature data, a plurality of target feature data are screened out from the plurality of first feature data.

[0065] Specifically, after the execution subject obtains the importance level information of multiple first feature data, it compares the importance level information of each first feature data with the second threshold, determines whether the importance level information of each first feature data is greater than the second threshold, and selects the first feature data corresponding to the importance level information greater than the second threshold from the multiple first feature data, thereby obtaining multiple target feature data. Among them, the second threshold can be a numerical value pre-set based on experience, and the present disclosure does not make specific limitations on this.

[0066] In this implementation, by using the regression model and the catboost model to perform association analysis and importance analysis on the candidate feature data, the candidate feature data is screened twice, so that the target feature data obtained is a more important factor for delivery efficiency, which improves the accuracy of the target feature data and obtains more important factors affecting delivery efficiency.

[0067] Step 240, using the Bayesian network model to perform causal inference calculations on the multiple target feature data, and generate information on the degree of influence of the multiple target feature data on the delivery efficiency of the target delivery station.

[0068] In this step, after the above-mentioned execution entity determines multiple target feature data, it can use a pre-trained Bayesian network model to perform causal inference calculations on the multiple target feature data, and calculate each target feature data based on the correlation between the multiple target feature data and the correlation with the distribution efficiency, to obtain the influence degree level information of each target feature data on the distribution efficiency of the target distribution station, thereby generating the influence degree level information of the multiple target feature data on the distribution efficiency of the target distribution station, wherein the influence degree level information can characterize the degree to which different values ​​of the target feature data can affect the distribution efficiency of the target distribution station with respect to the target feature data.

[0069] For example, the impact level information of multiple target characteristic data on the delivery efficiency of the target delivery station can be expressed as follows: when the target characteristic data is the waybill density, the different values ​​are [400, 600), [800, 1000), and the delivery efficiency is [250, +∞), the impact level information can be -0.3; when the target characteristic data is the vehicle arrival time, the different values ​​are earlier than 8 o'clock, 9 o'clock-10 o'clock, and the delivery efficiency is [200, 250), the impact level information can be 0.2.

[0070] As an optional implementation, the above step 240, using the Bayesian network model to perform causal inference calculation on multiple target feature data to generate the impact level information of the multiple target feature data on the delivery efficiency of the target delivery station, may include the following steps:

[0071] The first step is to discretize multiple target feature data to obtain multiple discrete target feature data.

[0072] Specifically, after the execution subject obtains a plurality of target feature data, it may perform discretization processing on each target feature data to obtain each discretized target feature data.

[0073] In the second step, the Bayesian network model is used to perform causal inference calculations on multiple discrete target feature data to generate information on the degree of influence of multiple discrete target feature data on the distribution efficiency of the target distribution station.

[0074] Specifically, after the above-mentioned execution entity obtains multiple discrete target feature data, it can use the pre-trained Bayesian network model to perform causal inference calculations on the multiple discrete target feature data, and calculate each discrete target feature data based on the correlation between the multiple discrete target feature data and the correlation with the distribution efficiency, to obtain the influence level information of each discrete target feature data on the distribution efficiency of the target distribution station, thereby generating the influence level information of multiple discrete target feature data on the distribution efficiency of the target distribution station.

[0075] In this implementation, by discretizing each target feature data to obtain discretized target feature data, it is possible to accurately determine that for the same target feature data, different values ​​can cause different distribution efficiencies, and accurately calculate the degree of influence of each discretized target feature data on the distribution efficiency of the target distribution station.

[0076] The information generation method provided by the embodiment of the present disclosure is as follows: the execution subject first obtains the historical distribution data corresponding to the target distribution station, then performs feature extraction on the historical distribution data, obtains multiple candidate feature data for affecting the distribution efficiency of the target distribution station and the distribution efficiency of the target distribution station, then performs feature screening on the multiple candidate feature data based on the distribution efficiency of the target distribution station, screens out multiple target feature data from the multiple candidate feature data, and finally uses the Bayesian network model to perform causal inference calculation on the multiple target feature data to generate multiple target feature data on the distribution efficiency of the target distribution station. Causal inference is performed using the historical distribution data of the target distribution station and the Bayesian network model, and attribution analysis is performed on the factors affecting the distribution efficiency, which can accurately and quickly determine the factors affecting the distribution efficiency, and further determine the influence level information of each factor affecting the distribution efficiency, so that users can better understand the important factors in the influencing factors, and realize the causal inference of the factors affecting the performance of the distribution station using the Bayesian network model at the terminal, without the need for human participation, thereby improving the accuracy and efficiency of determining the factors affecting the distribution efficiency of the distribution station.

