Freight rate index trend prediction method and device, equipment and storage medium

Abstract

The invention provides a freight rate index trend prediction method and device, equipment and a storage medium, and relates to the technical field of logistics data analysis. The invention discloses a method for predicting the trend of a freight rate index, and the method comprises the steps: obtaining the freight rate index and covariable data corresponding to the freight rate index; generating a candidate feature set based on the freight rate index and the covariable data according to a preset feature operator library; constructing a trend prediction model; and according to the candidate feature set, trend prediction of the freight rate index is carried out through a trend prediction model. According to the embodiment of the invention, the automatic machine learning technology can be applied to the trend prediction model, and the model prediction effect is improved.

Description

Technical Field

[0001] The present application relates to the technical field of logistics data analysis, and in particular to a method, device, equipment and storage medium for predicting the trend of a freight rate index. Background Art

[0002] The freight rate index is a set of indicators that can be constructed using industry logistics and transportation big data and expert knowledge, which can objectively guide market freight pricing. Predicting the freight rate index for multiple regions, multiple varieties, and multiple modes of transportation can guide the customization of freight rates for logistics routes in this industry.

[0003] Existing freight index forecasting usually uses the time series forecasting method, which requires manual data analysis, feature engineering, model selection and parameter adjustment of the forecast object. In the freight index forecasting scenario of multiple routes, multiple varieties and multiple modes of transportation, due to the large number of factors affecting cargo pricing and logistics routes, the traditional time series forecasting method needs to build feature engineering and forecasting models based on the freight index of each logistics route, which requires a large workload of model construction and parameter adjustment, and high operation and maintenance costs after model deployment. Summary of the invention

[0004] According to one aspect of the present application, a method for predicting the trend of a freight index is provided, comprising: obtaining a freight index and its corresponding covariate data; generating a candidate feature set based on the freight index and the covariate data according to a preset feature operator library; constructing a trend prediction model; and predicting the trend of the freight index through the trend prediction model according to the candidate feature set.

[0005] According to some embodiments, obtaining a freight rate index and its corresponding covariate data includes: obtaining transportation data under a preset scenario; and generating a freight rate index through the transportation data according to a preset configuration rule.

[0006] According to some embodiments, obtaining the freight rate index and its corresponding covariate data includes: obtaining the covariate data through a preset data interface; and preprocessing the covariate data based on a preset data processing operator.

[0007] According to some embodiments, a candidate feature set is generated based on a freight index and covariate data according to a preset feature operator library, including: obtaining multidimensional change features of the covariate data according to the feature operator library; obtaining multidimensional correlation features of the freight index and the covariate data according to the feature operator library; generating a candidate feature set based on the multidimensional change features and the multidimensional correlation features; and calculating importance parameters of the candidate features in the candidate feature set.

[0008] According to some embodiments, constructing a trend prediction model includes: configuring candidate sub-models corresponding to the trend prediction model according to a preset prediction model library; training the candidate sub-models through a preset training data set; optimizing parameters of the trained candidate sub-models through a preset test data set; and screening the candidate sub-models whose parameters have been optimized according to a preset fusion strategy library to form a trend prediction model.

[0009] According to some embodiments, according to a preset fusion strategy library, candidate sub-models that have been parameter-optimized are screened to form a trend prediction model, including: obtaining strategy matching parameters corresponding to the trend prediction model according to the fusion strategy library; determining candidate sub-models that match the strategy matching parameters among the candidate sub-models that have been parameter-optimized; and constructing a trend prediction model based on the candidate sub-models that match the strategy matching parameters.

[0010] According to some embodiments, trend prediction of the freight index is performed based on a candidate feature set through a trend prediction model, including: training the trend prediction model according to preset data indicators based on the candidate feature set; and outputting the trend prediction result of the freight index through the trained trend prediction model.

[0011] According to one aspect of the present application, a device for predicting the trend of a freight index is provided, comprising: a data acquisition module for acquiring a freight index and its corresponding covariate data; a data processing module for generating a candidate feature set based on the freight index and the covariate data according to a preset feature operator library; a model construction module for constructing a trend prediction model; and a data output module for predicting the trend of the freight index through the trend prediction model according to the candidate feature set.

[0012] According to one aspect of the present application, an electronic device is provided, comprising: 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 implement the aforementioned method.

[0013] According to one aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the aforementioned method is implemented.

[0014] According to the embodiments of the present application, data features can be automatically screened, and a trend prediction model can be constructed through freight index-related factors to predict the freight index, thereby improving the model prediction effect and reducing the model operation and maintenance costs.

[0015] It should be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application.

[0017] Figure 1 A flow chart showing a method for predicting a trend of a freight rate index according to an exemplary embodiment of the present application.

