CN110598989B

CN110598989B - Goods source quality evaluation method, device, equipment and storage medium

Info

Publication number: CN110598989B
Application number: CN201910748763.7A
Authority: CN
Inventors: 李轩增
Original assignee: Jiangsu Manyun Software Technology Co Ltd
Current assignee: Jiangsu Manyun Software Technology Co Ltd
Priority date: 2019-08-14
Filing date: 2019-08-14
Publication date: 2022-09-23
Anticipated expiration: 2039-08-14
Also published as: CN110598989A

Abstract

The embodiment of the invention discloses a goods source quality evaluation method, a device, equipment and a storage medium, wherein the method comprises the following steps: acquiring historical goods source data; extracting goods source characteristics from the historical goods source data, wherein the goods source characteristics comprise goods source attribute characteristics and scene guide characteristics, and the scene guide characteristics are used for quantitatively describing factors influencing the feedback condition of the goods source under a specific scene; and training by using the goods source characteristics to obtain a goods source quality evaluation model so as to evaluate the quality of goods sources generated on line in real time. According to the embodiment of the invention, by acquiring the comprehensive goods source attribute characteristics and combining the scene guide characteristics, the trained goods source quality evaluation model has higher accuracy, so that the goods source is accurately evaluated, and a data basis is provided for further improving the vehicle and goods matching rate.

Description

Goods source quality evaluation method, device, equipment and storage medium

Technical Field

The embodiment of the invention relates to the technical field of internet, in particular to a goods source quality evaluation method, a goods source quality evaluation device, goods source quality evaluation equipment and a storage medium.

Background

With the development of the internet and intelligent terminal technology, the birth of various applications brings great convenience to the life of people. In the field of freight transportation, a goods owner can place an order for a delivery source through the application on the intelligent terminal, fill order information such as source attributes and delivery routes, and the like, and the vehicle owner can select an order which is favorite and can be accepted according to the order information, so that the whole process from delivery to receipt can be realized on the internet, and the working efficiency is improved.

In addition to completing the most basic source delivery orders, many other downstream businesses have been derived to better serve users, such as counting and analyzing the failed orders to improve the response rate of the orders, helping the car owners to find the source meeting the requirements more quickly, and then improving the matching efficiency of the car and the goods. Due to the large number of goods owners, the orders are large in number and different in content, and therefore the demand for automatically evaluating the quality of goods sources is generated.

However, in the method for evaluating the quality of the goods sources in the prior art, the extracted features of the goods sources cannot effectively distinguish the goods sources, so that the accuracy of the method cannot meet the requirement.

Disclosure of Invention

The embodiment of the invention provides a goods source quality assessment method, a goods source quality assessment device, goods source quality assessment equipment and a storage medium, so that the accuracy of goods source quality assessment is improved.

In a first aspect, an embodiment of the present invention provides a method for evaluating quality of a source of goods, including:

acquiring historical goods source data;

extracting goods source characteristics from the historical goods source data, wherein the goods source characteristics comprise goods source attribute characteristics and scene guide characteristics, and the scene guide characteristics are used for quantitatively describing factors influencing the feedback condition of the goods source under a specific scene;

and training by using the goods source characteristics to obtain a goods source quality evaluation model so as to evaluate the quality of goods sources generated on line in real time.

In a second aspect, an embodiment of the present invention further provides a source quality assessment apparatus, including:

the data acquisition module is used for acquiring historical goods source data;

the characteristic extraction module is used for extracting goods source characteristics from the historical goods source data, wherein the goods source characteristics comprise goods source attribute characteristics and scene guide characteristics, and the scene guide characteristics are used for quantitatively describing factors influencing the feedback condition of the goods source under a specific scene;

and the model training module is used for obtaining a goods source quality evaluation model by utilizing the goods source characteristic training so as to evaluate the quality of the goods source generated on line in real time.

In a third aspect, an embodiment of the present invention further provides a computer device, including:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement a method of source quality assessment as described in any of the embodiments of the invention.

In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for evaluating the quality of a cargo source according to any embodiment of the present invention.

According to the embodiment of the invention, the goods source characteristics are extracted from the historical goods source data, the goods source quality evaluation model is obtained by utilizing the goods source characteristic training, so that the quality of the goods source generated on line in real time is evaluated, in addition, not only the goods source attribute characteristics but also the scene guide characteristics are considered in the goods source characteristics, namely, the characteristics of the factors influencing the feedback condition of the goods source under a specific scene are quantitatively described, so that the goods source is better distinguished by combining the feedback condition of the goods source under the specific scene, and the accuracy of the model for the quality evaluation of the goods source is improved.