[0077] See also Figure 3A , Figure 3A A flowchart 300 showing an embodiment of generating impact level information may include the following steps:

[0078] Step 310: construct a directed acyclic graph of multiple discretized target feature data based on the fitting data and importance level information corresponding to the multiple target feature data.

[0079] In this step, after the execution subject obtains multiple target feature data, it can obtain the fitting data and importance level information corresponding to the multiple target feature data according to the feature screening process of the multiple target feature data. Then, the execution subject can construct a directed acyclic graph of multiple discretized target feature data according to the fitting data and importance level information corresponding to the multiple target feature data, and the directed acyclic graph includes the causal relationship between the multiple target feature data and the delivery efficiency, and also includes the causal relationship between the multiple target feature data.

[0080] As an example, the above-mentioned multiple target feature data may include the collection-dispatch ratio, the parcel ratio, the earliest departure time, the average daily attendance time, the proportion of B network, the city level, the order density, the road area, etc. The constructed directed acyclic graph can be as follows: Figure 3B shown.

[0081] Step 320 , grouping the plurality of discretized target feature data according to the directed acyclic graph of the plurality of discretized target feature data to obtain a plurality of groups of target feature data.

[0082] In this step, after the above-mentioned execution subject obtains the directed acyclic graph of multiple discrete target feature data, the multiple discrete target feature data can be grouped according to the causal relationship between each discrete target feature data in the directed acyclic graph to obtain multiple groups of target feature data.

[0083] As an example, continue to refer to Figure 3B , multiple discrete target feature data can be divided into three groups, the collection and dispatch ratio and the package ratio as one group, the earliest departure time and the average daily attendance time as one group, and the B network proportion, city level, order density and road area as one group.

[0084] Step 330 , determining data types corresponding to the multiple groups of target feature data based on the fitting data and importance level information corresponding to the multiple target feature data.

[0085] In this step, after the above-mentioned execution entity obtains multiple groups of target feature data, for each group of target feature data, the data type corresponding to the target feature data is determined according to the fitting data and importance level information corresponding to the target feature data. The data type can characterize the type of the target feature data for distribution efficiency, and can include a first preset type that characterizes independent data and is not affected by other feature data, and can also include a second preset type that characterizes non-independent data and is affected by other feature data.

[0086] Step 340 , based on the data types corresponding to the multiple sets of target feature data, causal inference calculations are performed on the multiple sets of target feature data respectively to generate information on the degree of influence of multiple discrete target feature data on the delivery efficiency of the target delivery station.

[0087] In this step, after the above-mentioned execution entity obtains the data types corresponding to multiple groups of target feature data, for each group of target feature data, causal inference calculations can be performed on the target feature data using calculation methods of different data types. Different data types can use different algorithms for causal inference calculations, which can generate multiple discrete target feature data for the degree of influence of the distribution efficiency of the target distribution station.

[0088] In this embodiment, by grouping and determining the data type based on the association relationship between each target feature data in the directed acyclic graph, causal inference calculations can be performed on the feature data after different groups using the algorithm corresponding to the data type, so that the impact level information corresponding to each target feature data is more accurate.

[0089] refer to Figure 4 , Figure 4 A flowchart 400 of an embodiment of generating influence level information based on the data types corresponding to multiple sets of target feature data is shown, that is, the above step 340, based on the data types corresponding to the multiple sets of target feature data, causal inference calculations are performed on the multiple sets of target feature data respectively to generate influence level information of multiple discretized target feature data on the delivery efficiency of the target delivery station, which may include the following steps:

[0090] Step 410: construct a directed acyclic graph of multiple discretized target feature data based on the fitting data and importance level information corresponding to the multiple target feature data.

[0091] In this step, step 410 is the same as step 310 in the embodiment shown in FIG. 3 , and will not be described in detail here.

[0092] Step 420 , grouping the plurality of discrete target feature data according to the directed acyclic graph of the plurality of discrete target feature data to obtain a plurality of groups of target feature data.

[0093] In this step, step 420 is the same as step 320 in the embodiment shown in FIG. 3 , and will not be described in detail here.