[0018] Figure 2 A block diagram showing a device for predicting a trend of a freight rate index according to an exemplary embodiment of the present application.

[0019] Figure 3 A block diagram of an electronic device according to an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0020] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this application will be comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0021] The described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application may be put into practice without one or more of these specific details, or other modes, components, materials, devices or operations may be adopted. In these cases, known structures, methods, devices, realizations, materials or operations will not be shown or described in detail.

[0022] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0023] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0024] The present application provides a method, device, equipment and storage medium for predicting the trend of a freight rate index, which can improve the prediction effect of the model.

[0025] A method, device, equipment and storage medium for predicting the trend of a freight rate index according to an embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0026] Figure 1 A flow chart showing a method for predicting a trend of a freight rate index according to an exemplary embodiment of the present application.

[0027] like Figure 1 As shown, in step S100, the freight rate index and its corresponding covariate data are obtained.

[0028] For example, in step S100, the prediction device generates a freight rate index through transportation data, and obtains covariate data corresponding to the freight rate index.

[0029] The prediction device obtains transportation data under a preset scenario.

[0030] According to some embodiments, the preset scenario includes the region and mode of transportation (eg, truck transportation, coastal transportation, and river transportation) for calculating the freight rate index, etc. The prediction device obtains the transportation data under the preset scenario through a preset data interface.

[0031] According to the preset configuration rules, the forecasting device generates a freight rate index through the acquired transportation data.

[0032] According to some embodiments, the preset configuration rules include a whitelist of transport routes corresponding to the freight rate index, a calculation time period of a base freight rate for calculating the freight rate index, and a weight of the transport route.

[0033] According to some embodiments, the prediction device preprocesses the acquired transportation data, and processes the preprocessed transportation data according to preset configuration rules to generate a freight rate index under a preset scenario.

[0034] For example, take the scenario of steel products transported along the river by the target steel plant. For the freight rate area where the target steel plant is located, the prediction device obtains transportation data based on the logistics market of steel products through a preset data interface. Furthermore, the prediction device pre-processes data quality issues such as the address of the place of delivery and receipt, abnormal freight rates, etc. in the transportation data through data statistical analysis, data interpolation and other methods, and generates a standard data format. Based on the pre-processed transportation data, the prediction device calculates the average base price of transportation, and adjusts the base price in combination with the corresponding logistics business rules to generate a freight rate index corresponding to the current scenario.

[0035] Furthermore, the prediction device obtains covariate data corresponding to the freight rate index through a preset data interface.

[0036] According to some embodiments, the covariate data includes industry internal data (including production and sales, inventory, etc.), industry upstream and downstream industry data and macro factor data, etc. Among them, the industry upstream industry data includes cost factor data (such as gasoline and diesel prices, etc.), the industry downstream industry data includes industry factor data (such as prosperity index, truck and ship production and sales, etc.), and the macro factor data includes manufacturing investment data and infrastructure investment data, etc.

[0037] Based on the preset data processing operator, the prediction device preprocesses the covariate data to clean the abnormal data in the covariate data.

[0038] According to some embodiments, the preset data processing operator includes a data completion operator and a data cleaning operator. Among them, the data completion operator adopts a time series interpolation algorithm, including forward filling, backward filling and linear interpolation. The algorithm adopted by the data cleaning operator includes address normalization, standard address resolution of the transportation area, identification and smoothing of index outliers, etc.

[0039] According to some embodiments, the preprocessed covariate data are arranged in time series.

[0040] In step S200, a candidate feature set is generated based on a preset feature operator library, a freight index and covariate data.

[0041] For example, in step S200, the prediction device obtains the multi-dimensional change characteristics of the covariate data and the multi-dimensional correlation characteristics of the freight index and the covariate data according to a preset feature operator library, and generates a candidate feature set based on them.

[0042] According to the preset feature operator library, the prediction device obtains the multi-dimensional change characteristics of the covariate data.

[0043] According to some embodiments, the preset feature operator library may adopt a feature operator library of automatic machine learning, which includes a time domain variation feature operator class, a frequency domain variation feature operator class, a time series correlation feature operator class, and the like.

[0044] According to some embodiments, the prediction device calculates and outputs the time domain change sequence set, frequency domain change sequence set and correlation time series of the covariate data through a feature operator library as multi-dimensional change features of the covariate data.

[0045] Furthermore, based on a preset feature operator library, the prediction device obtains multi-dimensional correlation features between the freight rate index and the covariate data.

[0046] Based on the multi-dimensional variation characteristics of the covariate data and the multi-dimensional correlation characteristics between the freight index and the covariate data, the prediction device generates a candidate feature set.

