Logistics business data analysis method, apparatus, electronic device and storage medium

Abstract

The present application relates to the technical field of logistics information verification, and in particular relates to a logistics business data analysis method, an apparatus, an electronic device and a storage medium. The present application provides the logistics business data analysis method, the apparatus, the electronic device and the storage medium. The method comprises: first acquiring a business analysis instruction and logistics business data; on the basis of the business analysis instruction and the logistics business data, generating requirement analysis input information; and inputting the requirement analysis input information into a logistics large language model, such that the logistics large language model performs data analysis on the logistics business data according to logistics analysis requirements, so as to generate a data analysis result. The logistics large language model can perform data analysis on logistics business data according to logistics analysis requirements, so as to generate data analysis results, such that associated metrics and information between various logistics business data can be used to improve processing efficiency for analysis of a large amount of logistics business data.

Description

Logistics business data analysis method and device, electronic device, and storage medium Technical Field

[0001] The present application relates to the field of logistics information verification technology, and in particular to a logistics business data analysis method and its device, electronic equipment, and storage medium. Background Art

[0002] Logistics data analysis usually requires programming languages ​​(such as Python, SQL, etc.) and data analysis tools, as well as some statistical and mathematical knowledge. Therefore, the task of handling logistics data analysis places high demands on logistics business personnel. When the amount of logistics business data is large, relying on manpower to perform logistics data analysis is inefficient. However, since there may be related indicators or information between the various logistics business data in the same logistics data platform, if logistics data analysis is completely done without manpower, the correlation between the business data will be difficult to utilize, which will also result in low efficiency of efficiency data analysis. Based on this, how to use the related indicators and information between various logistics business data to improve processing efficiency when analyzing large amounts of logistics business data has become a problem that needs to be solved urgently in the industry. Summary of the Invention

[0003] The main purpose of the embodiments of the present application is to propose a logistics business data analysis method and its device, electronic device, and storage medium, which can utilize the related indicators and information between multiple logistics business data to improve processing efficiency when analyzing large amounts of logistics business data.

[0004] To achieve the above objectives, a first aspect of an embodiment of the present application proposes a logistics business data analysis method, comprising:

[0005] Obtaining business analysis instructions and logistics business data; wherein the business analysis instructions are used to reflect logistics analysis requirements for the logistics business data;

[0006] Generating demand analysis input information according to the business analysis instruction and the logistics business data;

[0007] The demand analysis input information is input into the logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein, the logistics big language model is pre-trained through multiple logistics training data with data association relationships in the logistics data platform, and the logistics data platform is used to store the logistics training data.

[0008] According to some embodiments provided by the present application, before inputting the demand analysis input information into the logistics large language model, the logistics large language model is pre-trained based on the logistics training data, specifically including:

[0009] Obtaining a logistics training data set from the logistics data platform, and configuring training annotation information for the logistics training data; wherein the logistics training data set includes a plurality of the logistics training data having the data association relationship;

[0010] The logistics training data set and the multiple training annotation information are input into the large pre-trained language model for model fine-tuning training to obtain the training output result of the large pre-trained language model for the logistics training data. When the training output result matches the training annotation information, the pre-trained logistics large language model is obtained.

[0011] According to some embodiments provided in this application, obtaining a logistics training dataset from the logistics data platform includes:

[0012] Extracting a plurality of logistics training data from the logistics data platform;

[0013] Performing association analysis on the plurality of logistics training data to obtain the data association relationship;

[0014] Dividing the first number of logistics training data into a second number of logistics training data groups according to the data association relationship; wherein the logistics training data in the same logistics training data group have the data association relationship;

[0015] The second number of the logistics training data groups are integrated to obtain the logistics training data set.

[0016] According to some embodiments provided by the present application, the logistics training data set and the plurality of training annotation information are input into the large pre-trained language model for model fine-tuning training, and a training output result of the large pre-trained language model for the logistics training data is obtained. When the training output result matches the training annotation information, the pre-trained logistics large language model is obtained, including:

[0017] Inputting the logistics training data of the same logistics training data group and the training annotation information corresponding to the logistics training data into the large pre-trained language model for model fine-tuning training;

[0018] During model fine-tuning training, the training output results of the large pre-trained language model for the same logistics training data set are matched with the corresponding training annotation information to obtain the pre-trained logistics large language model.

[0019] According to some embodiments provided by the present application, each of the logistics training data sets corresponds to a candidate analysis requirement;

[0020] During the model fine-tuning training, the training output results of the large pre-trained language model for the same logistics training data set are matched with the corresponding training annotation information to obtain the pre-trained logistics large language model, including:

[0021] In the model fine-tuning training, a target logistics training data group is determined from the plurality of logistics training data groups; wherein the candidate analysis requirements corresponding to the target logistics training data group are the target analysis requirements;

[0022] When the training output results of the large pre-trained language model for the target logistics training data group match the corresponding training annotation information, the logistics large language model suitable for processing the target analysis requirements is obtained.