[0094] Step 430 , determining data types corresponding to the multiple groups of target feature data based on the fitting data and importance level information corresponding to the multiple target feature data.

[0095] In this step, step 430 is the same as step 330 in the embodiment shown in FIG. 3 , and will not be described in detail here.

[0096] Step 440 : For each set of target characteristic data, in response to determining that the data type of the target characteristic data is a first preset type, obtaining the target characteristic data and the delivery efficiency of the target delivery station.

[0097] In this step, for each set of target feature data, the above-mentioned execution entity can determine, based on the fitting data and importance level information corresponding to the target feature data, that the data type corresponding to the target feature data is a first preset type that characterizes independent data and is not affected by other feature data. It can then be determined that the target feature data is not affected by other feature data, and there are no confounding variables, and the target feature data and the distribution efficiency of the target distribution station can be obtained.

[0098] As an example, refer to Figure 3B According to the correlation in the figure, it can be determined that the pick-up and dispatch ratio, the earliest departure time, the city level and the road area are not affected by other characteristic data, and the data type of the pick-up and dispatch ratio, the earliest departure time, the city level and the road area is determined to be the first preset type.

[0099] Step 450 , based on the target feature data and the delivery efficiency of the target delivery station, causal inference calculation is performed on the target feature data to generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0100] In this step, the above-mentioned execution entity obtains the target feature data and the distribution efficiency of the target distribution station, and can directly use the distribution efficiency of the target distribution station to perform causal inference calculations on the target feature data, and calculate the degree of influence of different values ​​corresponding to the target data features on the distribution efficiency.

[0101] As an example, if the target characteristic data of the first preset type is the earliest departure time, specifically, the earliest departure time can include three different values ​​of "earlier than 8 o'clock", "8 o'clock to 9 o'clock", and "9 o'clock to 10 o'clock" after discretization processing, and the delivery efficiency can be 6 different values ​​of [0,50), [50,100), [100,150), [150,200), [200,250), and [250,+∞). The influence level information of the earliest departure time on the delivery efficiency can be calculated for each delivery efficiency, and the difference between each two influence probabilities can be calculated. For example, the influence of the earliest departure time "earlier than 8 o'clock" and "8 o'clock to 9 o'clock" on the delivery efficiency [0,50) is calculated. If the calculated difference result is greater than 0, it means that the earliest departure time "earlier than 8 o'clock" has a greater impact on the delivery efficiency [0,50) than the earliest departure time "8 o'clock to 9 o'clock", and the difference is determined as the influence level information of the earliest departure time on the delivery efficiency.

[0102] In this embodiment, by using the above calculation method to calculate the target characteristic data of the first preset type, it is possible to accurately determine the level information of the impact degree of the target characteristic data on the delivery efficiency of the target delivery station.

[0103] Continue to refer Figure 4 The above step 340, based on the data types corresponding to the multiple sets of target feature data, performs causal inference calculations on the multiple sets of target feature data respectively to generate multiple discrete target feature data on the distribution efficiency of the target distribution station. It can also include the following steps:

[0104] Step 460 : For each set of target characteristic data, in response to determining that the data type of the target characteristic data is a second preset type, obtaining the target characteristic data, other characteristic data associated with the target characteristic data, and the delivery efficiency of the target delivery station.

[0105] In this step, for each set of target feature data, the above-mentioned execution entity can determine, based on the fitting data and importance level information corresponding to the target feature data, that the data type corresponding to the target feature data is a second preset type that characterizes non-independent data and is affected by other feature data. It can then be determined that the target feature data will be affected by other feature data and that there are confounding variables. The target feature data, other feature data associated with the target feature data, and the distribution efficiency of the target distribution station can be obtained.

[0106] As an example, refer to Figure 3B According to the correlation in the figure, it can be determined that the package ratio, average daily attendance time, and order density are all affected by other feature data, and the data type of the package ratio, average daily attendance time, and order density is determined to be the second preset type.

[0107] Step 470, based on the target feature data, other feature data associated with the target feature data and the delivery efficiency of the target delivery station, causal inference calculation is performed on the target feature data to generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0108] In this step, after the above-mentioned execution entity obtains the target characteristic data, other characteristic data associated with the target characteristic data and the distribution efficiency of the target distribution station, it can further analyze the target characteristic data to determine whether the target characteristic data is affected by one other characteristic data or multiple other characteristic data.

