A method, device, electronic device and storage medium for determining flight tags

Abstract

The present application relates to the field of air transportation, and in particular to a method, device, electronic device and storage medium for determining flight labels, the method comprising: obtaining a flight attribute feature matrix of a flight to be determined, as well as a historical flight attribute feature matrix and a historical flight label matrix of historical flights; inputting the flight attribute feature matrix of the flight to be determined into a first label probability module in a label determination model to obtain a first label probability matrix of the flight to be determined; inputting the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into a second label probability module in a label determination model to obtain a second label probability matrix of the flight to be determined; inputting the first label probability matrix and the second label probability matrix into a fusion module in a label determination model to obtain a flight label of the flight to be determined. The present application can determine the flight label of the flight to be determined through a label determination model, thereby improving the accuracy of determining the label.

Description

Technical Field

[0001] The present invention relates to the field of air transportation, and in particular to a method, device, electronic device and storage medium for determining a flight tag. Background Art

[0002] In the field of air transportation, flight tags, as an important part of the operation control system, play a key role in refined flight support management. Flight tags usually contain rich semantic information, such as flight number, take-off and landing time, aircraft type, parking space, terminal, etc. These flight attribute information are crucial for refined flight management. Through flight tags, the operation control system can realize real-time monitoring of flight status, adjust flight support plans in a timely manner, and ensure flight punctuality and passengers' travel experience. At the same time, flight tags also provide a basis for collaborative work for relevant departments such as airlines, airports, and ground services, and promote close cooperation and information sharing between various links.

[0003] The current generation of flight labels mainly relies on a series of fixed label generation rules. Although this method is simple and direct, it has its inherent limitations when facing the diversification of flight operations. Fixed label generation rules often fail to fully consider the uniqueness and complexity of each flight, resulting in the possibility that the generated labels may not be accurate enough. Summary of the invention

[0004] In view of this, the purpose of the present application is to provide a method, device, electronic device and storage medium for determining a flight tag, which can determine the flight tag and improve the accuracy of determining the tag.

[0005] In a first aspect, an embodiment of the present application provides a method for determining a flight label, and the method for determining a flight label includes:

[0006] Obtain the flight attribute feature matrix of the flight to be determined, as well as the historical flight attribute feature matrix and historical flight label matrix of the historical flights;

[0007] Inputting the flight attribute feature matrix of the flight to be determined into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined; the label determination model is obtained by training the flight attribute feature matrix of the sample flight and the corresponding flight attribute feature matrix;

[0008] Input the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined; the first label probability matrix and the second label probability matrix contain the label probabilities corresponding to each flight attribute;

[0009] The first label probability matrix and the second label probability matrix are input into a fusion module in the label determination model to obtain the flight label of the flight to be determined.

[0010] In a possible implementation, the historical flight attribute feature matrix and the historical flight label matrix of the historical flights are input into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined, including:

[0011] Determine the historical flights having a first preset number of historical flight attribute feature matrices that are most similar to the flight attribute feature matrix of the flight to be determined as target historical flights;

[0012] Perform weighted summation on the historical flight label matrix of the target historical flight to obtain a second label probability matrix of the flight to be determined;

[0013] The weight of the historical flight label matrix of each target historical flight is the similarity between the historical flight attribute feature matrix of each target historical flight and the flight attribute feature matrix of the flight to be determined.

[0014] In a possible implementation, the first label probability matrix and the second label probability matrix are input into a fusion module in a label determination model to obtain a flight label of a flight to be determined, including:

[0015] According to the first label probability matrix and the second label probability matrix, a target label probability matrix of the flight to be determined is calculated; the target label probability matrix contains the label probabilities corresponding to the attributes of each flight;

[0016] According to the target label probability matrix, the flight label of the flight to be determined is determined.

[0017] In a possible implementation, calculating a target label probability matrix of the flight to be determined according to the first label probability matrix and the second label probability matrix includes:

[0018] The first label probability matrix and the second label probability matrix are weightedly summed to obtain the target label probability matrix of the flight to be determined.

[0019] In a possible implementation, determining the flight label of the to-be-determined flight according to the target label probability matrix includes:

[0020] The second preset number of flight attribute features with the highest label probabilities in the target label probability matrix are determined as the flight labels of the flight to be determined.

