Logistics data transmission efficiency optimization method, device and equipment

Abstract

The invention discloses a logistics data transmission efficiency optimization method, device and equipment, and the method is applied to the technical field of logistics, and the method comprises the steps: obtaining the data of a logistics order in order downloading, track return and reset return stages, and carrying out the data cleaning and unified formatting processing of the data, and obtaining the preprocessing data; analyzing and processing the preprocessed data to obtain target attribute data of the logistics order; constructing a data transmission risk prediction model, inputting the target attribute data into the data transmission risk prediction model for data transmission risk prediction processing, and obtaining a risk prediction result of the logistics order; and if the risk prediction result represents that the logistics order has the data transmission risk, dynamically adjusting link parameters of the logistics order. According to the invention, the risk data in the logistics order can be quickly and accurately predicted and the logistics link parameters can be adjusted, so that the data transmission efficiency of the logistics order is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device and equipment for optimizing logistics data transmission efficiency. Background Art

[0002] In the modern e-commerce logistics system, efficient and accurate weight data feedback is crucial for supply chain management. Especially in large e-commerce platforms, the daily order volume is huge, involving complex logistics networks and multi-level warehousing systems. Traditional weight feedback links often face problems such as data processing delays, transmission errors, and uneven distribution of network resources, which will directly affect logistics efficiency and cost control, and even lead to reduced customer satisfaction. In particular, during peak hours or when the network environment is unstable, the instability and inefficiency of data transmission are more significant. Previous weight feedback systems mostly relied on fixed parameter settings and manual monitoring, which made it difficult to adapt to rapidly changing business needs and network conditions. Summary of the invention

[0003] The present invention provides a method, device and equipment for optimizing the efficiency of logistics data transmission, which can quickly and accurately predict the risk data in logistics orders and adjust the logistics link parameters, thereby improving the data transmission efficiency of logistics orders; and significantly improving the intelligent management efficiency of logistics data processing.

[0004] In one aspect, the present invention provides a method for optimizing logistics data transmission efficiency, the method comprising: Obtaining data of logistics orders at the stages of order download, track return, and reset return, and performing data cleaning and unified formatting on the data to obtain pre-processed data; Analyze and process the pre-processed data to obtain order processing time, track update frequency, and weight return delay data as target attribute data of the logistics order; Constructing a data transmission risk prediction model, inputting the target attribute data into the data transmission risk prediction model for data transmission risk prediction processing, and obtaining a risk prediction result of the logistics order; If the risk prediction result indicates that there is a data transmission risk for the logistics order, the link parameters of the logistics order are dynamically adjusted to optimize the data transmission efficiency of the logistics order in real time.

[0005] Optionally, performing data cleaning and unified formatting processing on the data to obtain pre-processed data includes: Calculate the mean, median and standard deviation of each data field; Determine the screening data range based on the difference between the mean and three times the standard deviation and the sum; Deleting the data that exceeds the screening data range in the data to obtain first processed data; Performing uniform formatting processing on the first processed data by using a regular expression to obtain second processed data; Obtain missing values ​​in the second processed data, and replace the missing values ​​with the mean, median or standard deviation.

[0006] Optionally, the method further includes: Use Z-score method to identify and remove outliers in order processing time; Use regular expressions to uniformly set the format of order IDs; The median strategy of SimpleImputer is used to fill the missing values ​​of the trajectory update frequency and convert all timestamp fields to a unified time format.

[0007] Optionally, the pre-processing data is analyzed and processed to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order, including: Analyzing and processing the preprocessed data to obtain a plurality of preprocessed features; Inputting the plurality of preprocessing features into a key feature recognition model to perform feature importance prediction processing to obtain an importance parameter of each preprocessing feature; Sorting the plurality of preprocessing features according to the importance parameter of each preprocessing feature, and selecting key features from the plurality of preprocessing features according to the sorting result; the key features include the order processing time, the track update frequency, and the weight return delay data; The screened key features are arranged in a preset order to form vector features as target attribute data of the logistics order.

[0008] Optionally, the training method of the key feature recognition model includes: Acquire multiple preprocessed training features of training logistics orders; the multiple preprocessed training features are labeled with training key feature labels; Inputting the plurality of preprocessing training features into a recursive feature elimination algorithm model, performing feature importance prediction processing, and obtaining importance parameters of each preprocessing training feature; According to the importance parameter of each preprocessing training feature, the plurality of preprocessing training features are sorted, and according to the sorting result, the feature with the smallest importance parameter is deleted from the plurality of preprocessing features to obtain a screening training feature; If the number of the screening training features meets the preset conditions, the recursive feature elimination algorithm model is trained according to the difference between the screening training features and the training key feature labels to obtain the key feature recognition model.

