Intelligent information release and management system for port entering and leaving service of aviation cargo station

Abstract

The invention relates to the technical field of aviation cargo station order data processing, and discloses an intelligent information release and management system for port entry and exit business of an aviation cargo station, which comprises the following steps of: firstly, collecting aviation cargo order data from each aviation business system including cargo processing, cargo check-in, cargo sorting, cargo reconfirmation and a waybill management system; features related to order recommendation and matching are extracted from the integrated data, the features comprise cargo types, weights, volumes, destinations, flight time and flight capacities, and features having great influences on recommendation and matching results are screened out by using a feature selection algorithm; an order recommendation model is constructed based on a machine learning algorithm, a flight matching model is constructed by using a classification algorithm, the model is trained and optimized, and the recommendation and matching accuracy and efficiency are improved; and outputting recommendation and matching results. The method can integrate and analyze port entry and exit business data, provides logistics distribution decision support for the aviation cargo station, and improves the resource utilization rate.

Description

Technical Field

[0001] The present invention relates to the technical field of air cargo terminal order data processing, in particular to an intelligent information publishing and management system for air cargo terminal inbound and outbound business. Background Art

[0002] There are many methods for order matching in aviation logistics, including rule-based matching, customer demand-based matching, cargo characteristics-based matching, and intelligent matching. These methods have their own advantages and can be flexibly selected and applied according to actual conditions. In the field of aviation logistics, the intelligent matching method that uses machine learning algorithms and big data technology to match orders and optimize logistics networks has brought many conveniences and efficiency improvements, but it still has shortcomings. At present, there is an information island phenomenon in the aviation logistics platform. Since the business systems are isolated from each other, information cannot flow freely between different systems, forming an information island. This makes it difficult for the aviation logistics platform to obtain comprehensive cargo information or order information, affecting logistics efficiency and accuracy.

[0003] To realize the release and sharing of various air logistics orders and achieve efficient matching of cargo and flight recommendations.

[0004] Therefore, in order to solve the above problems, an intelligent information release and management system for the air cargo terminal's inbound and outbound business is needed. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent information release and management system for the inbound and outbound business of an air cargo terminal. The system of the present invention can integrate and analyze the inbound and outbound business data to provide logistics allocation decision support for the air cargo terminal. Through data analysis, the logistics process can be optimized and resource utilization can be improved.

[0006] The present invention is achieved in that: The present invention provides an intelligent information publishing and management method for air cargo terminal inbound and outbound services, which is specifically performed in the following steps: S1: Collect air cargo order data from various aviation business systems including cargo handling, cargo check-in, cargo sorting, cargo reconfirmation, and waybill management systems; The collected data is cleaned, deduplicated, and format converted for preprocessing, and the processed data is imported into the data center; S2: Govern data in the data center, including data standard formulation, data quality monitoring, and structured processing of unstructured data including cargo images, videos, laser, RFID, and OCR recognition; In step S2, the following steps are specifically performed: S2.1: First, determine the data source, identify and list all relevant business systems, including data from the cargo handling system, cargo check-in system, cargo sorting system, cargo reconfirmation system, and waybill management system; S2.2: Extract data. Write data extraction scripts or configure data extraction tools according to data interfaces and data formats; extract data from various business systems regularly or in real time; S2.3: Perform data cleaning, identify and handle missing values, and fill missing data using the mean, median, mode, interpolation, or deletion; S2.4: Identify and handle outliers; remove duplicate data, using unique identifiers or business rules to identify and delete duplicate records; S2.5: Perform data format conversion to convert data in different formats into a unified format; for example, convert data in CSV, Excel, JSON and other formats into structured data in a database table or data warehouse. Or encode or decode data to ensure data consistency and readability.

[0007] S2.6: Perform data integration, import the cleaned and converted data into the data center, establish a data model in the data center, define the relationship and hierarchy between the data, and use ETL tools or custom scripts to integrate the data to ensure the accuracy and consistency of the data.