[0109] If it is determined that the target feature data is affected by other feature data, the above-mentioned execution entity may use the other feature data associated with the target feature data as a confounding variable of the target feature data, and use the other feature data associated with the target feature data and the delivery efficiency of the target delivery station to perform causal inference calculations on the target feature data, and calculate the degree of influence of different values ​​corresponding to the target data features on the delivery efficiency.

[0110] As an example, if the target characteristic data of the second preset type is the average daily attendance time, and the characteristic data related to the average daily attendance time is the earliest exit time, specifically, based on the known prior probability P (Z = z), the Bayesian network model can be used to calculate the posterior probability P (Y = y|X = x, Z = z), that is, the average daily attendance time, delivery efficiency and the confounding variable related to the average daily attendance time (earliest exit time) are calculated to obtain the impact probability of each average daily attendance time, and the difference between each two impact probabilities is calculated, and the difference is determined as the level of impact information of the average daily attendance on the delivery efficiency.

[0111] If it is determined that the target feature data is affected by multiple other feature data, the above-mentioned execution entity can use other feature data associated with the target feature data as confounding variables of the target feature data, and use the backdoor rule, other feature data associated with the target feature data, and the delivery efficiency of the target delivery station to perform causal inference calculations on the target feature data, and calculate the degree of influence of different values ​​corresponding to the target data features on the delivery efficiency.

[0112] As an example, if the target characteristic data of the second preset type is order density, and the characteristic data related to order density is road area and city level, specifically, the above-mentioned execution subject can use the backdoor criterion for calculation, that is, the order density, delivery efficiency and the confounding variables related to the order density (road area and city level) are calculated by the backdoor criterion. Among them, based on the known prior probabilities P(Z=z), P(W=w), the Bayesian network model can be used to calculate the posterior probability P(Y=y1|X=x1, Z=z, W=w), and calculate the degree of influence of different order densities on the delivery efficiency.

[0113] In this embodiment, by calculating the target characteristic data of the second preset type using the above calculation method, the influence level information of the target characteristic data on the delivery efficiency of the target delivery station can be accurately determined.

[0114] As an optional implementation, the above information generation method may further include the following steps:

[0115] In the first step, based on multiple target feature data and information on the degree of influence of the delivery efficiency of the target delivery station, efficiency prompt information corresponding to the target delivery station is generated.

[0116] Specifically, after the above-mentioned execution entity obtains the impact level information of multiple target characteristic data on the delivery efficiency of the target delivery station, it can determine the impact level information corresponding to each target characteristic data based on the impact level information of multiple target characteristic data on the delivery efficiency of the target delivery station, and generate corresponding efficiency prompt information.

[0117] Alternatively, the execution entity may also determine the target characteristic data with the greatest impact based on the impact level information of multiple target characteristic data on the delivery efficiency of the target delivery station, and generate corresponding efficiency prompt information based on the impact level information corresponding to the target characteristic data.

[0118] The second step is to send the efficiency reminder information to the target distribution station.

[0119] Specifically, after the execution subject generates the corresponding efficiency prompt information, the efficiency prompt information can be sent to the target delivery station, so that the target delivery station can improve and enhance the delivery process according to the efficiency prompt information.

[0120] In this implementation, by sending efficiency prompt information corresponding to the target delivery station to the target delivery station, the target delivery station can improve and enhance the delivery process according to the efficiency prompt information, thereby improving the delivery efficiency of the target delivery station.

[0121] refer to Figure 5As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an information generating device. Figure 2 The method embodiments shown correspond.

[0122] like Figure 5 As shown, the information generating device 500 of this embodiment may include: an acquisition module 510, a feature extraction module 520, a feature screening module 530 and a generation module 540.

[0123] The acquisition module 510 is configured to acquire the historical delivery data corresponding to the target delivery station;

[0124] The feature extraction module 520 is configured to extract features from the historical delivery data to obtain a plurality of candidate feature data affecting the delivery efficiency of the target delivery station and the delivery efficiency of the target delivery station;

[0125] The feature screening module 530 is configured to screen the plurality of candidate feature data based on the delivery efficiency of the target delivery station, and screen out a plurality of target feature data from the plurality of candidate feature data;

[0126] The generation module 540 is configured to perform causal inference calculation on the multiple target feature data using the Bayesian network model to generate the influence level information of the multiple target feature data on the delivery efficiency of the target delivery station.