[0009] Optionally, constructing a data transmission risk prediction model includes: Obtain a sample attribute data set of a sample logistics order; each sample attribute data in the sample attribute data set is annotated with a sample risk result label; dividing the sample attribute data set into a plurality of sample data sets; Generate multiple decision trees based on multiple machine learning models, and build the corresponding relationship between sample data sets and decision trees; According to each sample data set, the decision tree corresponding to each sample data set is trained for data transmission risk prediction to obtain the sample risk prediction result of each sample logistics order; Determining target loss data according to the difference between the sample risk prediction result and the sample risk result label; Training each decision tree according to the target loss data, and constructing the data transmission risk prediction model according to the multiple decision trees at the end of the training; Accordingly, the target attribute data is input into the data transmission risk prediction model for data transmission risk prediction processing to obtain the risk prediction result of the logistics order, including: Inputting the target attribute data into each decision tree in the data transmission risk prediction model to perform data transmission risk prediction processing, and obtaining an initial risk prediction result of each decision tree; According to the initial risk prediction result of each decision tree, the risk prediction result of the logistics order is determined.

[0010] Optionally, the obtaining of sample attribute data of the sample logistics order includes: Obtain sample data of sample logistics orders at the stages of order download, track return, and reset return, and perform data cleaning and unified formatting on the sample data to obtain pre-processed sample data; The pre-processed sample data is analyzed and processed to obtain order processing time, track update frequency and weight return delay data as sample attribute data of the sample logistics order.

[0011] On the other hand, a device for optimizing the efficiency of logistics data transmission is provided, the device comprising: The data acquisition module is used to obtain the data of the logistics order at each stage of order download, track return, and reset return, and to clean and uniformly format the data to obtain pre-processed data; A target data determination module is used to analyze and process the pre-processed data to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order; A model building module, used to build a data transmission risk prediction model, input the target attribute data into the data transmission risk prediction model to perform data transmission risk prediction processing, and obtain the risk prediction result of the logistics order; The parameter adjustment module is used to dynamically adjust the link parameters of the logistics order if the risk prediction result indicates that the logistics order has a data transmission risk, so as to optimize the data transmission efficiency of the logistics order in real time.

[0012] Optionally, the data acquisition module includes: The calculation unit is used to calculate the mean, median and standard deviation of each field corresponding to the data; A screening unit, used for determining a screening data range according to a difference between the mean and three times the standard deviation and a sum; A first processing unit, configured to delete data in the data that exceeds the screening data range to obtain first processed data; A second processing unit, configured to perform uniform formatting processing on the first processed data using a regular expression to obtain second processed data; A value replacement unit is used to obtain missing values ​​in the second processed data and replace the missing values ​​with the mean, median or standard deviation.

[0013] Optionally, the device further comprises: Removal module, which is used to identify and remove outliers in order processing time using the Z-score method; The order format setting module is used to uniformly set the format of the order ID using regular expressions; The time format conversion module is used to fill the missing values ​​of the trajectory update frequency using the median strategy of SimpleImputer and convert all timestamp fields into a unified time format.

[0014] Optionally, the target data determination module includes: An analysis unit, used for analyzing and processing the preprocessing data to obtain a plurality of preprocessing features; An importance parameter determination unit, used for inputting the plurality of preprocessing features into a key feature recognition model for feature importance prediction processing to obtain an importance parameter of each preprocessing feature; A sorting unit, used to sort the plurality of preprocessing features according to an importance parameter of each preprocessing feature, and select key features from the plurality of preprocessing features according to the sorting result; the key features include the order processing time, the track update frequency, and the weight return delay data; The target data determination unit is used to arrange the screened key features in a preset order to form vector features as the target attribute data of the logistics order.

[0015] Optionally, the device further comprises: A training feature acquisition module, used to acquire a plurality of pre-processed training features for training logistics orders; the plurality of pre-processed training features are annotated with training key feature labels; A parameter determination module, used for inputting the plurality of pre-processed training features into a recursive feature elimination algorithm model, performing feature importance prediction processing, and obtaining importance parameters of each pre-processed training feature; A sorting module, used to sort the plurality of preprocessing training features according to the importance parameter of each preprocessing training feature, and according to the sorting result, delete the feature with the smallest importance parameter from the plurality of preprocessing features to obtain the screening training feature; The key model training module is used to train the recursive feature elimination algorithm model according to the difference between the screening training features and the training key feature labels to obtain the key feature recognition model if the number of the screening training features meets the preset conditions.

[0016] Optionally, the model building module includes: A sample data acquisition unit is used to acquire a sample attribute data set of a sample logistics order; each sample attribute data in the sample attribute data set is marked with a sample risk result label; A grouping unit, used for dividing the sample attribute data set into a plurality of sample data sets; A relationship building unit, used to generate multiple decision trees according to multiple machine learning models, and to build a corresponding relationship between a sample data set and the decision trees; A sample result prediction unit is used to perform data transmission risk prediction training on the decision tree corresponding to each sample data set according to each sample data set, so as to obtain a sample risk prediction result for each sample logistics order; a target loss determination unit, configured to determine target loss data according to a difference between the sample risk prediction result and the sample risk result label; A model building unit, used to train each decision tree according to the target loss data, and build the data transmission risk prediction model according to the multiple decision trees at the end of the training; Accordingly, the model building module includes: An initial result prediction unit, used for inputting the target attribute data into each decision tree in the data transmission risk prediction model to perform data transmission risk prediction processing, and obtaining an initial risk prediction result of each decision tree; The risk result determination unit is used to determine the risk prediction result of the logistics order according to the initial risk prediction result of each decision tree.