[0021] In a possible implementation, obtaining a historical flight attribute feature matrix and a historical flight label matrix of historical flights includes:

[0022] Obtain the historical flight label data and historical flight attribute feature data of the initial historical flights;

[0023] Count the frequency of each historical flight tag among all historical flight tags;

[0024] Determine the initial historical flight whose frequency of the historical flight label is greater than or equal to the preset frequency as the final historical flight;

[0025] The final historical flight labels and historical flight attribute feature data of the historical flights are converted into feature vectors to obtain a historical flight label matrix and a historical flight attribute feature matrix.

[0026] In a second aspect, an embodiment of the present application further provides a device for determining a flight label, the device for determining a flight label comprising:

[0027] An acquisition module, used to acquire a flight attribute feature matrix of a to-be-determined flight, as well as a historical flight attribute feature matrix and a historical flight label matrix of historical flights;

[0028] An input module is used to input the flight attribute feature matrix of the flight to be determined into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined; the label determination model is obtained by training the flight attribute feature matrix of the sample flight and the corresponding flight attribute feature matrix;

[0029] The input module is also used to input the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined; the first label probability matrix and the second label probability matrix contain the label probability corresponding to each flight attribute;

[0030] The input module is also used to input the first label probability matrix and the second label probability matrix into the fusion module in the label determination model to obtain the flight label of the flight to be determined.

[0031] In a possible implementation, the input module is specifically used to determine the historical flights of a first preset number of historical flight attribute feature matrices that are most similar to the flight attribute feature matrix of the flight to be determined as the target historical flights; perform weighted summation on the historical flight label matrices of the target historical flights to obtain a second label probability matrix of the flight to be determined; wherein the weight of the historical flight label matrix of each target historical flight is the similarity between the historical flight attribute feature matrix of each target historical flight and the flight attribute feature matrix of the flight to be determined.

[0032] In a possible implementation, the input module is specifically used to calculate a target label probability matrix for the flight to be determined based on the first label probability matrix and the second label probability matrix; the target label probability matrix contains label probabilities corresponding to attribute features of each flight; and the flight label of the flight to be determined is determined based on the target label probability matrix.

[0033] In a possible implementation, the input module is specifically configured to perform a weighted summation on the first label probability matrix and the second label probability matrix to obtain a target label probability matrix for the flight to be determined.

[0034] In a possible implementation, the input module is specifically configured to determine a second preset number of flight attribute features with the highest label probabilities in the target label probability matrix as flight labels of the flight to be determined.

[0035] In a possible implementation, the acquisition module is specifically used to obtain historical flight label data and historical flight attribute feature data of an initial historical flight; count the frequency of each historical flight label among all historical flight labels; determine the initial historical flight whose frequency of historical flight labels is greater than or equal to a preset frequency as the final historical flight; convert the historical flight label and historical flight attribute feature data of the final historical flight into feature vectors to obtain a historical flight label matrix and a historical flight attribute feature matrix.

[0036] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of the method for determining a flight tag as described in any one of the first aspects.

[0037] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for determining a flight tag as described in any one of the first aspects are executed.

[0038] The embodiment of the present application provides a method, device, electronic device and storage medium for determining flight labels, the method comprising: obtaining a flight attribute feature matrix of a flight to be determined, as well as a historical flight attribute feature matrix and a historical flight label matrix of historical flights; inputting the flight attribute feature matrix of the flight to be determined into a first label probability module in a label determination model to obtain a first label probability matrix of the flight to be determined; inputting the historical flight attribute feature matrix and the historical flight label matrix of the historical flights into a second label probability module in a label determination model to obtain a second label probability matrix of the flight to be determined; inputting the first label probability matrix and the second label probability matrix into a fusion module in a label determination model to obtain a flight label of the flight to be determined. The present application can determine the flight label of the flight to be determined through a label determination model, thereby improving the accuracy of determining the label. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application 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 paying creative work.

[0040] Figure 1 A flow chart of a method for determining a flight tag provided in an embodiment of the present application is shown;

[0041] Figure 2 A schematic diagram of the structure of a first label probability module provided in an embodiment of the present application is shown;

[0042] Figure 3 A schematic diagram of the structure of a device for determining a flight tag provided in an embodiment of the present application is shown;

[0043] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0044] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0045] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0046] In order to enable those skilled in the art to use the content of this application, the following implementation is provided in conjunction with a specific application scenario "air transportation field". For those skilled in the art, the general principles defined herein can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application. Although this application is mainly described around the "air transportation field", it should be understood that this is only an exemplary embodiment.

