CN115310732B - Flight delay prediction method and system - Google Patents
Flight delay prediction method and system Download PDFInfo
- Publication number
- CN115310732B CN115310732B CN202211243543.7A CN202211243543A CN115310732B CN 115310732 B CN115310732 B CN 115310732B CN 202211243543 A CN202211243543 A CN 202211243543A CN 115310732 B CN115310732 B CN 115310732B
- Authority
- CN
- China
- Prior art keywords
- flight
- neural network
- network model
- delay
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000012545 processing Methods 0.000 claims abstract description 30
- 239000013598 vector Substances 0.000 claims abstract description 27
- 230000003111 delayed effect Effects 0.000 claims abstract description 25
- 238000003062 neural network model Methods 0.000 claims description 118
- 238000012549 training Methods 0.000 claims description 103
- 230000006870 function Effects 0.000 claims description 25
- 230000005477 standard model Effects 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 210000002569 neuron Anatomy 0.000 claims description 9
- 210000004205 output neuron Anatomy 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000011176 pooling Methods 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 239000000126 substance Substances 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 5
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 9
- 238000013135 deep learning Methods 0.000 description 8
- 238000005406 washing Methods 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 4
- 230000004044 response Effects 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 230000010006 flight Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/109—Time management, e.g. calendars, reminders, meetings or time accounting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a flight delay prediction method and a system, wherein the method comprises the steps of obtaining flight operation information of a target flight, weather information of a take-off and landing city of the target flight in a preset operation time and associated delay information of an airport associated with the target flight; after the flight operation information, the weather information and the associated delay information are converted into feature vectors, predicting whether the target flight will be delayed or not through a pre-constructed flight delay prediction model; and if the delay of the target flight is predicted, further determining the delay level of the target flight, and determining a processing scheme corresponding to the delay level according to the corresponding relation acquired in advance. The method disclosed by the invention can comprehensively influence the factors of flight delay to predict the flight delay and provide a corresponding processing scheme according to the prediction result.
Description
Technical Field
The disclosure relates to the technical field of big data analysis, in particular to a flight delay prediction method and a flight delay prediction system.
Background
Due to the weather, the flow, passengers and the like, part of flights are delayed or cancelled frequently, and a large number of flight delays are easily formed under the influence of time and space correlation of a flight network.
The application number is 202110436375.2, the invention is named as an intelligent flight delay prediction method based on deep learning, and discloses the following contents:
acquiring flight data of a target flight and weather data of a take-off and landing city within the running time of the target flight, calculating delay time of the target flight, and manufacturing a delay label according to the delay time; performing digital processing on flight data and weather data; preprocessing the flight data and the weather data after digital processing, dividing a plurality of delay categories according to delay labels, and setting a deep learning model; training a deep learning algorithm model by using the preprocessed flight data and weather data, and predicting the delay condition of the target flight by using a weather forecast and deep learning algorithm;
although the flight delay analysis is carried out through deep learning in the comparison file, the result prediction is carried out only through a single model, flight data are formed by summarizing various types of data, and the prediction result of the single model is often unsatisfactory; and as the delay time becomes longer, the probability that the predicted value deviates from the standard value is greatly increased, and the local optimal solution is easily trapped.
The application number is 201911066077.8, the invention is named as an 'departure flight delay prediction method based on aging information and deep learning', and the following contents are disclosed:
acquiring flight information data and meteorological information data; performing feature conversion according to the flight information data and the meteorological information data, establishing an departing flight delay prediction model based on deep learning, and predicting flight delay by adopting the departing flight delay prediction model based on the deep learning;
although the time-efficiency information is added to the comparison file on the basis of deep learning, the comparison file still adopts a single model to carry out delay prediction, and the prediction result of the single model is easy to fall into the difficulty of local optimal solution.
The information disclosed in this background section of the application is only for enhancement of understanding of the general background of the application and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art that is already known to a person skilled in the art.
Disclosure of Invention
The disclosed embodiment provides a flight delay prediction method and a flight delay prediction system, and the method can at least solve part of problems in the prior art.
In a first aspect of an embodiment of the present disclosure,
a flight delay prediction method is provided, and the method comprises
Acquiring flight operation information of a target flight, weather information of a take-off and landing city of the target flight within preset operation time, and associated delay information of an airport associated with the target flight;
after the flight operation information, the weather information and the associated delay information are converted into feature vectors, whether the target flight will be delayed or not is predicted through a pre-constructed flight delay prediction model,
if the target flight is predicted to be delayed, further determining the delay level of the target flight, and determining a processing scheme corresponding to the delay level according to a pre-acquired corresponding relation, wherein,
the flight delay prediction model is constructed based on a plurality of neural network models, and the output of the preceding neural network model is used as the input of the following neural network model and is used for performing delay prediction on the target flight.