Drawings

FIG. 1 is a flow chart of a method for evaluating the quality of a source of goods according to a first embodiment of the present invention;

FIG. 2 is a flowchart of a quality evaluation method for a source of goods according to a second embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a source quality evaluation apparatus according to a third embodiment of the present invention;

fig. 4 is a schematic structural diagram of a computer device in the fourth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It should be further noted that, for the convenience of description, only some of the structures associated with the present invention are shown in the drawings, not all of them.

According to the embodiment of the invention, the machine learning model is utilized according to the requirements in the vehicle and goods matching scene, the evaluation model is trained by combining the goods source attribute characteristic and the scene guide characteristic, and the goods source quality is quantitatively evaluated according to the preference degree of a driver for goods. The transverse characteristic data support is provided for each service scene of the vehicle and goods matching platform, the multi-dimensional description of goods attributes is facilitated, drivers are helped to find goods meeting requirements more quickly, and the vehicle and goods matching efficiency is improved finally.

Fig. 1 is a flowchart of a method for evaluating quality of a cargo source according to a first embodiment of the present invention, which is applicable to a situation where quality of a cargo source is evaluated, for example, in a scenario of matching vehicles and cargos, efficiency of matching vehicles and cargos is improved by evaluating quality of a cargo source. The method may be performed by a source quality assessment apparatus, which may be implemented in software and/or hardware, and may be configured in a computer device. As shown in fig. 1, the method specifically includes:

s101, obtaining historical goods source data.

For example, historical sourcing data may be obtained for the previous week, month, or more. The goods source is goods to be transported in the field of transportation, information of the goods to be transported, transportation related information or information required by other orders in each order is obtained as the goods source information, and the information can be used as a data base of the embodiment of the invention.

S102, extracting goods source characteristics from the historical goods source data, wherein the goods source characteristics comprise goods source attribute characteristics and scene guide characteristics, and the scene guide characteristics are used for quantitatively describing factors influencing the feedback condition of the goods source under a specific scene.

In particular, the source attribute characteristics generally refer to characteristics directly related to the source, including, but not limited to: a first type of cargo, a second type of cargo, a loading time, a unloading time, a loading ground warp latitude, an unloading ground warp latitude, a loading and unloading manner, a minimum required vehicle length, a truck type, a cargo weight, a cargo volume, whether an invoice is required, whether a receipt is required, a pricing unit, whether a piece goods source, whether a guaranty goods source, an amount of deposit, a length of a remark item, whether a local is not visible, whether a quoted goods source, a desired freight rate, whether a one-port price goods source, and the like. The extraction of the source attribute features may be based on the identification and extraction of specific fields in the source data.

The scene oriented features are features generated aiming at a specific scene and used for quantitatively describing factors influencing the feedback condition of the goods source under the specific scene, for example, the features generated aiming at the business item form under the scene of matching the goods and vehicles can describe the influence of iteration of the product form on the subsequent feedback condition of the goods source from the perspective of wave.

The scene oriented features may include, for example, source route length, source hot-end, and destination potential source supply conditions; wherein the route length is represented by a straight-line distance between a start location to a destination location in the source data; the hot degree of the goods source is represented by the positioning information of the starting place in the goods source data; the destination potential source supply condition is represented by an offset angle of the destination in azimuth relative to the origin. Specifically, the above features can be expressed by polar coordinate conversion, that is, the route is expressed by vinegar boiling latitude and longitude in a polar coordinate method.

Optionally, the scene guide feature may further include a degree of deviation of a size or volume of the source relative to a set vehicle space in the source property feature; wherein the degree of deviation is represented by a difference in size or volume of the source and the set vehicle space. Specifically, the size information of the vehicle can be known by the vehicle type included in the source attribute feature, and the cheapness degree refers to the difference between the size or volume of the source and the size information of the vehicle, for example, the offset degree is represented by how much the length of the source exceeds the length of the vehicle or how much the length of the source exceeds the length of the vehicle.

Because the scene guide characteristics can be used as factors influencing the preference degree of the car owner to the goods source in the scene, the feedback condition of the goods source can be learned in a more targeted manner by the model through the expansion of the goods source characteristics, particularly the scene guide characteristics of the goods source combined with the expansion of the specific scene, and the factors influencing the feedback condition of the goods source in the specific scene through quantitative description, so that the model has more accurate goods source distinguishing capability in the specific scene, and the goods source evaluation result is accurately given.

S103, training by using the goods source characteristics to obtain a goods source quality evaluation model so as to evaluate the quality of goods sources generated on line in real time.