[0047] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0048] The following is a detailed description of a method for determining a flight label provided in an embodiment of the present application.

[0049] Reference Figure 1 As shown, it is a flow chart of a method for determining a flight tag provided in an embodiment of the present application. The specific execution process of the method for determining a flight tag is as follows:

[0050] S101. Obtain a flight attribute feature matrix of a to-be-determined flight, as well as a historical flight attribute feature matrix and a historical flight label matrix of historical flights.

[0051] S102: Input the flight attribute feature matrix of the flight to be determined into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined.

[0052] S103: Input the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined.

[0053] S104: Input the first label probability matrix and the second label probability matrix into a fusion module in the label determination model to obtain a flight label of the flight to be determined.

[0054] The embodiment of the present application provides a method for determining a flight label, the method comprising: obtaining a flight attribute feature matrix of a flight to be determined, and a historical flight attribute feature matrix and a historical flight label matrix of historical flights; inputting the flight attribute feature matrix of the flight to be determined into a first label probability module in a label determination model to obtain a first label probability matrix of the flight to be determined; inputting the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into a second label probability module in the label determination model to obtain a second label probability matrix of the flight to be determined; inputting the first label probability matrix and the second label probability matrix into a fusion module in the label determination model to obtain a flight label of the flight to be determined. The present application can determine the flight label of the flight to be determined through the label determination model, thereby improving the accuracy of determining the label.

[0055] The following is a description of the exemplary steps of the present application embodiment:

[0056] S101. Obtain a flight attribute feature matrix of a to-be-determined flight, as well as a historical flight attribute feature matrix and a historical flight label matrix of historical flights.

[0057] In the implementation manner of the present application, the flight attribute feature matrix of the flight to be determined contains the feature vector of the flight attributes of the flight to be determined; the historical flight attribute feature matrix contains the feature vector of the flight attributes of the historical flight. The flight attribute feature matrix and the historical flight attribute feature matrix are both n*m matrices; wherein n is the number of flight attributes, and m is the length of the flight attribute feature vector. The historical flight label matrix contains the label status of each flight attribute, and the label status of the flight attribute includes two states: the flight attribute is a label and the flight attribute is not a label; the historical flight label matrix is ​​a 1*n matrix composed of 0 and 1, and the value 1 indicates that the flight attribute is a label, and the value 0 indicates that the flight attribute is not a label.

[0058] Among them, flight attributes may include flight number, terminal, flight mission, arrival and departure, aircraft position, boarding gate, aircraft model, flight attributes, airline, apron, aircraft model level, concourse, check-in counter, terminal, baggage claim carousel, terminal area, etc., depending on the actual situation.

[0059] Specifically, obtaining the flight attribute feature matrix of the flight to be determined includes: obtaining flight attribute feature data of the flight to be determined; converting each flight attribute feature data of the flight to be determined into a feature vector by Word2vec to obtain the flight attribute feature matrix of the flight to be determined.

[0060] The flight attribute characteristic data of the flight to be determined contains the values ​​of the flight attributes of the flight to be determined.

[0061] Here, Word2vec is a tool that converts words into vector form, which can simplify the processing of text content into vector operations in vector space.

[0062] Specifically, obtaining a historical flight attribute feature matrix and a historical flight label matrix of historical flights includes: obtaining historical flight label data and historical flight attribute feature data of an initial historical flight; counting the frequency of each historical flight label among all historical flight labels; determining an initial historical flight whose frequency of a historical flight label is greater than or equal to a preset frequency as a final historical flight; converting the historical flight label and the historical flight attribute feature data of the final historical flight into a feature vector to obtain a historical flight label matrix and a historical flight attribute feature matrix.

[0063] The historical flight tag data includes the values ​​of flight attributes of the historical flight tags; and the historical flight attribute feature data includes the values ​​of flight attributes of the historical flight.

[0064] Here, the lower the frequency of historical flight labels, the less accurate the determined flight labels. Filtering historical flights according to the frequency of historical flight labels can remove the noise of historical data and improve the accuracy of determining flight labels.

[0065] S102: Input the flight attribute feature matrix of the flight to be determined into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined.