[0017] Optionally, the sample data acquisition unit includes: The sample data acquisition subunit is used to acquire sample data of sample logistics orders at the stages of order download, track return, and reset return, and to perform data cleaning and unified formatting on the sample data to obtain pre-processed sample data; The sample attribute determination subunit is used to analyze and process the pre-processed sample data to obtain order processing time, track update frequency and weight return delay data as sample attribute data of the sample logistics order.

[0018] On the other hand, an electronic device is provided, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method for optimizing logistics data transmission efficiency as described above.

[0019] On the other hand, a computer storage medium is provided, which stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the method for optimizing logistics data transmission efficiency as described above.

[0020] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes to implement the above-mentioned method for optimizing the efficiency of logistics data transmission.

[0021] The method, device and equipment for optimizing logistics data transmission efficiency provided by the present invention have the following technical effects: The present invention obtains the data of the logistics order at each stage of order download, track return, and reset return, and performs data cleaning and unified formatting processing on the data to obtain pre-processed data; the pre-processed data is analyzed and processed to obtain the order processing time, track update frequency, and weight return delay data as the target attribute data of the logistics order; a data transmission risk prediction model is constructed, and the target attribute data is input into the data transmission risk prediction model for data transmission risk prediction processing to obtain the risk prediction result of the logistics order; if the risk prediction result indicates that the logistics order has a data transmission risk, the link parameters of the logistics order are dynamically adjusted to optimize the data transmission efficiency of the logistics order in real time. The present invention can quickly and accurately predict the risk data in the logistics order and adjust the logistics link parameters, thereby improving the data transmission efficiency of the logistics order; significantly improve the intelligent management efficiency of logistics data processing, and bring significant efficiency improvement and cost optimization to the e-commerce logistics field. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions and advantages of the embodiments of this specification or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This is a schematic diagram of a system for optimizing logistics data transmission efficiency provided by an embodiment of this specification; Figure 2 It is a flow chart of a method for optimizing logistics data transmission efficiency provided by an embodiment of this specification; Figure 3 It is a flowchart of a method for performing data cleaning and uniform formatting on the data to obtain pre-processed data provided by an embodiment of this specification; Figure 4 It is a flowchart of a method provided by an embodiment of this specification for analyzing and processing the pre-processed data to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order; Figure 5 It is a flowchart of a training method for a key feature recognition model provided in an embodiment of this specification; Figure 6 This is a schematic diagram of the structure of a device for optimizing logistics data transmission efficiency provided in an embodiment of this specification; Figure 7 It is a structural diagram of a server provided in an embodiment of this specification. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention 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 server 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.

[0026] See also Figure 1 , Figure 1 is a schematic diagram of a system for optimizing logistics data transmission efficiency provided by an embodiment of this specification, such as Figure 1 As shown, the logistics data transmission efficiency optimization system may include at least a server 01 and a client 02.

[0027] Specifically, in the embodiments of this specification, the server 01 may include an independently operated server, or a distributed server, or a server cluster composed of multiple servers, and may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Server 01 may include a network communication unit, a processor, a memory, and the like. Specifically, the server 01 can be used to obtain data on logistics orders at the stages of order download, track return, and reset return, and determine the risk prediction results of logistics orders, and dynamically adjust the link parameters of the logistics orders.

[0028] Specifically, in the embodiments of this specification, the client 02 may include physical devices such as smart phones, desktop computers, tablet computers, laptops, digital assistants, smart wearable devices, smart speakers, vehicle terminals, smart TVs, etc., and may also include software running in physical devices, such as web pages provided by some service providers to users, or applications provided by these service providers to users. Specifically, the client 02 may be used to display risk prediction results for logistics orders.

[0029] The following describes a method for optimizing logistics data transmission efficiency of the present invention. Figure 2 It is a flowchart of a method for optimizing logistics data transmission efficiency provided in an embodiment of this specification. This specification provides method operation steps as described in the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only order of execution. When the actual system or server product is executed, it can be executed in sequence or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the method shown in the embodiment or the accompanying drawings. Specifically, Figure 2 As shown, the method may include: S201: Obtain data of logistics orders at various stages including order download, track return, and reset return, and perform data cleaning and unified formatting on the data to obtain pre-processed data. In the embodiments of this specification, Figure 3 As shown, the data is cleaned and uniformly formatted to obtain pre-processed data, including: S20101: Calculate the mean, median and standard deviation of each data field; S20103: Determine the screening data range according to the difference between the mean and three times the standard deviation and the sum; S20105: deleting the data that exceeds the screening data range in the data to obtain first processed data; S20107: using a regular expression to uniformly format the first processed data to obtain second processed data; S20109: Obtain missing values ​​in the second processed data, and replace the missing values ​​with the mean, median or standard deviation.

[0030] In the embodiments of this specification, for the collected data, the mean, median and standard deviation of each field are calculated. For each data point, if its value exceeds the range of the mean plus or minus three times the standard deviation, it is considered an outlier and is removed. Regular expressions are used to define the format rules of the data, such as the date format is "YYYY-MM-DD", the number format is a number with a specific decimal place, etc. For missing values, if the missing value ratio of a field is low, the mean, median of the field can be used or filled through other related fields. For data type conversion, for example, a number of a string type is converted to a numeric type. The second processed data after the missing value is replaced can be used as preprocessed data.