[0023] According to some embodiments provided by this application, inputting the demand analysis input information into the logistics large language model so that the logistics large language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results includes:

[0024] Determining the logistics analysis requirements corresponding to the requirements analysis input information;

[0025] When the logistics analysis requirements match the target analysis requirements, the logistics business data is analyzed using the logistics large language model adapted to process the target analysis requirements to generate the data analysis results.

[0026] According to some embodiments provided by the present application, performing association analysis on the plurality of logistics training data to obtain the data association relationship includes:

[0027] configuring a data correlation coefficient for every two of the logistics training data in the plurality of logistics training data based on a preset logistics business rule;

[0028] Integrating the data correlation coefficients between every two pieces of logistics training data in the plurality of logistics training data to generate a correlation matrix;

[0029] The data association relationship between the plurality of logistics training data is determined according to the correlation matrix.

[0030] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a logistics business data analysis device, the device comprising:

[0031] An acquisition module, configured to acquire business analysis instructions and logistics business data; wherein the business analysis instructions are used to reflect logistics analysis requirements for the logistics business data;

[0032] An input information generation module, configured to generate demand analysis input information based on the business analysis instruction and the logistics business data;

[0033] The data analysis module is used to input the demand analysis input information into the logistics large language model, so that the logistics large language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein, the logistics large language model is pre-trained through multiple logistics training data with data association relationships in the logistics data platform, and the logistics data platform is used to store the logistics training data.

[0034] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the method described in the first aspect when executing the computer program.

[0035] To achieve the above-mentioned purpose, the fourth aspect of the embodiments of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method described in the first aspect.

[0036] This application proposes a logistics business data analysis method and its device, electronic device, and storage medium, which need to first obtain business analysis instructions and logistics business data; wherein, the business analysis instructions are used to reflect the logistics analysis requirements for the logistics business data; according to the business analysis instructions and the logistics business data, demand analysis input information is generated; the demand analysis input information is input into a logistics large language model, so that the logistics large language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein, the logistics large language model is pre-trained through multiple logistics training data with data association relationships in a logistics data platform, and the logistics data platform is used to store the logistics training data. Since the logistics large language model can perform data analysis on the logistics business data according to the logistics analysis requirements and generate data analysis results, it can improve processing efficiency when analyzing large amounts of logistics business data by utilizing the indicators and information associated between multiple logistics business data. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] FIG1 is a flow chart of a logistics business data analysis method provided in an embodiment of the present application;

[0038] FIG2 is a flowchart of pre-training the large logistics language model before step S103 in FIG1 ;

[0039] FIG3 is a flow chart of step S201 in FIG2 ;

[0040] FIG4 is a flow chart of step S302 in FIG3 ;

[0041] FIG5 is a flow chart of step S202 in FIG2 ;

[0042] FIG6 is a flow chart of step S402 in FIG4 ;

[0043] FIG7 is a flow chart of step S103 in FIG1 ;

[0044] FIG8 is a schematic diagram of the structure of a logistics business data analysis device provided in an embodiment of the present application;

[0045] FIG9 is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0047] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and so on, in the specification, claims, and drawings, are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application. In addition, the identification of specific steps hereinafter does not represent a limitation on the order of steps and execution logic, and the order of execution and execution logic between each step should be understood and inferred with reference to the contents described in the embodiments.

[0049] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0050] Logistics data analysis usually requires programming languages ​​(such as Python, SQL, etc.) and data analysis tools, as well as some statistical and mathematical knowledge. Therefore, the task of handling logistics data analysis places high demands on logistics business personnel. When the amount of logistics business data is large, relying on manpower to perform logistics data analysis is inefficient. However, since there may be related indicators or information between the various logistics business data in the same logistics data platform, if logistics data analysis is completely done without manpower, the correlation between the business data will be difficult to utilize, which will also result in low efficiency of efficiency data analysis. Based on this, how to use the related indicators and information between various logistics business data to improve processing efficiency when analyzing large amounts of logistics business data has become a problem that needs to be solved urgently in the industry.

[0051] The following is a further explanation based on the accompanying drawings.

[0052] Referring to Figure 1, the main purpose of the embodiment of the present application is to propose a logistics business data analysis method and its device, electronic device, and storage medium, which can utilize the related indicators and information between multiple logistics business data to improve processing efficiency when analyzing large amounts of logistics business data.