[0066] In the implementation mode of the present application, the first label probability matrix includes the label probabilities corresponding to each flight attribute, and the first label probability matrix is ​​a 1*n matrix. The label determination model is used to determine the flight label of the flight to be determined. The label determination model is obtained by training the flight attribute feature matrix of the sample flight and the corresponding flight label.

[0067] Specifically, the first label probability module is mainly composed of two network models: Bi-LSTM and CNN, which aims to deeply mine the semantic information contained in the flight attribute feature matrix to make accurate label recommendations. Figure 2 FIG. 1 is a schematic diagram of the structure of the first tag probability module provided in an embodiment of the present application, wherein the flight to be determined , its flight attribute feature matrix ,in Indicates the flight number to be confirmed The feature vector of flight attributes, Indicates the number of flight attributes. In the Bi-LSTM model, the flight attribute feature matrix It is fed into both the forward LSTM and the backward LSTM at the same time; these two LSTM networks process the input independently but in opposite directions.

[0068] Specifically, the flight attribute feature matrix of the flight to be determined is input into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined, including:

[0069] Step 1: Determine the forward hidden state corresponding to the feature vector of each flight attribute according to the hidden state of the feature vector of the previous flight attribute corresponding to the feature vector of each flight attribute in the flight attribute feature matrix.

[0070] Here, the forward LSTM is used to calculate the feature vector of each flight attribute. and the hidden state of the feature vector of the corresponding previous flight attribute To determine the feature vector of each flight attribute The corresponding forward hidden state :

[0071] The specific formula is: ;

[0072] in, is the LSTM function.

[0073] Step 2: Determine the backward hidden state corresponding to the feature vector of each flight attribute according to the hidden state of the feature vector of the next flight attribute corresponding to the feature vector of each flight attribute.

[0074] Here, the backward LSTM is used to calculate the feature vector of each flight attribute. and the corresponding feature vector of the next flight attribute The hidden state of To determine the feature vector of each flight attribute The corresponding backward hidden state :

[0075] The specific formula is: .

[0076] Step 3: Concatenate the forward hidden state and the backward hidden state to obtain the final hidden state corresponding to the feature vector of each flight attribute .

[0077] In the implementation manner of the present application, the forward hidden state and the backward hidden state are concatenated by the following formula.

[0078] .

[0079] Step 4: Hidden state corresponding to the feature vector of each flight attribute Attention weights corresponding to each time step , determine the context vector of the flight to be determined .

[0080] ;

[0081] in, is the attention weight, which indicates the model's The final hidden state corresponding to the feature vector of flight attributes degree of attention.

[0082] Step 5: The context vector of the flight to be determined Substitute into the following formula to determine the first label probability matrix of the flight to be determined .

[0083] ;

[0084] in, is the weight matrix of the fully connected layer Dense, is the bias vector of the fully connected layer Dense.

[0085] here, is a normalized exponential function in probability theory and related fields, which can "compress" a K-dimensional vector z containing any real number into another K-dimensional real vector σ(z) so that each element ranges between (0,1) and the sum of all elements is 1. This function is often used in multi-classification problems.

[0086] S103: Input the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined.

[0087] In an implementation manner of the present application, the second label probability matrix contains the label probabilities corresponding to each flight attribute, and the second label probability matrix is ​​a 1*n matrix; the second label probability matrix is ​​obtained by weighted summing the historical flight label matrices of a first preset number of historical flights that are most similar to the flight to be determined; the first preset number is obtained by training the label determination model.

[0088] Specifically, the historical flight attribute feature matrix and the historical flight label matrix of the historical flight are input into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined, including:

[0089] Step 1: determine the historical flights having a first preset number of historical flight attribute feature matrices that are most similar to the flight attribute feature matrix of the flight to be determined as target historical flights.

[0090] In the implementation manner of the present application, the target historical flight is a first preset number of historical flights that are most similar to the flight to be determined.

[0091] Here, the flight attribute feature matrix of the flight to be determined and the historical flight attribute feature matrix of each historical flight are substituted into the formula to obtain the similarity between the flight to be determined and the historical flights:

[0092] ;

[0093] in, For flights to be confirmed, is the jth historical flight, is the similarity between the historical flight attribute feature matrix of the historical flight and the flight attribute feature matrix of the flight to be determined, is the historical flight attribute feature matrix of the flight to be determined, is the historical flight attribute feature matrix of the j-th historical flight.