[0031] In the embodiment of this specification, the method further includes: Use Z-score method to identify and remove outliers in order processing time; Use regular expressions to uniformly set the format of order IDs; The median strategy of SimpleImputer is used to fill the missing values ​​of the trajectory update frequency and convert all timestamp fields to a unified time format.

[0032] In the embodiments of this specification, the Z-score method can be applied to identify and remove outliers in order processing time. Regular expressions are used to unify the format of order IDs. SimpleImputer's median strategy is used to fill in missing values ​​of track update frequency. All timestamp fields are converted to a unified time format.

[0033] S203: Analyze and process the pre-processed data to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order.

[0034] In the embodiments of this specification, Figure 4 As shown, the pre-processed data is analyzed and processed to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order, including: S2031: Analyze and process the preprocessed data to obtain a plurality of preprocessed features; S2033: Inputting the plurality of preprocessing features into a key feature recognition model to perform feature importance prediction processing to obtain importance parameters of each preprocessing feature; S2035: sorting the plurality of preprocessing features according to the importance parameter of each preprocessing feature, and selecting key features from the plurality of preprocessing features according to the sorting result; the key features include the order processing time, the track update frequency, and the weight return delay data; S2037: Arrange the selected key features in a preset order to form vector features as target attribute data of the logistics order.

[0035] In the embodiments of this specification, the multiple pre-processing features can be input into the key feature recognition model for feature importance prediction processing to obtain the importance parameter of each pre-processing feature; a feature selection algorithm (such as recursive feature elimination, feature selection based on correlation) can be used to determine the variable set with the most predictive value and construct an efficient feature vector. When constructing the feature vector, the selected key features are arranged in a fixed order to form a vector, thereby obtaining the target attribute data of the logistics order.

[0036] In the embodiments of this specification, Figure 5 As shown, the training method of the key feature recognition model includes: S501: Acquire multiple preprocessed training features of a training logistics order; the multiple preprocessed training features are labeled with training key feature labels; S503: Inputting the plurality of preprocessed training features into a recursive feature elimination algorithm model, performing feature importance prediction processing, and obtaining importance parameters of each preprocessed training feature; S505: sorting the plurality of preprocessing training features according to the importance parameter of each preprocessing training feature, and deleting the feature with the smallest importance parameter from the plurality of preprocessing features according to the sorting result to obtain a screening training feature; S507: If the number of the screening training features meets the preset conditions, the recursive feature elimination algorithm model is trained according to the difference between the screening training features and the training key feature labels to obtain the key feature recognition model.

[0037] In an embodiment of the present specification, if the number of the screening training features does not meet the preset conditions, the screening training features are re-input into the recursive feature elimination algorithm model for repeated training. For the recursive feature elimination algorithm, first use all the features to train an initial model and sort the features according to their importance. Then gradually delete the least important features and retrain the model until the preset number of features is reached or the model performance is no longer improved. The correlation-based feature selection algorithm calculates the correlation between each feature and the target variable and selects features with higher correlation. Training loss data can be constructed based on the difference between the screening training features and the training key feature labels, and the parameters of the recursive feature elimination algorithm model can be adjusted according to the training loss data until the model converges to obtain the key feature recognition model.

[0038] S205: Construct a data transmission risk prediction model, input the target attribute data into the data transmission risk prediction model to perform data transmission risk prediction processing, and obtain a risk prediction result for the logistics order.

[0039] In the embodiment of this specification, the construction of the data transmission risk prediction model includes: Obtain a sample attribute data set of a sample logistics order; each sample attribute data in the sample attribute data set is annotated with a sample risk result label; dividing the sample attribute data set into a plurality of sample data sets; Generate multiple decision trees based on multiple machine learning models, and build the corresponding relationship between sample data sets and decision trees; According to each sample data set, the decision tree corresponding to each sample data set is trained for data transmission risk prediction to obtain the sample risk prediction result of each sample logistics order; Determining target loss data according to the difference between the sample risk prediction result and the sample risk result label; Each decision tree is trained according to the target loss data, and the data transmission risk prediction model is constructed according to the multiple decision trees at the end of the training.

[0040] In the embodiments of this specification, a machine learning model (such as a random forest, a support vector machine, or a deep neural network) is used to train historical data to learn the behavior patterns of data streams. The model needs to be able to predict future data processing time and potential transmission bottlenecks; for the random forest model, multiple decision trees are randomly generated. Each decision tree is trained on a random subset of the training data and split based on randomly selected features. The final prediction result is obtained by averaging or voting the prediction results of multiple decision trees. The support vector machine divides the data by finding an optimal hyperplane so that the distance from data points of different categories to the hyperplane is maximized. The deep neural network consists of an input layer, a hidden layer, and an output layer, and learns by adjusting the connection weights between neurons through a back-propagation algorithm. These models are trained using historical data, and the model parameters are continuously adjusted to minimize the prediction error.