In an alternative embodiment of the method according to the invention,
the method further comprises training a flight delay prediction model, and the method for training the flight delay prediction model comprises the following steps:
constructing a model training data set, wherein the model training data set comprises historical flight operation information, historical weather information and historical associated delay information in a preset time period;
converting the model training data set into training characteristic vectors, inputting a flight delay prediction model to be trained, updating the model training data set based on a training output result of the flight delay prediction model to be trained until a preset training convergence condition is reached to finish training the flight delay prediction model to be trained, wherein,
the flight delay prediction model to be trained comprises a preceding neural network model and a succeeding neural network model, the preceding neural network model performs preceding training learning according to the training feature vector and outputs a first output result to the succeeding neural network model, and the succeeding neural network performs succeeding training learning according to the first output result and outputs a training prediction value.
In an alternative embodiment of the method according to the invention,
before inputting the training feature vector into a flight delay prediction model to be trained, the method further comprises:
determining a data preprocessing method corresponding to the training data characteristics according to the training data characteristics of the model training data set, and screening out dirty data in the model training data set to obtain a standard model training data set, wherein the standard model training data set comprises standard flight operation information, standard weather information and standard association delay information;
respectively converting standard flight operation information, standard weather information and standard associated delay information of the standard model training data set into corresponding standard data characteristics;
and scoring the standard flight operation information, the standard weather information and the standard association delay information according to the data type corresponding to the standard data characteristic, and converting the information with the score exceeding a preset threshold value into a training characteristic vector.
In an alternative embodiment of the method according to the invention,
the method for training the flight delay prediction model further comprises the step of constructing a prior neural network model for training the flight delay prediction model based on the deep neural network model, wherein the prior neural network model comprises an input layer, a convolutional layer, a hidden layer and an output layer,
the convolutional layer performs batch normalization processing on the training feature vectors and determines output features corresponding to all hidden layers;
determining a predicted value corresponding to the output feature based on a first loss function of the preceding neural network model, and judging whether the preceding neural network model converges according to the predicted value and a loss value of a preset target value,
if not, determining the gradient value of the hidden layer through a back propagation algorithm according to the loss value, and updating the weight value of each layer of the prior neural network model according to the gradient value until the prior neural network model converges, wherein,
the judgment basis of the convergence of the prior neural network model comprises the judgment of whether the time difference value between the estimated arrival time of the flight and the actual arrival time of the flight output by the prior neural network model is within a preset time range.
In an alternative embodiment of the method according to the invention,
the first loss function of the prior neural network model is shown in the following equation:
wherein, the first and the second end of the pipe are connected with each other,
、
respectively representing the number of hidden neurons and output neurons,
、
respectively representing the weight values corresponding to the hidden neuron and the output neuron,
、
representing values input to hidden neurons and output neurons, respectively;
the weight value of each layer of the prior neural network model is shown as the following formula:
wherein the content of the first and second substances,
the duration of the period is indicated,
a loss value representing the ith predicted value and a preset target value at time t, L representing an error function,
indicating the ith predicted value at time t.
In an alternative embodiment of the method according to the invention,
the method for training the flight delay prediction model further comprises the step of training a post-neural network model of the flight delay prediction model, wherein the post-neural network model is constructed on the basis of a convolutional neural network model, the post-neural network model comprises a convolutional layer, a pooling layer and a classification layer,
after local connection and weight sharing extraction are carried out on input feature information through the convolution layer of the posterior neural network, the weight value of the input feature information is calibrated in a self-adaptive mode, wherein the input feature information is the predicted value of flight arrival time output by the prior neural network model;
the pooling layer fuses the input characteristic information and the corresponding weight values into a comprehensive characteristic value;
outputting the probability of the corresponding category of the input characteristic information according to the comprehensive characteristic value through a softmax classifier of the classification layer;
and (3) carrying out loop iteration training on the post-neural network model until the probability of outputting and inputting the corresponding category of the characteristic information meets a preset condition, wherein,
the input characteristic information corresponding category comprises a delay level.
In an alternative embodiment of the method according to the invention,
the method for adaptively calibrating the weight value of the input characteristic information is shown in the following formula:
wherein the content of the first and second substances,
、
the sigmod function and ReLu function are represented separately,
indicates the number of input characteristic information,
represents the weight value corresponding to the kth input characteristic information,
representing the convolved values of the input feature information.