[0053] To achieve the above objectives, a first aspect of an embodiment of the present application proposes a logistics business data analysis method, which may include, but is not limited to, the following steps S101 to S103.

[0054] Step S101: obtaining a business analysis instruction and logistics business data; wherein the business analysis instruction is used to reflect the logistics analysis requirements for the logistics business data;

[0055] Step S102: Generate demand analysis input information based on the business analysis instructions and logistics business data;

[0056] In step S103, the demand analysis input information is input into the logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein, the logistics big language model is pre-trained through multiple logistics training data with data association relationships in the logistics data platform, and the logistics data platform is used to store the logistics training data.

[0057] Through steps S101 to S103, this application proposes a logistics business data analysis method and its device, electronic device, and storage medium. It is necessary to first obtain business analysis instructions and logistics business data; wherein the business analysis instructions are used to reflect the logistics analysis requirements for the logistics business data; according to the business analysis instructions and the logistics business data, demand analysis input information is generated; the demand analysis input information is input into the logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein, the logistics big language model is pre-trained through multiple logistics training data with data association relationships in the logistics data platform, and the logistics data platform is used to store the logistics training data. Since the logistics big language model can perform data analysis on the logistics business data according to the logistics analysis requirements and generate data analysis results, it is possible to use the indicators and information associated between multiple logistics business data to improve processing efficiency when analyzing large amounts of logistics business data.

[0058] In step S101 of some embodiments, business analysis instructions and logistics business data are obtained; wherein, the business analysis instructions are used to reflect the logistics analysis needs for the logistics business data. It should be noted that the logistics business data is data involved in the logistics business, which can cover a variety of types, such as the number of parcels received and delivered in a specific area; the number of parcels processed in a certain transfer site; the number of parcels from area A to area C in a certain time period, the transfer sites along the way, the transportation time, the type of vehicle used for transportation, etc. It should be pointed out that the business analysis instructions are used to reflect the logistics analysis needs for the logistics business data, and are instructions for starting the logistics business data analysis process. In some embodiments, in order to realize the logistics analysis needs, the demander needs to generate a business analysis instruction corresponding to the logistics analysis needs to start the analysis of the logistics business data.

[0059] According to some more specific embodiments of the present application, data related to various types of logistics services can be stored and managed on a logistics data platform. Specifically, the various types of logistics service data can come from the same logistics data platform or from different logistics data platforms. Based on this, logistics service data can be obtained by querying and downloading data from the logistics data platform.

[0060] In some embodiments, step S102 generates demand analysis input information based on the business analysis instructions and logistics business data. It should be noted that after obtaining the business analysis instructions and logistics business data, the demand analysis input information can be generated based on the business analysis instructions and logistics business data. It should be noted that the demand analysis input information is used to instruct the logistics large language model to perform data analysis on the logistics business data in accordance with the logistics analysis requirements. It should be understood that the demand analysis input information can be a guide for instructing the logistics large language model to perform data analysis on the logistics business data in accordance with the logistics analysis requirements.

[0061] It should be clarified that the prompt, as the name implies, is text information that plays a guiding role. It should be understood that in the process of applying the logistics big language model to analyze logistics business data in accordance with logistics analysis requirements, in order to obtain evaluation results corresponding to the logistics analysis requirements, it is necessary to pre-draft a prompt text for conveying the logistics analysis requirements to the logistics big language model. If the prompt text is arbitrarily drafted in the process of conveying the preset requirements to the logistics big language model, then the prompt text obtained in this way is difficult to conform to the expression paradigm of the logistics big language model. Therefore, some embodiments of the present application can generate corresponding prompts as demand analysis input information based on business analysis instructions and logistics business data. In this way, better analysis results can be demonstrated in the process of applying the logistics big language model to analyze logistics business data in accordance with logistics analysis requirements.

[0062] In step S103 of some embodiments, the demand analysis input information is input into the logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein, the logistics big language model is pre-trained through multiple logistics training data with data association relationships in the logistics data platform, and the logistics data platform is used to store the logistics training data.

[0063] Large-scale pre-trained language models are those pre-trained on large text corpora. These models typically employ self-supervised learning methods, trained on large amounts of unlabeled text data to learn the linguistic structure and semantic information within the text. These models possess powerful representational capabilities and can be applied to a variety of natural language processing tasks, such as text generation, text classification, sequence labeling, and machine translation. Furthermore, large-scale pre-trained language models can be adapted to specific task requirements through techniques such as fine-tuning, achieving even better performance. Logistics large-scale language models are large-scale pre-trained language models that can be used for data analysis of logistics business data.