[0008] S3: Extract features related to order recommendation and matching from the integrated data, including cargo type, weight, volume, destination, flight time, and flight capacity, and use feature selection algorithms to select features that have a great impact on recommendation and matching results; Specifically, the variance selection method is used to screen out features that have a great impact on recommendation and matching results; S3.1: First, calculate the variance of each feature cargo type, weight, volume, destination, flight time, and flight capacity, set a variance threshold, and select features whose variance is greater than the threshold; as follows: in, is the i-th value of feature x, is the mean of feature x, n represents the number of samples; S3.2: Select features by calculating the correlation coefficient between the features and the target variable. The features with larger correlation coefficients have stronger correlation with the target variable and may have greater impact on the model. S3.3: Specifically, by calculating the correlation coefficient between each feature and the target variable, a correlation coefficient threshold is set, and the features with correlation coefficients greater than the threshold are selected, as shown in the following formula; in, and Represent the i-th value of feature X and target variable y respectively, and They represent the mean of feature x and variable y respectively, and n represents the number of samples; S3.4: Combine the feature values ​​calculated in steps S3.2 and S3.3, comprehensively evaluate the impact of feature subsets on recommendation and matching results, and select the optimal feature subset.

[0009] Please follow the steps below: S4.1: For each feature subset, use k-fold cross validation to evaluate the performance of the model on different data subsets. Based on the results of cross validation, select the feature subset with the best performance as the final feature set; S4.2: Divide each feature subset into k subsets of equal size, initialize the performance evaluation index, and initialize a list for storing cross-validation results for each feature subset; Perform k iterations; in each iteration, select a subset as the validation set and the remaining k-1 subsets as the training set; use the training set data to train the model; Use the validation set data to evaluate the model performance and calculate the evaluation indicators including accuracy, recall, and F1 score; store the evaluation results in a list of the corresponding feature subset; S4.3: Calculate the average performance. For each feature subset, calculate its average performance in k iterations; S4.4: According to the average performance, select the feature subset with the best performance, determine the final feature set, and use the selected optimal feature subset as the final feature set.

[0010] S4: Build an order recommendation model based on machine learning algorithms, build a flight matching model using classification algorithms, train and optimize the model to improve the accuracy and efficiency of recommendations and matching; Among them, the construction of order recommendation model; First, the feature value data set is divided into a training set, a validation set, and a test set from the feature values ​​that have a great impact on the calculated matching results, including cargo type, weight, volume, destination, flight time, and flight capacity; Train the model, use the training set data to train the model, and adjust the model parameters to minimize the loss function; Use validation set data to evaluate model performance and calculate evaluation indicators including accuracy, recall, F1 score, and AUC-ROC; Re-optimize the model, adjust model parameters according to the evaluation results, add new features, and use regularization techniques to optimize the model; Flight matching model construction; First, flight data is collected from the flight information system, including flight number, departure time, arrival time, flight capacity, and flight status; Regenerate features related to flight matching, including flight time windows, the difference between departure and arrival times, and flight remaining capacity data; Using the logistic regression algorithm, the data set is divided into a training set, a validation set, and a test set; the model is trained using the training set data, and the model parameters are adjusted to maximize the classification accuracy or AUC-ROC indicator; Use validation set data to evaluate model performance and calculate classification accuracy and AUC-ROC indicators; deploy the trained order recommendation model and flight matching model into the aviation logistics system; Based on the order information entered by the user, the order recommendation model is used to generate recommendation results. Based on the recommendation results and flight information, the flight matching model is used to generate matching results. The recommendations and matching results are displayed to the user for reference and decision-making.

[0011] S5: Based on the order information input by the user, including the type, weight, volume, and destination of goods, the recommendation model is used to recommend suitable air logistics orders for the user; based on the recommendation results and flight information, including flight time and flight capacity, the matching model is used to match the most suitable flight for the user; the recommendation and matching results are output for the user's reference and decision-making.

[0012] Further, the present invention provides an intelligent information release management system for air cargo terminal inbound and outbound business, including a user input module for receiving order information input by a user, including cargo type, weight, volume, and destination; receiving user input via a Web interface, a mobile application, or an API interface; and outputting the order information input by the user to an order recommendation module; The order recommendation module uses the recommendation model to recommend suitable air logistics orders to users based on the order information input by users. It loads the pre-trained recommendation model, pre-processes the order information input by users, such as standardization and encoding, and inputs the pre-processed order information into the recommendation model to obtain the recommendation results. Obtain and store flight information, including flight time and flight capacity, obtain flight information from airlines or third-party data sources, store flight information, and pass flight information to the flight matching module; The flight matching module uses the matching model to match the most suitable flight for the user based on the recommendation results and flight information, loads the pre-trained matching model, and performs matching calculations based on the recommendation results and flight information; outputs the matching results, including the recommended flight, flight time, and estimated arrival time; and passes the matching results to the result output module; The result output module outputs the recommendation and matching results to the user for reference and decision-making. It displays the recommendation and matching results through the web interface, mobile application or API interface, and provides user interaction functions, including selecting, confirming or modifying the recommendation and matching results.