[0041] In the embodiment of this specification, the inputting of the target attribute data into the data transmission risk prediction model to perform data transmission risk prediction processing to obtain the risk prediction result of the logistics order includes: Inputting the target attribute data into each decision tree in the data transmission risk prediction model to perform data transmission risk prediction processing, and obtaining an initial risk prediction result of each decision tree; According to the initial risk prediction result of each decision tree, the risk prediction result of the logistics order is determined.

[0042] In the embodiment of the present specification, after determining the initial risk prediction result of each decision tree, the average value of the initial risk prediction result of each decision tree can be calculated as the risk prediction result of the logistics order. Voting can also be performed based on the initial risk prediction result of each decision tree to determine the risk prediction result of the logistics order.

[0043] In the embodiment of this specification, the step of obtaining sample attribute data of a sample logistics order includes: Obtain sample data of sample logistics orders at the stages of order download, track return, and reset return, and perform data cleaning and unified formatting on the sample data to obtain pre-processed sample data; The pre-processed sample data is analyzed and processed to obtain order processing time, track update frequency and weight return delay data as sample attribute data of the sample logistics order.

[0044] S207: If the risk prediction result indicates that there is a data transmission risk for the logistics order, dynamically adjust the link parameters of the logistics order to optimize the data transmission efficiency of the logistics order in real time.

[0045] In the embodiments of this specification, link parameters can be dynamically adjusted according to the risk prediction results, such as adjusting the packet size, optimizing network routing, scheduling resources, etc., to optimize data transmission efficiency in real time. Continuously obtain network status and data transmission conditions. When the prediction model output may have a risk of congestion or delay, adjust the link parameters according to the preset strategy. For example, if network congestion is predicted, increase the packet size to reduce the number of packets, or recalculate the network route using the shortest path algorithm. For resource scheduling, set priorities according to the importance and urgency of the data to ensure that critical data is transmitted first. The developed responsive algorithm framework is able to receive prediction results in real time and send optimization instructions to the corresponding network devices and systems for execution.

[0046] Monitor the model prediction results and trigger resource adjustment strategies when network congestion or extended processing time is predicted. Implement a dynamic packet size adjustment algorithm to adaptively adjust the TCP window size based on the predicted network conditions. Use SDN (software defined network) technology to dynamically modify network routes or call the cloud service provider's route optimization service through API. Develop a resource scheduling module and use priority queues and load balancing algorithms to ensure sufficient resources for critical data transmission, such as giving priority to the return of weight information during peak hours.

[0047] This technical solution provides a settlement weight return link method based on intelligent dynamic optimization. It cleans and formats the data collected from the order download, trajectory return, and reset return stages, removes outliers and redundant information, extracts key features from the pre-processed data to evaluate the data transmission efficiency and possible delay risks under different conditions, and dynamically adjusts link parameters to optimize data transmission efficiency in real time. It significantly improves the level of intelligence in logistics data processing, and brings significant efficiency improvements and cost optimization to the e-commerce logistics field.

[0048] The technical solution of this application may include the following steps: Step 1: Clean and format the data collected from order download, trajectory return, and reset return to remove outliers and redundant information; For the collected data, calculate the mean, median, and standard deviation of each field. For each data point, if its value exceeds the range of the mean plus or minus three times the standard deviation, it is considered an outlier and removed. Use regular expressions to define the format rules of the data, such as the date format is "YYYY-MM-DD", the number format is a number with a specific decimal place, etc. For missing values, if the proportion of missing values ​​in a field is low, you can use the mean, median of the field or fill it through other related fields. For data type conversion, for example, convert a string type number to a numeric type.

[0049] Apply the Z-score method to identify and remove outliers in order processing time. Use regular expressions to unify the format of order IDs. Use SimpleImputer's median strategy to fill in missing values ​​of trajectory update frequency. Convert all timestamp fields to a unified time format.

[0050] Step 2: Extract key features from the preprocessed data, such as order processing time, trajectory update frequency, weight return delay, etc., as algorithm input; Feature selection algorithms (such as recursive feature elimination and correlation-based feature selection) are used to determine the set of variables with the most predictive value and construct an efficient feature vector. For the recursive feature elimination algorithm, an initial model is first trained using all features and the features are sorted according to their importance. Then the least important features are gradually removed and the model is retrained until the preset number of features is reached or the model performance is no longer improved. The correlation-based feature selection algorithm calculates the correlation between each feature and the target variable and selects the features with higher correlation. When constructing the feature vector, the selected key features are arranged in a fixed order to form a vector.

[0051] Step 3: Establish a prediction model to evaluate the data transmission efficiency and possible delay risks under different conditions; Machine learning models (such as random forests, support vector machines, or deep neural networks) are trained on historical data to learn the behavior patterns of data streams. The model needs to be able to predict future data processing time and potential transmission bottlenecks; for random forest models, multiple decision trees are randomly generated. Each decision tree is trained on a random subset of the training data and split based on randomly selected features. The final prediction result is obtained by averaging or voting the prediction results of multiple decision trees. Support vector machines divide data by finding an optimal hyperplane that maximizes the distance from data points of different categories to the hyperplane. Deep neural networks consist of input layers, hidden layers, and output layers, and learn by adjusting the connection weights between neurons through the back-propagation algorithm. These models are trained using historical data, and the model parameters are continuously adjusted to minimize the prediction error.