[0064] It should be noted that by inputting demand analysis input information into the logistics big language model, under the guidance of the demand analysis input information, the logistics big language model can perform data analysis on logistics business data according to the logistics analysis requirements and generate data analysis results. It should be clarified that the logistics big language model is pre-trained using multiple logistics training data with data associations in the logistics data platform, and the logistics data platform is used to store logistics training data. If two logistics data in the logistics data platform serve the same logistics business, then these two logistics data can be considered to be related indicators or information, that is, there is a data association relationship between the two logistics data. It should be understood that data association relationships are not limited to two logistics data belonging to the same logistics business. If two types of logistics businesses are inherently related, then two logistics data with data association relationships can also exist in two already related logistics businesses. It should be noted that data association relationships can cover a variety of types, including, but not limited to, the examples mentioned above.

[0065] It should be understood that data related to various types of logistics operations can be stored and managed within the logistics data platform. Therefore, logistics training data can be extracted from the logistics data platform and used to pre-train the logistics large language model. The pre-trained logistics large language model is capable of analyzing logistics business data based on logistics analysis requirements. This allows it to leverage correlated indicators and information across diverse logistics business data to improve processing efficiency when analyzing large amounts of logistics business data.

[0066] Referring to Figure 2, according to some embodiments provided by the present application, before inputting the demand analysis input information into the logistics large language model in step S103, the logistics large language model is also pre-trained based on the logistics training data, which may specifically include steps S201 to S202.

[0067] Step S201: Obtain a logistics training data set from a logistics data platform and configure training annotation information for the logistics training data; wherein the logistics training data set includes a plurality of logistics training data with data association relationships;

[0068] In step S202, the logistics training data set and multiple training annotation information are input into the large pre-trained language model for model fine-tuning training to obtain the training output results of the large pre-trained language model for the logistics training data. When the training output results match the training annotation information, a pre-trained logistics large language model is obtained.

[0069] It should be noted that although large-scale pre-trained language models have powerful language representation capabilities, they have their limitations. If they are not optimized for logistics data analysis tasks, they may not be able to obtain analysis results that meet the needs of logistics analysis. When a large-scale pre-trained language model that has not been optimized is applied to analyze logistics business data according to logistics analysis requirements, there may be situations where effective analysis results cannot be obtained. In this regard, the present disclosure provides an embodiment shown in steps S201 to S202,

[0070] In some embodiments, steps S201 and S202 involve obtaining a logistics training dataset from a logistics data platform and configuring training annotation information for the logistics training data. The logistics training dataset includes multiple logistics training data with data associations. The logistics training dataset and the multiple training annotations are input into a large pre-trained language model for model fine-tuning, resulting in a training output of the large pre-trained language model for the logistics training data. When the training output matches the training annotation information, a pre-trained logistics large language model is obtained. It should be emphasized that if two pieces of logistics data on the logistics data platform serve the same logistics business, then the two pieces of logistics data can be considered related indicators or information, meaning that the two pieces of logistics data have a data association relationship. It should be understood that data association relationships are not limited to two pieces of logistics data belonging to the same logistics business. If two types of logistics businesses are inherently related, then two pieces of logistics data with a data association relationship can also exist within two already related types of logistics businesses. It should be noted that data association relationships can encompass a variety of types, including, but not limited to, the examples above. Based on this, the logistics training dataset includes multiple sets of logistics training data with data correlations. When the logistics training dataset and multiple training annotations are input into a large pre-trained language model for fine-tuning, the data correlations between the multiple logistics training data sets can be leveraged to improve the large pre-trained language model's analytical capabilities for the logistics training data. When the training output matches the training annotations, the large pre-trained language model's analytical capabilities for the logistics training data have met the expected requirements. At this point, fine-tuning the large pre-trained language model can be completed, resulting in the pre-trained large logistics language model.

[0071] In the embodiment of the present application shown in step S201 to step S202, a logistics training data set is obtained from a logistics data platform, and training annotation information is configured for the logistics training data; wherein, the logistics training data set includes a plurality of logistics training data with data association relationships; the logistics training data set and the plurality of training annotation information are input into a large pre-trained language model for model fine-tuning training, and the training output result of the large pre-trained language model for the logistics training data is obtained. When the training output result matches the training annotation information, a pre-trained logistics large language model is obtained. Since the data association relationship between the plurality of logistics training data can be utilized in the process of model fine-tuning training for the large pre-trained language model, the analysis capability of the large pre-trained language model for the logistics training data can be improved. Therefore, the logistics large language model pre-trained in this way is used to process demand analysis input information, so as to perform data analysis on the logistics business data according to the logistics analysis requirements and generate data analysis results, which can improve the accuracy of the data analysis results while improving the efficiency of logistics business data processing.