[0047] According to some embodiments, the candidate feature set includes statistical features such as variance, covariance, kurtosis, skewness, etc. of covariate data, and time series features such as lag terms, seasonal terms, trend terms, etc.

[0048] According to some embodiments, the prediction device can calculate the importance parameters of candidate features in the candidate feature set through linear regression of the basic model, GBDT (GradientBoosting Decision Tree) and other algorithms, and sort them by importance for feature selection when training the trend prediction model to prevent the prediction effect from deteriorating due to overfitting.

[0049] In step S300, a trend prediction model is constructed.

[0050] For example, in step S300, the prediction device configures a candidate sub-model corresponding to the trend prediction model and performs parameter optimization to form a trend prediction model.

[0051] According to the preset prediction model library, the prediction device configures a plurality of candidate sub-models corresponding to the trend prediction model.

[0052] According to some embodiments, before configuring multiple candidate sub-models, the prediction device may construct a training data set and a test data set through a pre-configured automatic machine learning model library for parameter tuning of the candidate sub-models.

[0053] According to some embodiments, the prediction model library may be a time series model library for automatic machine learning. The candidate sub-models configured by the prediction device through the prediction model library include statistical time series models (such as ARIMA, ETS, etc.), machine learning models (such as xgboost, LR, etc.) and deep learning models (such as deepAR, TFT, NBEATS, FCNN, TCN, etc.).

[0054] The prediction device trains the candidate sub-models through a preset training data set to obtain a set of candidate sub-models corresponding to the trend prediction model.

[0055] Furthermore, the prediction device verifies and optimizes the parameters of the trained candidate sub-models through a preset test data set.

[0056] According to the preset fusion strategy library, the prediction device selects candidate sub-models whose parameters have been optimized to form a trend prediction model.

[0057] According to some embodiments, the fusion strategy library may adopt a fusion candidate strategy library of automatic machine learning. The prediction device obtains strategy matching parameters corresponding to the trend prediction model from the fusion strategy library to improve the accuracy of the prediction. Among them, the strategy matching parameters include fusion strategy parameters such as ensemble learning and meta-learning.

[0058] The prediction device determines a candidate sub-model that matches the strategy matching parameter from the set of candidate sub-models that have undergone parameter optimization to complete the screening of the candidate sub-models.

[0059] Furthermore, the prediction device constructs a trend prediction model based on the screened candidate sub-models.

[0060] In step S400, a trend prediction model is used to predict the trend of the freight rate index according to the candidate feature set.

[0061] For example, in step S400, the prediction device trains a trend prediction model according to the candidate feature set, and outputs a trend prediction result of the freight rate index through the trend prediction model.

[0062] According to the preset data indicators, the prediction device trains the trend prediction model through the candidate feature set so that the prediction results output by the trend prediction model can be freely converted according to the preset data indicators.

[0063] The prediction device outputs the trend prediction result of the freight rate index through the trained trend prediction model.

[0064] According to some embodiments, the trend prediction result of the freight rate index includes the future trend of the freight rate index and the increase or decrease trend and increase or decrease range of the freight rate index in the next cycle.

[0065] According to some embodiments, the prediction device may continuously update the parameters of the trend prediction model based on the automatic machine learning model library to adapt to changes in the distribution of data features.

[0066] According to the embodiments of the present application, a trend prediction model can be constructed through freight rate index related factors to predict the freight rate index, thereby improving the model prediction effect.

[0067] Figure 2 A block diagram showing a device for predicting a trend of a freight rate index according to an exemplary embodiment of the present application.

[0068] like Figure 2 As shown, the prediction device 100 includes a data acquisition module 110 , a data processing module 120 , a model building module 130 and a data output module 140 .

[0069] The data acquisition module 110 acquires transportation data under a preset scenario.

[0070] According to preset configuration rules, the data acquisition module 110 generates a freight rate index through the acquired transportation data.

[0071] The data acquisition module 110 acquires the covariate data corresponding to the freight rate index through a preset data interface.

[0072] Based on a preset data processing operator, the data acquisition module 110 preprocesses the covariate data.

[0073] According to the preset feature operator library, the data processing module 120 obtains the multi-dimensional change characteristics of the covariate data.

[0074] According to the preset feature operator library, the data processing module 120 obtains the multi-dimensional correlation features between the freight index and the covariate data.

[0075] Based on the multi-dimensional variation characteristics of the covariate data and the multi-dimensional correlation characteristics between the freight rate index and the covariate data, the data processing module 120 generates a candidate feature set.

[0076] According to the preset prediction model library, the model construction module 130 configures a plurality of candidate sub-models corresponding to the trend prediction model.