[0052] Step 4: According to the prediction results, dynamically adjust link parameters, such as adjusting data packet size, optimizing network routing, scheduling resources, etc., to optimize data transmission efficiency in real time.

[0053] Continuously obtain network status and data transmission. When the prediction model output may have the risk of congestion or delay, adjust the link parameters according to the preset strategy. For example, if network congestion is predicted, increase the packet size to reduce the number of packets, or recalculate the network route using the shortest path algorithm. For resource scheduling, set priorities based on the importance and urgency of the data to ensure that critical data is transmitted first. The developed responsive algorithm framework can receive the prediction results in real time and send optimization instructions to the corresponding network devices and systems for execution.

[0054] Monitor the model prediction results and trigger resource adjustment strategies when network congestion or extended processing time is predicted. Implement a dynamic packet size adjustment algorithm to adaptively adjust the TCP window size based on the predicted network conditions. Use SDN (software defined network) technology to dynamically modify network routes or call the cloud service provider's route optimization service through API. Develop a resource scheduling module and use priority queues and load balancing algorithms to ensure sufficient resources for critical data transmission, such as giving priority to the return of weight information during peak hours.

[0055] It can be seen from the technical solutions provided by the above embodiments of this specification that the embodiments of this specification obtain the data of the logistics order at each stage of order download, track return, and reset return, and perform data cleaning and unified formatting on the data to obtain pre-processed data; analyze and process the pre-processed data to obtain order processing time, track update frequency, and weight return delay data as the target attribute data of the logistics order; construct a data transmission risk prediction model, input the target attribute data into the data transmission risk prediction model for data transmission risk prediction processing, and obtain the risk prediction result of the logistics order; if the risk prediction result indicates that the logistics order has a data transmission risk, dynamically adjust the link parameters of the logistics order to optimize the data transmission efficiency of the logistics order in real time. The present invention can quickly and accurately predict the risk data in the logistics order and adjust the logistics link parameters, thereby improving the data transmission efficiency of the logistics order; significantly improve the intelligent management efficiency of logistics data processing, and bring significant efficiency improvement and cost optimization to the e-commerce logistics field.

[0056] The embodiment of this specification also provides a device for optimizing the efficiency of logistics data transmission, such as Figure 6 As shown, the device comprises: The data acquisition module 610 is used to acquire the data of the logistics order at each stage of order download, track return, and reset return, and perform data cleaning and unified formatting on the data to obtain pre-processed data; The target data determination module 620 is used to analyze and process the pre-processed data to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order; The model building module 630 is used to build a data transmission risk prediction model, input the target attribute data into the data transmission risk prediction model to perform data transmission risk prediction processing, and obtain the risk prediction result of the logistics order; The parameter adjustment module 640 is used to dynamically adjust the link parameters of the logistics order if the risk prediction result indicates that the logistics order has a data transmission risk, so as to optimize the data transmission efficiency of the logistics order in real time.

[0057] Optionally, the data acquisition module includes: The calculation unit is used to calculate the mean, median and standard deviation of each field corresponding to the data; A screening unit, used for determining a screening data range according to a difference between the mean and three times the standard deviation and a sum; A first processing unit, configured to delete data in the data that exceeds the screening data range to obtain first processed data; A second processing unit, configured to perform uniform formatting processing on the first processed data using a regular expression to obtain second processed data; A value replacement unit is used to obtain missing values ​​in the second processed data and replace the missing values ​​with the mean, median or standard deviation.

[0058] Optionally, the device further comprises: Removal module, which is used to identify and remove outliers in order processing time using the Z-score method; The order format setting module is used to uniformly set the format of the order ID using regular expressions; The time format conversion module is used to fill the missing values ​​of the trajectory update frequency using the median strategy of SimpleImputer and convert all timestamp fields into a unified time format.

[0059] Optionally, the target data determination module includes: An analysis unit, used for analyzing and processing the preprocessing data to obtain a plurality of preprocessing features; An importance parameter determination unit, used for inputting the plurality of preprocessing features into a key feature recognition model for feature importance prediction processing to obtain an importance parameter of each preprocessing feature; A sorting unit, used to sort the plurality of preprocessing features according to an importance parameter of each preprocessing feature, and select key features from the plurality of preprocessing features according to the sorting result; the key features include the order processing time, the track update frequency, and the weight return delay data; The target data determination unit is used to arrange the screened key features in a preset order to form vector features as the target attribute data of the logistics order.

[0060] Optionally, the device further comprises: A training feature acquisition module, used to acquire a plurality of pre-processed training features for training logistics orders; the plurality of pre-processed training features are annotated with training key feature labels; A parameter determination module, used for inputting the plurality of pre-processed training features into a recursive feature elimination algorithm model, performing feature importance prediction processing, and obtaining importance parameters of each pre-processed training feature; A sorting module, used to sort the plurality of preprocessing training features according to the importance parameter of each preprocessing training feature, and according to the sorting result, delete the feature with the smallest importance parameter from the plurality of preprocessing features to obtain the screening training feature; The key model training module is used to train the recursive feature elimination algorithm model according to the difference between the screening training features and the training key feature labels to obtain the key feature recognition model if the number of the screening training features meets the preset conditions.