[0013] Furthermore, the present invention provides a computer-storable medium, wherein the computer-readable storage medium includes an embedded processing system and a stored program, and when the embedded system control program runs, the computer-storable storage medium controls the execution of any one of the above-mentioned intelligent information publishing and management methods for the inbound and outbound business of an air cargo terminal.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. Through the intelligent information release system, the air cargo terminal can realize the automated processing and intelligent management of inbound and outbound business. The system can automatically collect, analyze and release cargo information, reduce manual intervention and improve logistics efficiency.

[0015] Real-time update and monitoring, the system can update the status of goods in real time, including the location of goods, transportation progress, etc., to ensure the accuracy and timeliness of information. At the same time, through the monitoring function, problems in the logistics process can be discovered and solved in time to avoid delays and errors.

[0016] 2. The system can integrate and analyze inbound and outbound business data to provide decision support for air cargo terminals. Through data analysis, logistics processes can be optimized and resource utilization can be improved.

[0017] Intelligent scheduling: Based on real-time data and algorithms, the system can realize intelligent scheduling, reasonably arrange the time of goods entering and leaving the port, and improve logistics efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It is understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a system structure diagram of the present invention; Figure 3 is the feature value extraction code of the present invention Figure 1 ; Figure 4 is the feature value extraction code of the present invention Figure 2 ; Figure 5 It is the order and flight recommendation code of the present invention Figure 1 ; Figure 6 It is the order and flight recommendation code of the present invention Figure 2 . DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 belong to the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but is only for selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0021] See also Figure 1-Figure 6 The present invention provides an intelligent information release and management system for air cargo terminal inbound and outbound services, which is specifically implemented in the following steps: S1: Collect air cargo order data from various aviation business systems including cargo handling, cargo check-in, cargo sorting, cargo reconfirmation, and waybill management systems; The collected data is cleaned, deduplicated, and format converted for preprocessing, and the processed data is imported into the data center; S2: Govern data in the data center, including data standard formulation, data quality monitoring, and structured processing of unstructured data including cargo images, videos, laser, RFID, and OCR recognition; In step S2, the following steps are specifically performed: S2.1: First, determine the data source, identify and list all relevant business systems, including data from the cargo handling system, cargo check-in system, cargo sorting system, cargo reconfirmation system, and waybill management system; S2.2: Extract data. Write data extraction scripts or configure data extraction tools according to data interfaces and data formats; extract data from various business systems regularly or in real time; S2.3: Perform data cleaning, identify and handle missing values, and fill missing data using the mean, median, mode, interpolation, or deletion; S2.4: Identify and handle outliers; remove duplicate data, using unique identifiers or business rules to identify and delete duplicate records; S2.5: Perform data format conversion to convert data in different formats into a unified format; for example, convert data in CSV, Excel, JSON and other formats into structured data in a database table or data warehouse. Or encode or decode data to ensure data consistency and readability.

[0022] S2.6: Perform data integration, import the cleaned and converted data into the data center, establish a data model in the data center, define the relationship and hierarchy between the data, and use ETL tools or custom scripts to integrate the data to ensure the accuracy and consistency of the data.

[0023] S3: Extract features related to order recommendation and matching from the integrated data, including cargo type, weight, volume, destination, flight time, and flight capacity, and use feature selection algorithms to select features that have a great impact on recommendation and matching results; specifically, use variance selection methods to select features that have a great impact on recommendation and matching results; S3.1: First, calculate the variance of each feature cargo type, weight, volume, destination, flight time, and flight capacity, set a variance threshold, and select features whose variance is greater than the threshold; as follows: in, is the i-th value of feature x, is the mean of feature x, n represents the number of samples; S3.2: Select features by calculating the correlation coefficient between the features and the target variable. The features with larger correlation coefficients have stronger correlation with the target variable and may have greater impact on the model. S3.3: Specifically, by calculating the correlation coefficient between each feature and the target variable, a correlation coefficient threshold is set, and the features with correlation coefficients greater than the threshold are selected, as shown in the following formula; in, and Represent the i-th value of feature X and target variable y respectively, and They represent the mean of feature x and variable y respectively, and n represents the number of samples; S3.4: Combine the feature values ​​calculated in steps S3.2 and S3.3, comprehensively evaluate the impact of feature subsets on recommendation and matching results, and select the optimal feature subset.