[0072] 3 , according to some embodiments provided in the present application, step S201 of acquiring a logistics training data set from a logistics data platform may include, but is not limited to, the following steps S301 to S304 .

[0073] Step S301, extracting multiple logistics training data from the logistics data platform;

[0074] Step S302: performing association analysis on multiple logistics training data to obtain data association relationships;

[0075] Step S303: dividing the first number of logistics training data into a second number of logistics training data groups according to the data association relationship; wherein the logistics training data in the same logistics training data group have a data association relationship;

[0076] Step S304 : Integrate the second number of logistics training data sets to obtain a logistics training data set.

[0077] In some embodiments, step S301 involves extracting multiple logistics training data from a logistics data platform. It should be noted that data related to various types of logistics operations can be stored and managed on the logistics data platform. The various types of logistics training data can come from the same logistics data platform or from different logistics data platforms. Therefore, the logistics training data can be obtained by querying or downloading data from the logistics data platform.

[0078] In step S302 of some embodiments, association analysis is performed on multiple logistics training data to obtain data association relationships. It should be noted that the purpose of performing association analysis on multiple logistics training data is to determine which logistics training data, among the multiple logistics training data, have data association relationships. It should be emphasized that if two logistics data in the logistics data platform serve the same logistics business, then the two logistics data can be considered to be related indicators or information, that is, the two logistics data have a data association relationship. Based on this, the data association relationships can be clarified by determining which logistics training data, among the multiple logistics training data, serve the same logistics business.

[0079] 4 , according to some embodiments provided in the present application, step S302 performs association analysis on multiple logistics training data to obtain data association relationships, which may include, but is not limited to, the following steps S401 to S403 .

[0080] Step S401, configuring a data correlation coefficient for every two logistics training data in a plurality of logistics training data based on a preset logistics business rule;

[0081] Step S402, integrating the data correlation coefficients between every two logistics training data in the plurality of logistics training data to generate a correlation matrix;

[0082] Step S403: determining the data association relationship between the plurality of logistics training data according to the correlation matrix.

[0083] In some embodiments, step S401 configures a data correlation coefficient for each pair of logistics training data in the plurality of logistics training data based on preset logistics business rules. It should be noted that the preset logistics business rules refer to pre-set business rules related to the logistics business, which disclose logistics data serving the same logistics business. Based on this, by comparing the plurality of logistics training data within the logistics business data, the strength of the correlation between the logistics training data can be quantified, thereby configuring the corresponding data correlation coefficient.

[0084] In some embodiments, step S402 integrates the data correlation coefficients between every two pieces of logistics training data in the plurality of logistics training data to generate a correlation matrix. It should be noted that a correlation matrix, also called a correlation coefficient matrix, is composed of the data correlation coefficients between the columns of the matrix. If the plurality of logistics training data is divided into a first group and a second group, wherein the first group of logistics training data is arranged as rows of the correlation matrix and the second group of logistics training data is arranged as columns of the correlation matrix, then the i-th row and j-th column of the correlation matrix represents the data correlation coefficient between the i-th piece of logistics training data in the first group of logistics training data and the j-th piece of logistics training data in the second group of logistics training data.

[0085] In some embodiments, step S403 determines the data association relationship between the multiple logistics training data sets based on the correlation matrix. It should be noted that the data association relationship between the multiple logistics training data sets is determined based on the strength of the correlation between the logistics training data sets based on the correlation matrix. For example, if the correlation matrix indicates that the data association coefficient between two pieces of logistics training data sets exceeds a preset correlation determination threshold, then this indicates that a data association relationship exists between the two pieces of logistics training data sets.

[0086] Through steps S401 to S403, a data correlation coefficient is assigned to each pair of logistics training data in the plurality of logistics training data based on preset logistics business rules. The data correlation coefficients between each pair of logistics training data in the plurality of logistics training data are integrated to generate a correlation matrix. Based on the correlation matrix, the data correlation relationships between the plurality of logistics training data are determined. Because the correlation matrix reflects the data correlation coefficients between each pair of logistics training data in the plurality of logistics training data, the data correlation relationships between the plurality of logistics training data can be quickly determined based on the correlation matrix. This helps improve the efficiency of dividing the first number of logistics training data into the second number of logistics training data groups.

[0087] It should be understood that there are various implementation methods for determining the data association relationship between multiple logistics training data, which may include, but are not limited to, the specific embodiments listed above.

[0088] In some embodiments, step S303 involves dividing the first number of logistics training data into a second number of logistics training data groups based on data association relationships; wherein the logistics training data within the same logistics training data group have data association relationships. It should be noted that only after clarifying which logistics training data serve the same logistics business and clarifying the data association relationships can the first number of logistics training data be further divided into the second number of logistics training data groups based on the data association relationships, wherein the logistics training data within the same logistics training data group have data association relationships.