[0127] In some optional implementations of the present embodiment, the feature screening module 530 is further configured to: based on multiple candidate feature data and the distribution efficiency value of the target distribution station, perform regression fitting on multiple candidate feature data to obtain fitting data corresponding to the multiple candidate feature data; based on a first threshold and the fitting data corresponding to the multiple candidate feature data, select multiple first feature data from the multiple candidate feature data; perform importance level calculation on the multiple first feature data to obtain importance level information of the multiple first feature data; based on a second threshold and the importance level information of the multiple first feature data, screen out multiple target feature data from the multiple first feature data.

[0128] In some optional implementations of the present embodiment, the generation module 540 includes: a discrete unit, configured to: discretize multiple target feature data to obtain multiple discrete target feature data; a generation unit, configured to: use a Bayesian network model to perform causal inference calculations on the multiple discrete target feature data to generate multiple discrete target feature data for the distribution efficiency of the target distribution station Impact level information.

[0129] In some optional implementations of the present embodiment, the generating unit is further configured to: construct a directed acyclic graph of multiple discrete target feature data based on the fitting data and importance level information corresponding to the multiple target feature data; group the multiple discrete target feature data according to the directed acyclic graph of the multiple discrete target feature data to obtain multiple groups of target feature data; determine the data types corresponding to the multiple groups of target feature data based on the fitting data and importance level information corresponding to the multiple groups of target feature data; perform causal inference calculations on the multiple groups of target feature data based on the data types corresponding to the multiple groups of target feature data, and generate influence level information of the multiple discrete target feature data on the distribution efficiency of the target distribution station.

[0130] In some optional implementations of the present embodiment, the generation unit is further configured to: for each set of target feature data, in response to determining that the data type of the target feature data is a first preset type, obtain the target feature data and the delivery efficiency of the target delivery station; based on the target feature data and the delivery efficiency of the target delivery station, perform causal inference calculations on the target feature data, and generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0131] In some optional implementations of the present embodiment, the generation unit is further configured to: for each set of target feature data, in response to determining that the data type of the target feature data is a second preset type, obtain the target feature data, other feature data associated with the target feature data, and the delivery efficiency of the target delivery station; based on the target feature data, other feature data associated with the target feature data, and the delivery efficiency of the target delivery station, perform causal inference calculations on the target feature data, and generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

[0132] In some optional implementations of the present embodiment, the device also includes a sending module; a generating unit is further configured to: generate efficiency prompt information corresponding to the target distribution station based on the impact level information of multiple target characteristic data on the distribution efficiency of the target distribution station; the sending module is configured to: send the efficiency prompt information to the target distribution station.

[0133] The information generating device provided by the above-mentioned embodiment of the present disclosure, the above-mentioned execution subject first obtains the historical distribution data corresponding to the target distribution station, and then extracts features from the historical distribution data to obtain multiple candidate feature data and the distribution efficiency of the target distribution station that affect the distribution efficiency of the target distribution station, and then performs feature screening on the multiple candidate feature data based on the distribution efficiency of the target distribution station, and screens out multiple target feature data from the multiple candidate feature data, and finally uses the Bayesian network model to perform causal inference calculation on the multiple target feature data to generate multiple target feature data on The distribution efficiency of the target distribution station is generated, and the historical distribution data of the target distribution station and the Bayesian network model are used to perform causal inference, and the factors affecting the distribution efficiency are analyzed attribution, which can accurately and quickly determine the factors affecting the distribution efficiency, and further determine the degree of influence of each factor affecting the distribution efficiency, so that users can better understand the important factors in the influencing factors, and realize the use of the Bayesian network model at the terminal to perform causal inference on the factors affecting the performance of the distribution station, without the need for human participation, thereby improving the accuracy and efficiency of determining the factors affecting the distribution efficiency of the distribution station.

[0134] Those skilled in the art will appreciate that the above device also includes some other well-known structures, such as a processor, a memory, etc. In order to unnecessarily obscure the embodiments of the present disclosure, these well-known structures are described in detail. Figure 5 Not shown.

[0135] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0136] Reference below Figure 6 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing an embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as smart screens, notebook computers, PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The terminal device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0137] like Figure 6As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0138] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 6 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0139] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, the above functions defined in the method of the embodiment of the present disclosure are executed. It should be noted that the computer-readable medium of the embodiment of the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In an embodiment of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, an apparatus, or a device. In an embodiment of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, an apparatus, or a device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wire, optical cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

[0140] Computer program code for performing the operations of the embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet).