[0024] Please follow the steps below: like Figure 3-Figure 4; S4.1: For each feature subset, use k-fold cross validation to evaluate the performance of the model on different data subsets. Based on the results of cross validation, select the feature subset with the best performance as the final feature set; S4.2: Divide each feature subset into k subsets of equal size, initialize the performance evaluation index, and initialize a list for storing cross-validation results for each feature subset; Perform k iterations; in each iteration, select a subset as the validation set and the remaining k-1 subsets as the training set; use the training set data to train the model; Use the validation set data to evaluate the model performance and calculate the evaluation indicators including accuracy, recall, and F1 score; store the evaluation results in a list of the corresponding feature subset; S4.3: Calculate the average performance. For each feature subset, calculate its average performance in k iterations; S4.4: According to the average performance, select the feature subset with the best performance, determine the final feature set, and use the selected optimal feature subset as the final feature set.

[0025] S4: Build an order recommendation model based on machine learning algorithms, build a flight matching model using classification algorithms, train and optimize the model to improve the accuracy and efficiency of recommendations and matching; like Figure 5-Figure 6 , including the construction of order recommendation model; First, the feature value data set is divided into a training set, a validation set, and a test set from the feature values ​​that have a great impact on the calculated matching results, including cargo type, weight, volume, destination, flight time, and flight capacity; Train the model, use the training set data to train the model, and adjust the model parameters to minimize the loss function; Use validation set data to evaluate model performance and calculate evaluation indicators including accuracy, recall, F1 score, and AUC-ROC; Re-optimize the model, adjust model parameters according to the evaluation results, add new features, and use regularization techniques to optimize the model; Flight matching model construction; First, flight data is collected from the flight information system, including flight number, departure time, arrival time, flight capacity, and flight status; Regenerate features related to flight matching, including flight time windows, the difference between departure and arrival times, and flight remaining capacity data; Using the logistic regression algorithm, the data set is divided into a training set, a validation set, and a test set; the model is trained using the training set data, and the model parameters are adjusted to maximize the classification accuracy or AUC-ROC indicator; Use validation set data to evaluate model performance and calculate classification accuracy and AUC-ROC indicators; deploy the trained order recommendation model and flight matching model into the aviation logistics system; Based on the order information entered by the user, the order recommendation model is used to generate recommendation results. Based on the recommendation results and flight information, the flight matching model is used to generate matching results. The recommendations and matching results are displayed to the user for reference and decision-making.

[0026] S5: Based on the order information input by the user, including the type, weight, volume, and destination of goods, the recommendation model is used to recommend suitable air logistics orders for the user; based on the recommendation results and flight information, including flight time and flight capacity, the matching model is used to match the most suitable flight for the user; the recommendation and matching results are output for the user's reference and decision-making.

[0027] In this embodiment, the present invention provides an intelligent information release management system for air cargo terminal inbound and outbound business, including a user input module for receiving order information input by a user, including cargo type, weight, volume, and destination; receiving user input via a Web interface, a mobile application, or an API interface; and outputting the order information input by the user to an order recommendation module; The order recommendation module uses the recommendation model to recommend suitable air logistics orders to users based on the order information input by users. It loads the pre-trained recommendation model, pre-processes the order information input by users, such as standardization and encoding, and inputs the pre-processed order information into the recommendation model to obtain the recommendation results. Obtain and store flight information, including flight time and flight capacity, obtain flight information from airlines or third-party data sources, store flight information, and pass flight information to the flight matching module; The flight matching module uses the matching model to match the most suitable flight for the user based on the recommendation results and flight information, loads the pre-trained matching model, and performs matching calculations based on the recommendation results and flight information; outputs the matching results, including the recommended flight, flight time, and estimated arrival time; and passes the matching results to the result output module; The result output module outputs the recommendation and matching results to the user for reference and decision-making. It displays the recommendation and matching results through the web interface, mobile application or API interface, and provides user interaction functions, including selecting, confirming or modifying the recommendation and matching results.