[0077] The model building module 130 trains the candidate sub-models using a preset training data set to obtain a set of candidate sub-models corresponding to the trend prediction model.

[0078] The model building module 130 verifies and optimizes the parameters of the trained candidate sub-models using a preset test data set.

[0079] According to the preset fusion strategy library, the model building module 130 selects candidate sub-models whose parameters have been optimized to form a trend prediction model.

[0080] According to the preset data indicators, the data output module 140 trains the trend prediction model through the candidate feature set.

[0081] The data output module 140 outputs the trend prediction result of the freight rate index through the trained trend prediction model.

[0082] Figure 3 A block diagram of an electronic device according to an exemplary embodiment of the present application is shown.

[0083] like Figure 3 As shown, the electronic device 600 is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0084] like Figure 3As shown, the electronic device 600 is in the form of a general-purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processing unit 610), a display unit 640, etc. The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 executes the methods described in this specification according to various exemplary embodiments of the present application. For example, the processing unit 610 may execute the following: Figure 1 The method shown in .

[0085] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0086] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0087] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0088] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also 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 a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0089] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by combining software with necessary hardware. The technical solution according to the embodiment of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal or a network device, etc.) to execute the method according to the embodiment of the present application.

[0090] The software product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable 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 thereof.

[0091] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0092] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0093] The computer-readable medium carries one or more programs. When the one or more programs are executed by a device, the computer-readable medium implements the aforementioned functions.

[0094] Those skilled in the art will appreciate that the above modules can be distributed in the device according to the description of the embodiment, or can be changed accordingly and only used in one or more devices different from the embodiment. The modules of the above embodiments can be combined into one module, or further divided into multiple sub-modules.

[0095] The embodiments of the present application are described in detail above, and the description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, changes or deformations made by those skilled in the art based on the idea of ​​the present application, the specific implementation method and the scope of application of the present application, all belong to the scope of protection of the present application. In summary, the content of this specification should not be construed as limiting the present application.

Claims

1. A method for predicting the trend of a freight rate index, characterized in that: include: Obtain freight rate index and its corresponding covariate data; Generate a candidate feature set based on the freight rate index and the covariate data according to a preset feature operator library; Build trend prediction models; According to the candidate feature set, the trend prediction model is used to perform trend prediction of the freight rate index.

2. The method according to claim 1, characterized in that Get the freight index and its corresponding covariate data, including: Obtain transportation data under preset scenarios; The freight rate index is generated through the transportation data according to a preset configuration rule.

3. The method according to claim 1, characterized in that Get the freight index and its corresponding covariate data, including: Acquire the covariate data through a preset data interface; The covariate data is preprocessed based on a preset data processing operator.

4. The method according to claim 1, characterized in that: According to a preset feature operator library, a candidate feature set is generated based on the freight rate index and the covariate data, including: According to the feature operator library, obtaining multidimensional change characteristics of the covariate data; According to the feature operator library, obtaining multi-dimensional correlation features of the freight rate index and the covariate data; generating the candidate feature set based on the multi-dimensional variation feature and the multi-dimensional correlation feature; and Calculate the importance parameters of the candidate features in the candidate feature set.

5. The method according to claim 1, characterized in that Build trend prediction models, including: According to a preset prediction model library, configuring candidate sub-models corresponding to the trend prediction model; Training the candidate sub-model using a preset training data set; Optimize the parameters of the trained candidate sub-models through a preset test data set; According to the preset fusion strategy library, candidate sub-models whose parameters have been optimized are screened to form the trend prediction model.

6. The method according to claim 5, characterized in that According to the preset fusion strategy library, the candidate sub-models that have been parameter-optimized are screened to form the trend prediction model, including: According to the fusion strategy library, obtaining strategy matching parameters corresponding to the trend prediction model; Determining a candidate sub-model that matches the strategy matching parameter among the candidate sub-models whose parameters have been optimized; The trend prediction model is constructed based on the candidate sub-models that match the strategy matching parameters.

7. The method according to claim 1, characterized in that According to the candidate feature set, the trend prediction model is used to predict the trend of the freight rate index, including: According to the candidate feature set, the trend prediction model is trained according to preset data indicators; The trend prediction result of the freight rate index is outputted through the trained trend prediction model.

8. A device for predicting the trend of a freight rate index, characterized in that: include: The data acquisition module is used to obtain the freight index and its corresponding covariate data; A data processing module, used to generate a candidate feature set based on the freight index and the covariate data according to a preset feature operator library; Model building module, used to build trend prediction models; The data output module is used to perform trend prediction of the freight rate index according to the candidate feature set through the trend prediction model.

9. An electronic device, characterized in that: 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 implement the method according to any one of claims 1 to 7.

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

Images