[0061] Optionally, the model building module includes: A sample data acquisition unit is used to acquire a sample attribute data set of a sample logistics order; each sample attribute data in the sample attribute data set is marked with a sample risk result label; A grouping unit, used for dividing the sample attribute data set into a plurality of sample data sets; A relationship building unit, used to generate multiple decision trees according to multiple machine learning models, and to build a corresponding relationship between a sample data set and the decision trees; A sample result prediction unit is used to perform data transmission risk prediction training on the decision tree corresponding to each sample data set according to each sample data set, so as to obtain a sample risk prediction result for each sample logistics order; a target loss determination unit, configured to determine target loss data according to a difference between the sample risk prediction result and the sample risk result label; A model building unit, used to train each decision tree according to the target loss data, and build the data transmission risk prediction model according to the multiple decision trees at the end of the training; Accordingly, the model building module includes: An initial result prediction unit, used for inputting the target attribute data into each decision tree in the data transmission risk prediction model to perform data transmission risk prediction processing, and obtaining an initial risk prediction result of each decision tree; The risk result determination unit is used to determine the risk prediction result of the logistics order according to the initial risk prediction result of each decision tree.

[0062] Optionally, the sample data acquisition unit includes: The sample data acquisition subunit is used to acquire sample data of sample logistics orders at the stages of order download, track return, and reset return, and to perform data cleaning and unified formatting on the sample data to obtain pre-processed sample data; The sample attribute determination subunit is used to analyze and process the pre-processed sample data to obtain order processing time, track update frequency and weight return delay data as sample attribute data of the sample logistics order.

[0063] The device and method embodiments in the described device embodiments are based on the same inventive concept.

[0064] An embodiment of the present specification provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement a method for optimizing logistics data transmission efficiency as provided in the above method embodiment.

[0065] An embodiment of the present invention also provides a computer storage medium, which can be set in a terminal to store at least one instruction or at least one program related to a method for optimizing logistics data transmission efficiency in a method embodiment. The at least one instruction or at least one program is loaded and executed by the processor to implement the method for optimizing logistics data transmission efficiency provided in the above method embodiment.

[0066] The embodiment of the present invention also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method for optimizing the logistics data transmission efficiency provided by the above method embodiment.

[0067] Optionally, in the embodiments of this specification, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0068] The memory described in the embodiments of this specification can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, application programs required for functions, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0069] The method for optimizing the efficiency of logistics data transmission provided in the embodiments of this specification can be executed in a mobile terminal, a computer terminal, a server or a similar computing device. Taking running on a server as an example, Figure 71 is a hardware structure diagram of a server of a method for optimizing logistics data transmission efficiency provided in an embodiment of this specification. Figure 7 As shown, the server 700 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 710 (the central processing unit 710 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 730 for storing data, and one or more storage media 720 (such as one or more mass storage devices) for storing application programs 723 or data 722. Among them, the memory 730 and the storage medium 720 can be short-term storage or permanent storage. The program stored in the storage medium 720 may include one or more modules, each of which may include a series of instruction operations on the server. Furthermore, the central processing unit 710 can be configured to communicate with the storage medium 720 and execute a series of instruction operations in the storage medium 720 on the server 700. The server 700 may also include one or more power supplies 760, one or more wired or wireless network interfaces 750, one or more input and output interfaces 740, and / or one or more operating systems 721, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0070] The input / output interface 740 may be used to receive or send data via a network. A specific example of the network may include a wireless network provided by a communication provider of the server 700. In one example, the input / output interface 740 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices via a base station so as to communicate with the Internet. In one example, the input / output interface 740 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0071] It can be understood by those skilled in the art that Figure 7 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 7 More or fewer components as shown, or with Figure 7 Different configurations are shown.

[0072] It can be seen from the embodiments of the optimization method, device, electronic device or storage medium for logistics data transmission efficiency provided by the present invention that the present invention obtains data of logistics orders at each stage of order download, track return, and reset return, and performs data cleaning and unified formatting on the data to obtain pre-processed data; the pre-processed data is analyzed and processed to obtain order processing time, track update frequency, and weight return delay data as target attribute data of the logistics order; a data transmission risk prediction model is constructed, and the target attribute data is input into the data transmission risk prediction model for data transmission risk prediction processing to obtain the risk prediction result of the logistics order; if the risk prediction result indicates that the logistics order has a data transmission risk, the link parameters of the logistics order are dynamically adjusted to optimize the data transmission efficiency of the logistics order in real time. The present invention can quickly and accurately predict the risk data in the logistics order and adjust the logistics link parameters, thereby improving the data transmission efficiency of the logistics order; significantly improve the intelligent management efficiency of logistics data processing, and bring significant efficiency improvement and cost optimization to the field of e-commerce logistics.