[0089] In some embodiments, step S304 involves integrating the second number of logistics training data sets to obtain a logistics training data set. It should be noted that the second number of logistics training data sets can only be integrated to obtain the logistics training data set after the first number of logistics training data sets are divided into the second number of logistics training data sets based on the data association relationship.

[0090] In the embodiment of the present application shown in steps S301 to S304, multiple logistics training data are extracted from the logistics data platform, and correlation analysis is performed on the multiple logistics training data to obtain data correlation relationships. The first number of logistics training data are divided into a second number of logistics training data groups according to the data correlation relationships; wherein, the logistics training data in the same logistics training data group have data correlation relationships, and the second number of logistics training data groups are integrated to obtain a logistics training data set. Different logistics training data groups can be determined based on different data correlation relationships, so as to form a logistics training data set. When the logistics training data set and multiple training annotation information are further input into a large pre-trained language model for model fine-tuning training, the training effect can be further improved. Therefore, the logistics large language model pre-trained in this way is used to process demand analysis input information to perform data analysis on logistics business data according to logistics analysis requirements and generate data analysis results. This can improve the accuracy of data analysis results while improving the efficiency of logistics business data processing.

[0091] Referring to Figure 5, according to some embodiments provided by the present application, step S202 inputs the logistics training data set and multiple training annotation information into the large pre-trained language model for model fine-tuning training to obtain the training output results of the large pre-trained language model for the logistics training data. When the training output results match the training annotation information, a pre-trained logistics large language model is obtained, which may include, but is not limited to, the following steps S501 to S502.

[0092] Step S501: Input the logistics training data of the same logistics training data group and the training annotation information corresponding to the logistics training data into a large pre-trained language model for model fine-tuning training;

[0093] In step S502 , during model fine-tuning training, the training output results of the large pre-trained language model for the same logistics training data set are matched with the corresponding training annotation information to obtain a pre-trained logistics large language model.

[0094] In some embodiments, steps S501 and S502 involve inputting the logistics training data and the corresponding training annotation information from the same logistics training data set into a large pre-trained language model for model fine-tuning training. During model fine-tuning training, the training output of the large pre-trained language model for the same logistics training data set matches the corresponding training annotation information, resulting in a pre-trained logistics large language model. It should be noted that, since different logistics training data sets were previously determined based on different data associations during the formation of the logistics training data set, each logistics training data set can correspond to a specific logistics business. On this basis, the logistics training data and the corresponding training annotation information from the same logistics training data set are input into the large pre-trained language model for model fine-tuning training, thereby improving the data analysis capabilities of the large pre-trained language model for this specific logistics business. When different logistics training data sets correspond to different logistics businesses, after inputting multiple logistics training data sets and the corresponding training annotation information from the logistics training data into the large pre-trained language model for model fine-tuning training, the large pre-trained language model can perform different types of logistics data analysis based on different types of logistics businesses. It should be emphasized that when the logistics training data set includes multiple logistics training data groups, when the training output results of the large pre-trained language model for the same logistics training data group match the corresponding training annotation information during model fine-tuning training, it means that the large pre-trained language model's analysis capabilities for logistics data under the same logistics business have reached the expected requirements. At this time, the model fine-tuning training of the large pre-trained language model can be terminated to obtain the pre-trained logistics large language model. In this way, the logistics large language model pre-trained in this way is used to process demand analysis input information to perform data analysis on logistics business data according to logistics analysis requirements and generate data analysis results. This can improve the efficiency of logistics business data processing while improving the applicability of data analysis results for different types of logistics businesses.

[0095] Referring to Figure 6 , according to some embodiments provided herein, each logistics training data set corresponds to a candidate analysis requirement. Step S402 During model fine-tuning training, the training output of the large pre-trained language model for the same logistics training data set is matched with the corresponding training annotation information to obtain a pre-trained logistics large language model. This process may include, but is not limited to, steps S601 to S602 .

[0096] Step S601: During model fine-tuning training, a target logistics training data set is determined from multiple logistics training data sets; wherein the candidate analysis requirements corresponding to the target logistics training data set are the target analysis requirements;

[0097] In step S602 , when the training output of the large pre-trained language model for the target logistics training data set matches the corresponding training annotation information, a large logistics language model adapted to process the target analysis requirements is obtained.

[0098] In step S601 of some embodiments, during model fine-tuning training, a target logistics training data set is determined from multiple logistics training data sets; the candidate analysis requirement corresponding to the target logistics training data set is the target analysis requirement. It should be noted that each logistics training data set corresponds to a candidate analysis requirement for a certain type of logistics business. Based on this, during model fine-tuning training, the target logistics training data set is determined from multiple logistics training data sets, with the goal of using the target logistics training data set to specifically train the data analysis capabilities of the large pre-trained language model for a specific candidate analysis requirement.