The goods source quality evaluation model quantitatively evaluates corresponding scores based on the prediction of the probability that goods sources are fed back in a specific time. Specifically, a machine learning method, such as lightGBM (fast, distributed, high-performance gradient boosting framework based on decision tree algorithm), may be employed to train the model.

The process of obtaining the cargo quality assessment model by utilizing the cargo characteristic training can comprise the following steps:

classifying the source characteristics according to the weeks generated by the source data to which the source characteristics belong to obtain source characteristic sets with different weeks;

and respectively taking the source feature set of each week number as the input of the model, taking the labeling result of whether the source corresponding to each source feature is fed back within a specific time as the output, and training to obtain the source quality evaluation model corresponding to different week numbers.

It should be noted that, in the vehicle-cargo matching scenario, the behavior of the driver and the cargo owner appears obviously periodic, for example, the cargo sources on monday are generally more and the probability that the cargo sources are selected by the vehicle owner is also higher, whereas the cargo sources on friday are generally fewer and the probability that the cargo sources are selected by the vehicle owner is also lower. Therefore, the source characteristics can be divided according to seven days in a week to obtain source characteristic sets of days from Monday to Sunday respectively, then the source characteristic sets of each week number are used as the input of the model for training respectively, the output of the model can be the result of whether the pre-labeled source is fed back within a specific time, for example, if the pre-labeled source is fed back, the pre-labeled source can be labeled as 1, and if the pre-labeled source is not fed back, the pre-labeled source can be labeled as 0, so that the model can predict the probability that the source can be fed back according to the characteristics of the current source, and quantitatively evaluate the corresponding score. And training by using the characteristics of different week numbers to obtain the source quality assessment model corresponding to each week number, and enabling each source quality assessment model to learn the source characteristics of the corresponding week numbers in a targeted manner, so that the accuracy of the model in source assessment can be further improved.

In addition, after the offline training model is deployed, the processing process of the offline features can be transferred into the flink real-time calculation module, and the evaluation result of the quality of the goods source obtained through calculation is sent to the preset feature library to be stored. Under the scene of vehicle-cargo matching, the quality evaluation score of the cargo source can be read from the key value of the feature library, and the score can be integrated into other features of the scene and provides a data basis for the subsequent customized adjustment of downstream services.

For a specific scene of vehicle and cargo matching, although the offline performance of the complex model (500 boosting iterations) is good, the return time in the real-time computation module is greatly increased, which results in that the cargo search request generated before the computation is completed cannot apply the computed evaluation result, so in a specific implementation manner, at most 200 boosting iterations can be selected to balance the offline performance of the model and the online performance efficiency.

The embodiment of the invention extracts the goods source characteristics from the historical goods source data, obtains the goods source quality evaluation model by utilizing the goods source characteristic training, and evaluates the quality of the goods source generated on line in real time, and considers not only the goods source attribute characteristics but also the scene guide characteristics in the goods source characteristics, namely the characteristics of the factors for quantitatively describing the influence on the feedback condition of the goods source in a specific scene, thereby better distinguishing the goods source by combining the feedback condition of the goods source in the specific scene, quantitatively evaluating the corresponding score by utilizing the goods source quality evaluation model based on the prediction of the probability that the goods source is fed back in a specific time, and improving the accuracy of the model for evaluating the quality of the goods source.

Example two

Fig. 2 is a flowchart of a cargo source quality evaluation method in the second embodiment of the present invention, and the second embodiment is further optimized based on the first embodiment. As shown in fig. 2, the method includes:

s201, obtaining historical goods source data.

S202, determining whether target source attribute data with a value exceeding a preset truncation threshold exists or not according to at least one target source attribute data in the historical source data, wherein different types of target source attribute data correspond to different preset truncation thresholds. If so, executing S203, otherwise, directly executing S204.

S203, adjusting the value of the target source attribute data to be the corresponding preset truncation threshold value.

Since the source data is usually filled in on the platform by the party to be delivered, there are many subjective factors, even those who are distracting. Therefore, in order to improve the accuracy of model prediction and evaluation, the embodiment of the invention also needs to perform rationality prediction on the goods source data. Specifically, the overall industry information may be combined with investigation in advance to determine target source attributes of multiple dimensions, where unreasonable data may exist, such as source weight, source volume, and transportation distance, and then perform rationality detection on the data of the target source attributes.