[0073] It should be noted that the above sequence of the embodiments of this specification is for description only and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0075] A person skilled in the art will appreciate that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0076] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing logistics data transmission efficiency, characterized in that: The method comprises: Obtaining data of logistics orders at the stages of order download, track return, and reset return, and performing data cleaning and unified formatting on the data to obtain pre-processed data; Analyze and process the pre-processed data to obtain order processing time, track update frequency, and weight return delay data as target attribute data of the logistics order; Constructing a data transmission risk prediction model, inputting the target attribute data into the data transmission risk prediction model for data transmission risk prediction processing, and obtaining a risk prediction result of the logistics order; If the risk prediction result indicates that there is a data transmission risk for the logistics order, the link parameters of the logistics order are dynamically adjusted to optimize the data transmission efficiency of the logistics order in real time.

2. The method according to claim 1, characterized in that The data is cleaned and uniformly formatted to obtain pre-processed data, including: Calculate the mean, median and standard deviation of each data field; Determine the screening data range based on the difference between the mean and three times the standard deviation and the sum; Deleting the data that exceeds the screening data range in the data to obtain first processed data; Performing uniform formatting processing on the first processed data by using a regular expression to obtain second processed data; Obtain missing values ​​in the second processed data, and replace the missing values ​​with the mean, median or standard deviation.

3. The method according to claim 2, characterized in that The method further comprises: Use Z-score method to identify and remove outliers in order processing time; Use regular expressions to uniformly set the format of order IDs; The median strategy of SimpleImputer is used to fill the missing values ​​of the trajectory update frequency and convert all timestamp fields to a unified time format.

4. The method according to claim 1, characterized in that: The pre-processing data is analyzed and processed to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order, including: Analyzing and processing the preprocessed data to obtain a plurality of preprocessed features; Inputting the plurality of preprocessing features into a key feature recognition model to perform feature importance prediction processing to obtain an importance parameter of each preprocessing feature; Sorting the plurality of preprocessing features according to the importance parameter of each preprocessing feature, and selecting key features from the plurality of preprocessing features according to the sorting result; the key features include the order processing time, the track update frequency, and the weight return delay data; The screened key features are arranged in a preset order to form vector features as target attribute data of the logistics order.

5. The method according to claim 1, characterized in that The training method of the key feature recognition model includes: Acquire multiple preprocessed training features of training logistics orders; the multiple preprocessed training features are labeled with training key feature labels; Inputting the plurality of preprocessing training features into a recursive feature elimination algorithm model, performing feature importance prediction processing, and obtaining importance parameters of each preprocessing training feature; According to the importance parameter of each preprocessing training feature, the plurality of preprocessing training features are sorted, and according to the sorting result, the feature with the smallest importance parameter is deleted from the plurality of preprocessing features to obtain a screening training feature; If the number of the screening training features meets the preset conditions, the recursive feature elimination algorithm model is trained according to the difference between the screening training features and the training key feature labels to obtain the key feature recognition model.

6. The method according to claim 1, characterized in that The constructing of the data transmission risk prediction model comprises: Obtain a sample attribute data set of a sample logistics order; each sample attribute data in the sample attribute data set is annotated with a sample risk result label; dividing the sample attribute data set into a plurality of sample data sets; Generate multiple decision trees based on multiple machine learning models, and build the corresponding relationship between sample data sets and decision trees; According to each sample data set, the decision tree corresponding to each sample data set is trained for data transmission risk prediction to obtain the sample risk prediction result of each sample logistics order; Determining target loss data according to the difference between the sample risk prediction result and the sample risk result label; Training each decision tree according to the target loss data, and constructing the data transmission risk prediction model according to the multiple decision trees at the end of the training; Accordingly, the target attribute data is input into the data transmission risk prediction model for data transmission risk prediction processing to obtain the risk prediction result of the logistics order, including: Inputting the target attribute data into each decision tree in the data transmission risk prediction model to perform data transmission risk prediction processing, and obtaining an initial risk prediction result of each decision tree; According to the initial risk prediction result of each decision tree, the risk prediction result of the logistics order is determined.

7. The method according to claim 6, characterized in that The obtaining of sample attribute data of the sample logistics order includes: Obtain sample data of sample logistics orders at the stages of order download, track return, and reset return, and perform data cleaning and unified formatting on the sample data to obtain pre-processed sample data; The pre-processed sample data is analyzed and processed to obtain order processing time, track update frequency and weight return delay data as sample attribute data of the sample logistics order.

8. A device for optimizing logistics data transmission efficiency, characterized in that: The device comprises: The data acquisition module is used to obtain the data of the logistics order at each stage of order download, track return, and reset return, and to clean and uniformly format the data to obtain pre-processed data; A target data determination module is used to analyze and process the pre-processed data to obtain order processing time, track update frequency and weight return delay data as target attribute data of the logistics order; A model building module, used to build a data transmission risk prediction model, input the target attribute data into the data transmission risk prediction model to perform data transmission risk prediction processing, and obtain the risk prediction result of the logistics order; The parameter adjustment module is used to dynamically adjust the link parameters of the logistics order if the risk prediction result indicates that the logistics order has a data transmission risk, so as to optimize the data transmission efficiency of the logistics order in real time.

9. An electronic device, characterized in that: The device includes: a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the method for optimizing logistics data transmission efficiency as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the method for optimizing logistics data transmission efficiency as described in any one of claims 1-7.

Images