[0141] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0142] The units involved in the embodiments described in the present application may be implemented by software or hardware. The units described may also be set in a processor, for example, it may be described as: a processor includes an acquisition module, a feature extraction module, a feature screening module and a generation module, wherein the names of these modules do not constitute a limitation on the modules themselves in some cases.

[0143] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device; or may exist independently without being assembled into the electronic device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: obtains the historical distribution data corresponding to the target distribution station; performs feature extraction on the historical distribution data to obtain multiple candidate feature data and the distribution efficiency of the target distribution station that affect the distribution efficiency of the target distribution station; performs feature screening on multiple candidate feature data based on the distribution efficiency of the target distribution station, and screens out multiple target feature data from the multiple candidate feature data; uses a Bayesian network model to perform causal inference calculations on multiple target feature data, and generates multiple target feature data The impact level information on the distribution efficiency of the target distribution station.

[0144] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) to form a technical solution.

Claims

1. A method for generating information, the method comprising: include: Obtain the historical delivery data corresponding to the target delivery station; Extracting features from the historical delivery data to obtain a plurality of candidate feature data affecting the delivery efficiency of the target delivery station and the delivery efficiency of the target delivery station; Performing feature screening on the plurality of candidate feature data based on the delivery efficiency of the target delivery station, and screening out a plurality of target feature data from the plurality of candidate feature data; A Bayesian network model is used to perform causal inference calculation on the multiple target feature data to generate information on the degree of influence of the multiple target feature data on the delivery efficiency of the target delivery station.

2. The method according to claim 1, in, The step of screening the plurality of candidate feature data based on the delivery efficiency of the target delivery station and screening out a plurality of target feature data from the plurality of candidate feature data comprises: Based on the plurality of candidate feature data and the delivery efficiency value of the target delivery station, performing regression fitting on the plurality of candidate feature data to obtain fitting data corresponding to the plurality of candidate feature data; Selecting a plurality of first feature data from the plurality of candidate feature data based on a first threshold and fitting data corresponding to the plurality of candidate feature data; Calculating the importance levels of the plurality of first feature data to obtain importance level information of the plurality of first feature data; A plurality of target feature data are screened out from the plurality of first feature data based on a second threshold and the importance level information of the plurality of first feature data.

3. The method according to claim 1 or 2, in, The using of the Bayesian network model to perform causal inference calculation on the plurality of target feature data to generate information on the degree of influence of the plurality of target feature data on the delivery efficiency of the target delivery station includes: Discretizing the plurality of target feature data to obtain a plurality of discretized target feature data; A Bayesian network model is used to perform causal inference calculation on the plurality of discrete target feature data to generate information on the degree of influence of the plurality of discrete target feature data on the delivery efficiency of the target delivery station.

4. The method according to claim 3, in, The using of the Bayesian network model to perform causal inference calculation on the plurality of discrete target feature data to generate information on the degree of influence of the plurality of discrete target feature data on the delivery efficiency of the target delivery station includes: Based on the fitting data and importance level information corresponding to the plurality of target feature data, construct a directed acyclic graph of the plurality of discretized target feature data; According to the directed acyclic graph of the plurality of discretized target feature data, the plurality of discretized target feature data are grouped to obtain a plurality of groups of target feature data; Determining the data types corresponding to the multiple groups of target feature data based on the fitting data and the importance level information corresponding to the multiple target feature data; Based on the data types corresponding to the multiple groups of target feature data, causal inference calculations are performed on the multiple groups of target feature data respectively to generate influence degree level information of the multiple discretized target feature data on the delivery efficiency of the target delivery station.

5. The method according to claim 4, in, The method of performing causal inference calculations on the multiple sets of target feature data based on the data types corresponding to the multiple sets of target feature data to generate information on the degree of influence of the multiple discretized target feature data on the delivery efficiency of the target delivery station includes: For each set of target characteristic data, in response to determining that the data type of the target characteristic data is a first preset type, obtaining the target characteristic data and the distribution efficiency of the target distribution station; Based on the target feature data and the delivery efficiency of the target delivery station, a causal inference calculation is performed on the target feature data to generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

6. The method according to claim 4, in, The method of performing causal inference calculations on the multiple sets of target feature data based on the data types corresponding to the multiple sets of target feature data to generate information on the degree of influence of the multiple discretized target feature data on the delivery efficiency of the target delivery station includes: For each set of target characteristic data, in response to determining that the data type of the target characteristic data is a second preset type, acquiring the target characteristic data, other characteristic data associated with the target characteristic data, and the delivery efficiency of the target delivery station; Based on the target feature data, other feature data associated with the target feature data and the delivery efficiency of the target delivery station, causal inference calculation is performed on the target feature data to generate impact level information of the target feature data on the delivery efficiency of the target delivery station.