Firstly, extracting at least one target source attribute data from the historical source data, then comparing the target source attribute data with a preset truncation threshold, wherein the truncation threshold can be an initial value of the target source attribute data, applying the initial value to an actual method, checking whether the evaluation result of the model reaches the expectation, iterating and optimizing the initial value in the process, and determining the truncation threshold when the model reaches the expectation. Here, different target source attribute data correspond to different preset cutoff thresholds. And finally, adjusting the numerical value of the target source attribute data exceeding the corresponding preset cut-off threshold value to the corresponding preset cut-off threshold value.

For example, when the target source attribute data is 100 tons in weight, and the cutoff threshold corresponding to the source attribute data of the cargo weight is 50 tons in weight, the value of 100 tons in the target source attribute data is modified to 50 tons; for another example, when the target cargo source attribute data is a transportation distance of 1 kilometer, and the cutoff threshold corresponding to the cargo source attribute data is 3000 kilometers, the value of 1 kilometer of the target cargo source attribute data is modified to 5000 kilometers. The above is merely an example, and the embodiment of the present invention is not limited in this respect.

Through the numerical modification of the target source attribute data exceeding the preset truncation threshold, the target source attribute data is equivalent to truncation of the overlong tail part data of the target source attribute data, a small number of outlier data points are removed after truncation, the reasonability of the data and the accurate assessment of the whole source can be guaranteed, and meanwhile, the characteristics exceeding the truncation point can still be reasonably estimated.

S204, extracting goods source characteristics from the historical goods source data, wherein the goods source characteristics comprise goods source attribute characteristics and scene guide characteristics, and the scene guide characteristics are used for quantitatively describing factors influencing the feedback condition of the goods source under a specific scene.

S205, training by using the goods source characteristics to obtain a goods source quality evaluation model so as to evaluate the quality of goods sources generated in real time on line, wherein the goods source quality evaluation model quantitatively evaluates corresponding scores based on the prediction of the probability that the goods sources are fed back in a specific time.

According to the embodiment of the invention, the goods source quality evaluation model is trained by utilizing the comprehensive goods source attribute characteristics and the scene guide characteristics, the goods sources are better distinguished by combining the feedback condition of the goods sources under a specific scene, the quality evaluation of the goods sources generated on line in real time is realized, and the accuracy is higher. And through the rationality detection to the goods source data, cut off overlength afterbody score data, can not guarantee the rationality of data and the accurate aassessment to whole goods source accurately, simultaneously, the characteristic that surpasss the cut-off point also can obtain comparatively reasonable prediction yet to further improve the degree of accuracy that the model was appraised.

EXAMPLE III

Fig. 3 is a schematic structural diagram of a source quality assessment apparatus according to a third embodiment of the present invention, which is applicable to a situation of assessing source quality, for example, in a scenario of matching vehicles and cargos, the efficiency of matching vehicles and cargos is improved by assessing the source quality. As shown in fig. 3, the apparatus includes:

the data acquisition module 310 is used for acquiring historical source data;

a feature extraction module 320, configured to extract source features from the historical source data, where the source features include source attribute features and scene guide features, and the scene guide features are used to quantitatively describe factors that affect the feedback condition of the source under a specific scene;

and the model training module 330 is configured to train the characteristics of the goods sources to obtain a goods source quality assessment model, so as to perform quality assessment on the goods sources generated in real time on the line.

Optionally, the scene guidance features at least comprise a source route length, a source hot degree and a destination potential source supply condition;

wherein the route length is represented by a straight-line distance between a start location to a destination location in the source data;

the hot degree of the goods source is represented by the positioning information of the starting place in the goods source data;

the destination potential sourcing condition is represented by an offset angle of the destination in azimuth relative to the origin.

Optionally, the scene guidance feature at least includes a degree of deviation of a size or volume of the source relative to a set vehicle space in the source attribute feature;

wherein the degree of deviation is represented by a difference in size or volume of the source and the set vehicle space.

Optionally, the apparatus further includes a data truncation module, specifically configured to:

before the characteristic extraction module extracts the goods source characteristics from the historical goods source data, determining whether the object goods source attribute characteristics with the numerical value exceeding a preset truncation threshold exist or not according to at least one object goods source attribute characteristic in the historical goods source data, wherein different object goods source attribute characteristics correspond to different preset truncation thresholds;

and if so, adjusting the numerical value of the target source attribute characteristic to be a corresponding preset truncation threshold value.

Optionally, the target source attribute data includes at least one of the following: source weight, source volume, and travel distance.

Optionally, the source quality evaluation model quantitatively evaluates a corresponding score based on a prediction of a probability that a source is fed back within a specific time;

accordingly, the model training module 330 is specifically configured to:

The goods source quality evaluation device provided by the embodiment of the invention can execute the goods source quality evaluation method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the goods source quality evaluation method.