[0099] In step S602 of some embodiments, when the training output of the large pre-trained language model for the target logistics training data set matches the corresponding training annotation information, a large logistics language model adapted to process the target analysis requirements is obtained. It should be noted that when the training output of the large pre-trained language model for the target logistics training data set matches the corresponding training annotation information, it means that the data analysis capability of the large pre-trained language model for the target analysis requirements has met the expected requirements. At this point, model fine-tuning training of the large pre-trained language model can be terminated, resulting in the pre-trained large logistics language model.

[0100] In the embodiment of the present application shown in steps S601 to S602, during model fine-tuning training, a target logistics training data group is determined from multiple logistics training data groups; wherein, the candidate analysis requirements corresponding to the target logistics training data group are target analysis requirements; when the training output results of the large pre-trained language model for the target logistics training data group match the corresponding training annotation information, a large logistics language model adapted to process the target analysis requirements is obtained. It should be noted that the large logistics language model pre-trained in this way is used to process demand analysis input information, so as to perform data analysis on logistics business data according to logistics analysis requirements and generate data analysis results. This can improve the efficiency of logistics business data processing while further improving the applicability of data analysis results to different types of analysis requirements.

[0101] Referring to Figure 7, according to some embodiments provided in the present application, step S103 inputs the demand analysis input information into the logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results, which may include, but is not limited to, the following steps S701 to S702.

[0102] Step S701, determining the logistics analysis requirements corresponding to the demand analysis input information;

[0103] Step S702: When the logistics analysis requirements match the target analysis requirements, the logistics business data is analyzed using a logistics large language model adapted to process the target analysis requirements to generate data analysis results.

[0104] In some embodiments, steps S701 and S702 determine the logistics analysis requirements corresponding to the demand analysis input information. When the logistics analysis requirements match the target analysis requirements, a logistics large language model adapted to handle the target analysis requirements is used to perform data analysis on the logistics business data, generating a data analysis result. It should be noted that during the pre-training process for the logistics large language model, different types of target logistics training data sets can be utilized to train the logistics large language model's logistics data analysis capabilities for different target analysis requirements. Based on this, after obtaining the demand analysis input information, it is necessary to determine the logistics analysis requirements corresponding to the demand analysis input information. If the logistics analysis requirements match the target analysis requirements, it means that the logistics analysis requirements correspond to the target analysis requirements. In this case, the logistics business data can be analyzed using the logistics large language model adapted to handle the target analysis requirements to generate a data analysis result. In this way, performing data analysis on the logistics business data using the logistics large language model adapted to handle the target analysis requirements can generate more accurate data analysis results, improve logistics business data processing efficiency, and further enhance the applicability and accuracy of the data analysis results for different types of analysis requirements.

[0105] 8 , an embodiment of the present application provides a logistics business data analysis device 800, including:

[0106] The acquisition module 801 is used to acquire business analysis instructions and logistics business data; wherein the business analysis instructions are used to reflect the logistics analysis requirements for the logistics business data;

[0107] Input information generation module 802, used to generate demand analysis input information based on business analysis instructions and logistics business data;

[0108] The data analysis module 803 is used to input the demand analysis input information into the logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein, the logistics big language model is pre-trained through multiple logistics training data with data association relationships in the logistics data platform, and the logistics data platform is used to store the logistics training data.

[0109] It can be seen that the contents of the above-mentioned logistics business data analysis method embodiment are all applicable to the embodiment of the present logistics business data analysis device. The functions specifically implemented by the embodiment of the present logistics business data analysis device are the same as those in the above-mentioned logistics business data analysis method embodiment, and the beneficial effects achieved are also the same as those achieved by the above-mentioned logistics business data analysis method embodiment.

[0110] The present application also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-described logistics business data analysis method when executing the computer program. The electronic device can be any smart terminal, including a tablet computer and an in-vehicle computer.

[0111] 9 , which illustrates a hardware structure of an electronic device according to another embodiment, the electronic device includes:

[0112] The processor 901 can be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0113] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called by the processor 901 to execute the logistics business data analysis method of the embodiments of this application.

[0114] Input / output interface 903, used to implement information input and output;

[0115] Communication interface 904, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0116] Bus 905 , which transmits information between various components of the device (e.g., processor 901 , memory 902 , input / output interface 903 , and communication interface 904 );

[0117] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0118] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned logistics business data analysis method.