[0028] In this embodiment, for example, there is a batch of logistics goods, and the information obtained from the air cargo terminal system is as follows: Table 1 Initial information table of goods Cargo ID Product Name Initial weight Initial volume 001 Electronics 2.5 kg 0.03 m³ 002 clothing 5.0 kg 0.10 m³ 003 food 1.2 kg 0.02 m³ The packaging information is as follows: Table 2 Packaging information of goods Cargo ID Packaging Type Packaging Materials Packed weight Volume after packaging 001 Anti-static bags Foam 2.6 kg 0.032 m³ 002 Carton Bubble Wrap 5.2 kg 0.11 m³ 003 Refrigerator ice bag 1.5 kg 0.03 m³ The recommended flights are as follows; Table 3 Recommended flight schedule Cargo ID Flight number destination Estimated time of departure 001 CA123 Beijing 2023-04-03 08:00 002 CZ456 Shanghai 2023-04-03 10:00 003 MU789 Guangzhou 2023-04-03 12:00 In this embodiment, the recommended cargo according to the flight is as follows: Table 4 Logistics Order Table Order Number Place of shipment Place of delivery Type of cargo Cargo weight Delivery time Customer needs OD001 Beijing Shanghai Electronics 3.0 kg 2024-11-14 Urgent delivery OD002 Beijing Guangzhou clothing 5.5 kg 2024-11-14 Next day delivery OD003 Shanghai Beijing food 2.2 kg 2024-11-14 Standard delivery OD004 Beijing Shanghai Fresh 1.8 kg 2024-11-14 Same-day delivery The flights pushed according to the order are as follows; Table 5 Recommended flight schedule Recommended Order Flight number Place of shipment Place of delivery Type of cargo Cargo weight Estimated time of departure OD001 KN5955 Beijing Shanghai Electronics 3.0 kg 2024-11-14 16:55 OD004 KN5955 Beijing Shanghai Fresh 1.8 kg 2024-11-14 16:55 OD001 CA1509 Beijing Shanghai Electronics 3.0 kg 2024-11-14 17:30 OD004 CA1509 Beijing Shanghai Fresh 1.8 kg 2024-11-14 17:30 In this embodiment, the present invention provides a computer-storable medium, wherein the computer-readable storage medium includes an embedded processing system and a stored program, and when the embedded system control program runs, the method for intelligent information release and management of air cargo terminal inbound and outbound business described in any one of the above is controlled and executed.

[0029] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention has various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent information publishing and management method for air cargo terminal inbound and outbound business, characterized by: Please follow the steps below: S1: Collect air cargo order data from various aviation business systems including cargo handling, cargo check-in, cargo sorting, cargo reconfirmation, and waybill management systems; The collected data is cleaned, deduplicated, and format converted for preprocessing, and the processed data is imported into the data center; S2: Govern data in the data center, including data standard formulation, data quality monitoring, and structured processing of unstructured data including cargo images, videos, laser, RFID, and OCR recognition; S3: Extract features related to order recommendation and matching from the integrated data, including cargo type, weight, volume, destination, flight time, and flight capacity, and use feature selection algorithms to select features that have a great impact on recommendation and matching results; Specifically, the variance selection method is used to screen out features that have a great impact on recommendation and matching results; S3.1: First, calculate the variance of each feature cargo type, weight, volume, destination, flight time, and flight capacity, set a variance threshold, and select features whose variance is greater than the threshold; as follows: in, is the i-th value of feature x, is the mean of feature x, n represents the number of samples; S3.2: Select features by calculating the correlation coefficient between the features and the target variable; S3.3: Specifically, by calculating the correlation coefficient between each feature and the target variable, a correlation coefficient threshold is set, and the features with correlation coefficients greater than the threshold are selected, as shown in the following formula; in, and Represent the i-th value of feature X and target variable y respectively, and They represent the mean of feature x and variable y respectively, and n represents the number of samples; S3.4: Combine the feature values ​​calculated in steps S3.2 and S3.3, comprehensively evaluate the impact of feature subsets on recommendation and matching results, and select the optimal feature subset; Then, for each feature subset, k-fold cross validation is used to evaluate the performance of the model on different data subsets. Based on the results of cross validation, the feature subset with the best performance is selected as the final feature set; Divide each feature subset into k subsets of equal size, initialize the performance evaluation index, and initialize a list for storing cross-validation results for each feature subset; Perform k iterations; in each iteration, select a subset as the validation set and the remaining k-1 subsets as the training set; use the training set data to train the model; Use the validation set data to evaluate the model performance and calculate the evaluation indicators including accuracy, recall, and F1 score; store the evaluation results in a list of the corresponding feature subset; Calculate the average performance. For each feature subset, calculate its average performance in k iterations; According to the average performance, the feature subset with the best performance is selected to determine the final feature set, and the selected optimal feature subset is used as the final feature set; S4: Build an order recommendation model based on machine learning algorithms, build a flight matching model using classification algorithms, train and optimize the model to improve the accuracy and efficiency of recommendations and matching; S5: Based on the order information input by the user, including the type, weight, volume, and destination of goods, the recommendation model is used to recommend suitable air logistics orders for the user; based on the recommendation results and flight information, including flight time and flight capacity, the matching model is used to match the most suitable flight for the user; the recommendation and matching results are output for the user's reference and decision-making.