Example four

Fig. 4 is a schematic structural diagram of a computer device according to a fourth embodiment of the present invention. FIG. 4 illustrates a block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention. The computer device 12 shown in fig. 4 is only an example and should not bring any limitations to the function and scope of use of the embodiments of the present invention.

As shown in FIG. 4, computer device 12 is in the form of a general purpose computing device. The components of computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including the system memory 28 and the processing unit 16.

Bus 18 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Computer device 12 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)30 and/or cache memory 32. Computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 4, and commonly referred to as a "hard drive"). Although not shown in FIG. 4, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 by one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.

A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally carry out the functions and/or methodologies of embodiments of the invention as described.

Computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), with one or more devices that enable a user to interact with computer device 12, and/or with any devices (e.g., network card, modem, etc.) that enable computer device 12 to communicate with one or more other computing devices. Such communication may be through an input/output (I/O) interface 22. Also, computer device 12 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via network adapter 20. As shown, the network adapter 20 communicates with the other modules of the computer device 12 over the bus 18. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, to implement the cargo source quality assessment method provided by the embodiment of the present invention, including:

acquiring historical goods source data;

extracting goods source characteristics from the historical goods source data, wherein the goods source characteristics comprise goods source attribute characteristics and scene guide characteristics, and the scene guide characteristics are used for quantitatively describing factors influencing the feedback condition of goods sources under a specific scene;

EXAMPLE five

The fifth embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for evaluating the quality of a source of goods provided by the fifth embodiment of the present invention, and the method includes:

acquiring historical goods source data;

Computer storage media for embodiments of the present invention may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It is to be noted that the foregoing description is only exemplary of the invention and that the principles of the technology may be employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for evaluating the quality of a source of goods, comprising:

acquiring historical goods source data;

training by using the goods source characteristics to obtain a goods source quality evaluation model so as to evaluate the quality of goods sources generated on line in real time;

the goods source quality evaluation model quantitatively evaluates corresponding scores based on the prediction of the probability that the goods source is fed back in a specific time;

correspondingly, the obtaining of the goods source quality assessment model by using the goods source characteristic training comprises the following steps:

and respectively taking the source characteristic set of each week number as the input of the model, taking the labeling result of whether the source corresponding to each source characteristic is fed back within a specific time as the output, and training to obtain the source quality evaluation models corresponding to different week numbers.

2. The method of claim 1, wherein the scenario-oriented features include at least a source route length, a source hot-end degree, and a destination potential source supply condition;

wherein the route length is represented by a straight-line distance between a start location to a destination in the source data;

the destination potential source supply condition is represented by an offset angle of the destination in azimuth relative to the origin.

3. The method of claim 1, wherein the scene guidance features include at least a degree of deviation of a size or volume of a source relative to a set vehicle space in the source attribute features;

4. The method of any of claims 1-3, wherein prior to extracting source features from the historical source data, the method further comprises:

determining whether target goods source attribute data with a numerical value exceeding a preset truncation threshold exists or not according to at least one target goods source attribute data in the historical goods source data, wherein different types of target goods source attribute data correspond to different preset truncation thresholds;

and if so, adjusting the value of the target source attribute data to be a corresponding preset truncation threshold value.

5. The method of claim 4, wherein the target source property data comprises at least one of: source weight, source volume, and travel distance.

6. A source quality assessment apparatus, comprising:

the data acquisition module is used for acquiring historical goods source data;

the characteristic extraction module is used for extracting goods source characteristics from the historical goods source data, wherein the goods source characteristics comprise goods source attribute characteristics and scene guide characteristics, the scene guide characteristics are used for quantitatively describing factors influencing the feedback condition of a goods source in a specific scene, and the goods source characteristics are classified according to the week number generated by the goods source data to which the goods source characteristics belong to, so that goods source characteristic sets with different week numbers are obtained;

the model training module is used for training by using the source characteristics to obtain a source quality evaluation model, respectively using the source characteristic set of each week number as the input of the model, using the labeling result of whether the source corresponding to each source characteristic is fed back within a specific time as the output, training to obtain the source quality evaluation model corresponding to different week numbers, and performing quality evaluation on the source generated in real time, wherein the source quality evaluation model quantitatively evaluates corresponding scores based on the prediction of the probability that the source is fed back within the specific time.

7. The apparatus according to claim 6, wherein the apparatus further comprises a data truncation module, specifically configured to:

8. A computer device, comprising:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the sourcing quality assessment method of any of claims 1-5.

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method of source quality assessment according to any one of claims 1-5.