[0119] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0120] The logistics business data analysis method and its device, electronic device and storage medium provided in the embodiments of the present application capture file change events in the file system and use metadata information to efficiently determine when the data storage space used by each user folder reaches the quota and promptly perform restriction actions on it. In artificial intelligence technology, multiple users of the same file system can fairly use computing resources within the established quota of storage space.

[0121] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0122] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0123] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0124] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0125] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. 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 necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0126] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0128] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0129] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0130] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0131] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A logistics business data analysis method, characterized in that: include: Acquire business analysis instructions and logistics business data; wherein the business analysis instructions are used to reflect logistics analysis requirements for the logistics business data; Generate demand analysis input information according to the business analysis instruction and the logistics business data; The demand analysis input information is input into the logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein the logistics big language model is pre-trained via multiple logistics training data with data association relationships in a logistics data platform, and the logistics data platform is used to store the logistics training data.

2. The method according to claim 1, characterized in that: Before inputting the demand analysis input information into the logistics large language model, the method further includes pre-training the logistics large language model based on the logistics training data, specifically including: Acquire a logistics training data set from the logistics data platform, and configure training annotation information for the logistics training data; wherein the logistics training data set includes a plurality of the logistics training data having the data association relationship; The logistics training data set and the multiple training annotation information are input into the large pre-trained language model for model fine-tuning training to obtain the training output result of the large pre-trained language model for the logistics training data. When the training output result matches the training annotation information, the pre-trained logistics large language model is obtained.

3. The method according to claim 2, characterized in that The obtaining of a logistics training data set from the logistics data platform includes: Extracting a plurality of the logistics training data from the logistics data platform; Performing association analysis on the plurality of logistics training data to obtain the data association relationship; Dividing the first number of the logistics training data into a second number of logistics training data groups according to the data association relationship; wherein the logistics training data in the same logistics training data group have the data association relationship; The second number of the logistics training data sets are integrated to obtain the logistics training data set.

4. The method according to claim 3, characterized in that The step of inputting the logistics training data set and the plurality of training annotation information into the large pre-trained language model for model fine-tuning training, obtaining the training output result of the large pre-trained language model for the logistics training data, and obtaining the pre-trained logistics large language model when the training output result matches the training annotation information, includes: Inputting the logistics training data of the same logistics training data group and the training annotation information corresponding to the logistics training data into the large-scale pre-trained language model for model fine-tuning training; During model fine-tuning training, the training output results of the large pre-trained language model for the same logistics training data set are matched with the corresponding training annotation information to obtain the pre-trained logistics large language model.

5. The method according to claim 4, characterized in that Each of the logistics training data sets corresponds to a candidate analysis requirement; During the model fine-tuning training, the training output result of the large pre-trained language model for the same logistics training data group is matched with the corresponding training annotation information to obtain the pre-trained logistics large language model, including: In the model fine-tuning training, a target logistics training data group is determined from the multiple logistics training data groups; wherein the candidate analysis requirements corresponding to the target logistics training data group are the target analysis requirements; When the training output results of the large pre-trained language model for the target logistics training data group match the corresponding training annotation information, the logistics large language model suitable for processing the target analysis requirements is obtained.

6. The method according to claim 5, characterized in that The step of inputting the demand analysis input information into the logistics big language model so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis demand and generates data analysis results includes: Determining the logistics analysis requirements corresponding to the requirements analysis input information; When the logistics analysis requirement matches the target analysis requirement, the logistics business data is analyzed using the logistics large language model adapted to process the target analysis requirement to generate the data analysis result.

7. The method according to claim 3, characterized in that The performing association analysis on the plurality of logistics training data to obtain the data association relationship includes: Based on a preset logistics business rule, configuring a data correlation coefficient for every two of the logistics training data in the plurality of logistics training data; Integrate the data correlation coefficients between every two of the logistics training data in the plurality of logistics training data to generate a correlation matrix; The data association relationship between the plurality of logistics training data is determined according to the correlation matrix.

8. A logistics business data analysis device, characterized in that: The device comprises: An acquisition module, used to acquire business analysis instructions and logistics business data; wherein the business analysis instructions are used to reflect logistics analysis requirements for the logistics business data; An input information generation module, used to generate demand analysis input information according to the business analysis instruction and the logistics business data; A data analysis module is used to input the demand analysis input information into a logistics big language model, so that the logistics big language model performs data analysis on the logistics business data according to the logistics analysis requirements and generates data analysis results; wherein the logistics big language model is pre-trained via a plurality of logistics training data having data association relationships in a logistics data platform, and the logistics data platform is used to store the logistics training data.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the logistics business data analysis method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the logistics business data analysis method described in any one of claims 1 to 7 is implemented.

Images