2. The intelligent information publishing and management method for the inbound and outbound business of an air cargo terminal according to claim 1, characterized in that: In step S2, the following steps are specifically performed: S2.1: First, determine the data source, identify and list all relevant business systems, including data from the cargo handling system, cargo check-in system, cargo sorting system, cargo reconfirmation system, and waybill management system; S2.2: Extract data. Write data extraction scripts or configure data extraction tools according to data interfaces and data formats; extract data from various business systems regularly or in real time; S2.3: Perform data cleaning, identify and handle missing values, and fill missing data using the mean, median, mode, interpolation, or deletion; S2.4: Identify and handle outliers; remove duplicate data, using unique identifiers or business rules to identify and delete duplicate records; S2.5: Perform data format conversion to convert data in different formats into a unified format; or encode or decode data to ensure data consistency and readability; S2.6: Perform data integration, import the cleaned and converted data into the data center, establish a data model in the data center, define the relationship and hierarchy between the data, and use ETL tools or custom scripts to integrate the data.

3. The intelligent information publishing and management method for the inbound and outbound business of an air cargo terminal according to claim 1, characterized in that: In step S4, an order recommendation model is constructed; First, the feature value data set is divided into a training set, a validation set, and a test set from the feature values ​​that have a great impact on the calculated matching results, including cargo type, weight, volume, destination, flight time, and flight capacity; Train the model, use the training set data to train the model, and adjust the model parameters to minimize the loss function; Use validation set data to evaluate model performance and calculate evaluation indicators including accuracy, recall, F1 score, and AUC-ROC; Re-optimize the model, adjust model parameters according to the evaluation results, add new features, and use regularization techniques to optimize the model.

4. The intelligent information publishing and management method for the inbound and outbound business of an air cargo terminal according to claim 1, characterized in that: In step S4, a flight matching model is constructed; First, flight data is collected from the flight information system, including flight number, departure time, arrival time, flight capacity, and flight status; Regenerate features related to flight matching, including flight time windows, the difference between departure and arrival times, and flight remaining capacity data; Using the logistic regression algorithm, the data set is divided into training set, validation set and test set; Train the model, use the training set data to train the model, and adjust the model parameters to maximize the classification accuracy or AUC-ROC indicator; Use validation set data to evaluate model performance and calculate classification accuracy and AUC-ROC indicators; deploy the trained order recommendation model and flight matching model into the aviation logistics system; Based on the order information entered by the user, the order recommendation model is used to generate recommendation results. Based on the recommendation results and flight information, the flight matching model is used to generate matching results. The recommendations and matching results are displayed to the user for reference and decision-making.

5. An intelligent information release management system for air cargo terminal inbound and outbound business, characterized by: It includes a user input module, which is used to receive the order information input by the user, including the type, weight, volume, and destination of the goods; receive the user input through a web interface, a mobile application, or an API interface; output, and pass the order information input by the user to the order recommendation module; The order recommendation module uses the recommendation model to recommend suitable air logistics orders to users based on the order information entered by the user. It loads the pre-trained recommendation model and pre-processes the order information entered by the user. Obtain and store flight information, including flight time and flight capacity, obtain flight information from airlines or third-party data sources, store flight information, and pass flight information to the flight matching module; The flight matching module uses the matching model to match the most suitable flight for the user based on the recommendation results and flight information, loads the pre-trained matching model, and performs matching calculations based on the recommendation results and flight information; outputs the matching results, including the recommended flight, flight time, and estimated arrival time; and passes the matching results to the result output module; The result output module outputs the recommendation and matching results to the user for reference and decision-making. It displays the recommendation and matching results through the web interface, mobile application or API interface, and provides user interaction functions, including selecting, confirming or modifying the recommendation and matching results.

6. A computer storable medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a method for intelligent information release and management of air cargo terminal inbound and outbound business as described in any one of claims 1 to 4.

Images