[0094] Step 2: Perform weighted summation on the historical flight label matrices of the target historical flight to obtain a second label probability matrix of the flight to be determined.

[0095] The weight of the historical flight label matrix of each target historical flight is the similarity between the historical flight attribute feature matrix of each target historical flight and the flight attribute feature matrix of the flight to be determined.

[0096] In the implementation manner of the present application, the historical flight label matrix of the target historical flight is substituted into the following formula to obtain the second label probability matrix of the flight to be determined:

[0097] ;

[0098] in, is the second label probability matrix of the flight to be determined, is the first preset quantity, For flights to be confirmed, is the jth target historical flight, is the similarity between the historical flight attribute feature matrix of the target historical flight and the flight attribute feature matrix of the flight to be determined, is the historical flight label matrix of the j-th target historical flight.

[0099] Example, assuming and is with The two most similar target historical flights are The similarities are 0.3 and 0.2 respectively, and and The historical flight label matrix is and . The second label probability matrix The calculation process is as follows:

[0100] .

[0101] S104: Input the first label probability matrix and the second label probability matrix into a fusion module in the label determination model to obtain a flight label of the flight to be determined.

[0102] Specifically, a target label probability matrix of the flight to be determined is calculated according to the first label probability matrix and the second label probability matrix.

[0103] In the implementation manner of the present application, the target label probability matrix contains the label probabilities corresponding to the attributes of each flight. Specifically, the first label probability matrix and the second label probability matrix are weighted and summed to obtain the target label probability matrix of the flight to be determined.

[0104] Here, the first label probability matrix and the second label probability matrix are weighted and summed by the following formula to obtain the target label probability matrix of the flight to be determined:

[0105] ;

[0106] ;

[0107] in, is the target label probability matrix, is the first label probability matrix The weight of is the second label probability matrix The weight of .

[0108] Specifically, the flight label of the flight to be determined is determined according to the target label probability matrix.

[0109] In an embodiment of the present application, the second preset number of flight attribute features with the highest label probability in the target label probability matrix are determined as the flight label of the flight to be determined; or the flight attribute features with label probability greater than the preset probability in the target label probability matrix are determined as the flight label of the flight to be determined.

[0110] Furthermore, the label determination model in the implementation manner of the present application includes a first label probability module, a second label probability module and a fusion module, and the label determination model is trained specifically through the following steps:

[0111] Step 1: Obtain the flight attribute feature matrix and the corresponding label probability matrix of the sample flights.

[0112] Step 2: Input the flight attribute feature matrix of the sample flight into the label determination model to obtain the predicted flight label of the sample flight.

[0113] Step 3: Determine the accuracy of the label determination model based on the flight labels of the sample flights and the predicted flight labels.

[0114] Step 4: If the accuracy is less than the preset threshold, the label determination model is updated to obtain the latest label determination model; and the flight attribute feature matrix of the sample flight is input into the label determination model to obtain the predicted flight label of the sample flight to continue execution.

[0115] In an embodiment of the present application, when the label determination model is updated, the dimension of the feature vector in the first label probability module and the number of units in the LSTM hidden layer, the first preset number in the second label probability module, the learning rate, batch size, the number of training rounds, and the optimization function are specifically updated.

[0116] Step 5: If the accuracy is greater than or equal to the preset threshold, the current label determination model is determined as the final label determination model.

[0117] Furthermore, the performance index of the label determination model is calculated by the following formula:

[0118] Assume that TP is a positive sample judged as positive, TN is a negative sample judged as negative, FP is a negative sample judged as positive, and FN is a positive sample judged as negative.

[0119] Recall is the probability of being predicted as a positive sample among the samples that are actually positive. The calculation process is as follows:

[0120] ;

[0121] Precision is the probability that all samples predicted as positive are actually positive samples, representing the accuracy of the prediction in the positive sample results. The calculation process is as follows:

[0122] ;

[0123] The F1 score can be understood as the recall rate And precision The harmonic mean combines the results of precision and recall, and its calculation process is as follows:

[0124] .