7. The method according to any one of claims 1 to 6, further comprising: include: Based on the impact level information of the plurality of target characteristic data on the delivery efficiency of the target delivery station, generating efficiency prompt information corresponding to the target delivery station; The efficiency prompt information is sent to the target delivery station.

8. An information generating device, the device include: an acquisition module, configured to acquire historical delivery data corresponding to a target delivery station; a feature extraction module configured to extract features from the historical delivery data to obtain a plurality of candidate feature data affecting the delivery efficiency of the target delivery station and the delivery efficiency of the target delivery station; A feature screening module is configured to perform feature screening on the plurality of candidate feature data based on the delivery efficiency of the target delivery station, and screen out a plurality of target feature data from the plurality of candidate feature data; The generation module is configured to perform causal inference calculation on the multiple target feature data using a Bayesian network model to generate impact level information of the multiple target feature data on the delivery efficiency of the target delivery station.

9. The device according to claim 8, in, The feature screening module is further configured to: Based on the plurality of candidate feature data and the delivery efficiency value of the target delivery station, performing regression fitting on the plurality of candidate feature data to obtain fitting data corresponding to the plurality of candidate feature data; Selecting a plurality of first feature data from the plurality of candidate feature data based on a first threshold and fitting data corresponding to the plurality of candidate feature data; Calculating the importance levels of the plurality of first feature data to obtain importance level information of the plurality of first feature data; A plurality of target feature data are screened out from the plurality of first feature data based on a second threshold and the importance level information of the plurality of first feature data.

10. The device according to claim 8 or 9, in, The generating module comprises: The discretization unit is configured to: discretize the plurality of target feature data to obtain a plurality of discretized target feature data; The generating unit is configured to: perform causal inference calculation on the plurality of discrete target feature data using a Bayesian network model to generate influence level information of the plurality of discrete target feature data on the delivery efficiency of the target delivery station.

11. The device according to claim 10, in, The generating unit is further configured to: Based on the fitting data and importance level information corresponding to the plurality of target feature data, construct a directed acyclic graph of the plurality of discretized target feature data; According to the directed acyclic graph of the plurality of discretized target feature data, the plurality of discretized target feature data are grouped to obtain a plurality of groups of target feature data; Determining the data types corresponding to the multiple groups of target feature data based on the fitting data and the importance level information corresponding to the multiple target feature data; Based on the data types corresponding to the multiple groups of target feature data, causal inference calculations are performed on the multiple groups of target feature data respectively to generate influence degree level information of the multiple discretized target feature data on the delivery efficiency of the target delivery station.

12. The device according to claim 11, in, The generating unit is further configured to: For each set of target characteristic data, in response to determining that the data type of the target characteristic data is a first preset type, obtaining the target characteristic data and the distribution efficiency of the target distribution station; Based on the target feature data and the delivery efficiency of the target delivery station, a causal inference calculation is performed on the target feature data to generate information on the degree of influence of the target feature data on the delivery efficiency of the target delivery station.

13. The device according to claim 11, in, The generating unit is further configured to: For each set of target characteristic data, in response to determining that the data type of the target characteristic data is a second preset type, acquiring the target characteristic data, other characteristic data associated with the target characteristic data, and the delivery efficiency of the target delivery station; Based on the target feature data, other feature data associated with the target feature data and the delivery efficiency of the target delivery station, causal inference calculation is performed on the target feature data to generate impact level information of the target feature data on the delivery efficiency of the target delivery station.

14. The device according to any one of claims 8 to 13, further comprising a sending module; The generating unit is further configured to: generate efficiency prompt information corresponding to the target delivery station based on the impact level information of the plurality of target feature data on the delivery efficiency of the target delivery station; The sending module is configured to send the efficiency prompt information to the target delivery station.

15. An electronic device, include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 1 to 7.

16. A computer readable medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.