[0125] Based on the same inventive concept, the embodiments of the present application also provide a device for determining a flight tag corresponding to the method for determining a flight tag. Since the principle of solving the problem by the device in the embodiments of the present application is similar to the method for determining a flight tag in the embodiments of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0126] Reference Figure 3 FIG. 1 is a schematic diagram of a device for determining a flight tag provided in an embodiment of the present application, and the device for determining a flight tag includes:

[0127] In a second aspect, an embodiment of the present application further provides a device for determining a flight label, the device for determining a flight label comprising:

[0128] An acquisition module 301 is used to acquire a flight attribute feature matrix of a to-be-determined flight, as well as a historical flight attribute feature matrix and a historical flight label matrix of historical flights;

[0129] An input module 302 is used to input the flight attribute feature matrix of the flight to be determined into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined; the label determination model is obtained by training the flight attribute feature matrix of the sample flight and the corresponding flight attribute feature matrix;

[0130] The input module 302 is also used to input the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined; the first label probability matrix and the second label probability matrix contain the label probability corresponding to each flight attribute;

[0131] The input module 302 is further used to input the first label probability matrix and the second label probability matrix into the fusion module in the label determination model to obtain the flight label of the flight to be determined.

[0132] In one possible implementation, the input module 302 is specifically used to determine the historical flights whose first preset number of historical flight attribute feature matrices are most similar to the flight attribute feature matrix of the flight to be determined as the target historical flights; perform weighted summation on the historical flight label matrices of the target historical flights to obtain a second label probability matrix of the flight to be determined; wherein the weight of the historical flight label matrix of each target historical flight is the similarity between the historical flight attribute feature matrix of each target historical flight and the flight attribute feature matrix of the flight to be determined.

[0133] In a possible implementation, the input module 302 is specifically used to calculate a target label probability matrix for the flight to be determined based on the first label probability matrix and the second label probability matrix; the target label probability matrix contains label probabilities corresponding to attribute features of each flight; and the flight label of the flight to be determined is determined based on the target label probability matrix.

[0134] In a possible implementation, the input module 302 is specifically configured to perform a weighted summation on the first label probability matrix and the second label probability matrix to obtain a target label probability matrix for the flight to be determined.

[0135] In a possible implementation, the input module 302 is specifically configured to determine a second preset number of flight attribute features with the highest label probabilities in the target label probability matrix as flight labels of the flight to be determined.

[0136] In a possible implementation, the acquisition module 301 is specifically used to obtain historical flight label data and historical flight attribute feature data of an initial historical flight; count the frequency of each historical flight label among all historical flight labels; determine the initial historical flight whose frequency of historical flight labels is greater than or equal to a preset frequency as the final historical flight; convert the historical flight label and historical flight attribute feature data of the final historical flight into feature vectors to obtain a historical flight label matrix and a historical flight attribute feature matrix.

[0137] The embodiment of the present application provides a device for determining flight labels, which includes: an acquisition module 301, which is used to obtain a flight attribute feature matrix of a flight to be determined, as well as a historical flight attribute feature matrix and a historical flight label matrix of a historical flight; an input module 302, which is used to input the flight attribute feature matrix of the flight to be determined into a first label probability module in a label determination model to obtain a first label probability matrix of the flight to be determined; the label determination model is obtained by training the flight attribute feature matrix of a sample flight and the corresponding flight; the input module 302 is also used to input the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into a second label probability module in the label determination model to obtain a second label probability matrix of the flight to be determined; the first label probability matrix and the second label probability matrix contain label probabilities corresponding to each flight attribute; the input module 302 is also used to input the first label probability matrix and the second label probability matrix into a fusion module in the label determination model to obtain the flight label of the flight to be determined. The present application can determine the flight label of the flight to be determined through the label determination model, thereby improving the accuracy of determining the label.

[0138] like Figure 4 As shown, an electronic device 400 provided in an embodiment of the present application includes: a processor 401, a memory 402 and a bus, the memory 402 stores machine-readable instructions executable by the processor 401, when the electronic device is running, the processor 401 and the memory 402 communicate through the bus, and the processor 401 executes the machine-readable instructions to perform the steps of the method for determining the flight tag as described above.

[0139] Specifically, the memory 402 and the processor 401 can be general-purpose memories and processors, which are not specifically limited here. When the processor 401 runs the computer program stored in the memory 402, the method for determining the flight tag can be executed.

[0140] Corresponding to the above-mentioned method for determining the flight label, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for determining the flight label are executed.

[0141] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0142] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0143] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0144] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the information processing method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0145] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for determining a flight label, characterized in that: The method for determining the flight label includes: Obtain the flight attribute feature matrix of the flight to be determined, as well as the historical flight attribute feature matrix and historical flight label matrix of the historical flights; Inputting the flight attribute feature matrix of the flight to be determined into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined; the label determination model is obtained by training the flight attribute feature matrix of the sample flight and the corresponding flight label; Inputting the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined; the first label probability matrix and the second label probability matrix contain the label probabilities corresponding to each flight attribute; The first label probability matrix and the second label probability matrix are input into a fusion module in a label determination model to obtain a flight label of the flight to be determined.

2. The method for determining a flight tag according to claim 1, characterized in that: The step of inputting the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined includes: Determine the historical flights having a first preset number of historical flight attribute feature matrices that are most similar to the flight attribute feature matrix of the flight to be determined as target historical flights; Performing weighted summation on the historical flight label matrices of the target historical flight to obtain a second label probability matrix of the flight to be determined; The weight of the historical flight label matrix of each target historical flight is the similarity between the historical flight attribute feature matrix of each target historical flight and the flight attribute feature matrix of the flight to be determined.

3. The method for determining a flight tag according to claim 1, characterized in that: The step of inputting the first label probability matrix and the second label probability matrix into a fusion module in a label determination model to obtain the flight label of the flight to be determined includes: Calculating a target label probability matrix of the to-be-determined flight according to the first label probability matrix and the second label probability matrix; the target label probability matrix includes label probabilities corresponding to attributes of each flight; The flight label of the to-be-determined flight is determined according to the target label probability matrix.

4. The method for determining a flight tag according to claim 3, characterized in that: The step of calculating the target label probability matrix of the to-be-determined flight according to the first label probability matrix and the second label probability matrix includes: A weighted sum is performed on the first label probability matrix and the second label probability matrix to obtain a target label probability matrix of the flight to be determined.

5. The method for determining a flight tag according to claim 3, characterized in that: Determining the flight label of the to-be-determined flight according to the target label probability matrix includes: A second preset number of flight attribute features with the highest label probabilities in the target label probability matrix are determined as flight labels of the flight to be determined.

6. The method for determining a flight tag according to claim 1, characterized in that: Get the historical flight attribute feature matrix and historical flight label matrix of historical flights, including: Obtain the historical flight label data and historical flight attribute feature data of the initial historical flights; Count the frequency of each historical flight tag among all historical flight tags; Determine the initial historical flight whose frequency of historical flight tags is greater than or equal to a preset frequency as the final historical flight; The historical flight label and historical flight attribute feature data of the final historical flight are converted into feature vectors to obtain a historical flight label matrix and a historical flight attribute feature matrix.

7. A device for determining a flight tag, characterized in that: The flight tag determination device includes: An acquisition module, used to acquire a flight attribute feature matrix of a to-be-determined flight, as well as a historical flight attribute feature matrix and a historical flight label matrix of historical flights; An input module, used to input the flight attribute feature matrix of the flight to be determined into the first label probability module in the label determination model to obtain the first label probability matrix of the flight to be determined; the label determination model is obtained by training the flight attribute feature matrix of the sample flight and the corresponding flight label; The input module is further used to input the historical flight attribute feature matrix and the historical flight label matrix of the historical flight into the second label probability module in the label determination model to obtain the second label probability matrix of the flight to be determined; the first label probability matrix and the second label probability matrix contain the label probability corresponding to each flight attribute; The input module is also used to input the first label probability matrix and the second label probability matrix into a fusion module in the label determination model to obtain the flight label of the flight to be determined.

8. The device for determining a flight tag according to claim 7, characterized in that: The input module is specifically used for: Determine the historical flights having a first preset number of historical flight attribute feature matrices that are most similar to the flight attribute feature matrix of the flight to be determined as target historical flights; Performing weighted summation on the historical flight label matrices of the target historical flight to obtain a second label probability matrix of the flight to be determined; The weight of the historical flight label matrix of each target historical flight is the similarity between the historical flight attribute feature matrix of each target historical flight and the flight attribute feature matrix of the flight to be determined.

9. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of the method for determining a flight tag as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for determining a flight tag according to any one of claims 1 to 6 are executed.

Images