In a second aspect of an embodiment of the present disclosure,
there is provided a flight delay prediction system, the system comprising:
the system comprises a first unit, a second unit and a third unit, wherein the first unit is used for acquiring flight operation information of a target flight, weather information of a take-off and landing city of the target flight in a preset operation time and associated delay information of an airport associated with the target flight;
a second unit, configured to convert the flight operation information, the weather information, and the associated delay information into feature vectors, predict whether the target flight will be delayed through a pre-constructed flight delay prediction model,
a third unit configured to further determine a delay level of the target flight if it is predicted that the target flight will be delayed, and determine a processing scenario corresponding to the delay level according to a correspondence relationship obtained in advance, wherein,
the flight delay prediction model is constructed based on a plurality of neural network models, and the output of the prior neural network model is used as the input of the subsequent neural network model and is used for performing delay prediction on the target flight.
In a third aspect of an embodiment of the present disclosure,
there is provided a flight delay prediction apparatus comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the memory-stored instructions to perform the method of any of the preceding.
In a fourth aspect of an embodiment of the present disclosure,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method of any of the preceding.
The present disclosure provides a flight delay prediction method, comprising
Acquiring flight operation information of a target flight, weather information of a take-off and landing city of the target flight within preset operation time, and associated delay information of an airport associated with the target flight;
through flight operation information, weather information and associated delay information, the most main factors influencing flight delay can be gathered, the accuracy of flight delay prediction is improved, and the fact that data to be processed by a flight delay prediction model is necessary and small in quantity is guaranteed.
After the flight operation information, the weather information and the associated delay information are converted into feature vectors, whether the target flight will be delayed or not is predicted through a pre-constructed flight delay prediction model,
the flight delay prediction model is constructed based on a plurality of neural network models, and the output of the prior neural network model is used as the input of the subsequent neural network model and is used for performing delay prediction on the target flight.
The flight delay model is constructed based on the neural network models, compared with the existing single model for flight delay prediction, the flight delay prediction model is higher in accuracy and higher in stability, and the prediction accuracy after the multiple models are fused is exponentially improved compared with that of the single model, so that the generalization capability is strong.
If the target flight will be delayed, further determining the delay level of the target flight, and determining a processing scheme corresponding to the delay level according to a pre-acquired corresponding relation, wherein,
the method disclosed by the embodiment of the disclosure can not only predict whether the flight is delayed, but also further determine the flight delay level, and set a corresponding scheme according to the flight delay level, thereby providing a solution of the system.
Drawings
FIG. 1 is a schematic flow chart of a flight delay prediction method according to an embodiment of the disclosure;
FIG. 2 is a schematic flow chart of a prior neural network model for training a flight delay prediction model according to an embodiment of the present disclosure;
FIG. 3 is a schematic flow chart of a post neural network model for training a flight delay prediction model according to an embodiment of the present disclosure;
FIG. 4 is a schematic diagram illustrating a probability of a classification layer outputting a corresponding class of feature information according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of a flight delay prediction system according to an embodiment of the disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions in the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present disclosure and in the drawings described above, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in other sequences than those illustrated or described herein.
It should be understood that, in various embodiments of the present disclosure, the sequence numbers of the processes do not mean the execution sequence, and the execution sequence of the processes should be determined by the functions and the inherent logic of the processes, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.
It should be understood that, in the present disclosure, "comprising" and "having" and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements explicitly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that in the present disclosure, "a plurality" means two or more. "and/or" is merely an association relationship describing an associated object, meaning that there may be three relationships, for example, and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "comprising a, B and C", "comprising a, B, C" means that all three of a, B, C are comprised, "comprising a, B or C" means comprising one of a, B, C, "comprising a, B and/or C" means comprising any 1 or any 2 or 3 of a, B, C.
It should be understood that in this disclosure, "B corresponding to a," "a corresponds to B," or "B corresponds to a," means that B is associated with a, from which B may be determined. Determining B from a does not mean determining B from a alone, but may also be determined from a and/or other information. And the matching of A and B means that the similarity of A and B is greater than or equal to a preset threshold value.
As used herein, the term "if" may be interpreted as "at \8230; …" or "in response to a determination" or "in response to a detection" depending on the context.
The technical solution of the present disclosure is explained in detail below with specific examples. These several specific embodiments may be combined with each other below, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 1 is a schematic flow chart of a flight delay prediction method according to an embodiment of the present disclosure, as shown in fig. 1, the method includes:
s101, acquiring flight operation information of a target flight, weather information of a take-off and landing city of the target flight in a preset operation time and associated delay information of an airport associated with the target flight;
in practical application, weather reasons, air traffic management reasons and airline company reasons are main factors causing flight delay, wherein the associated delay information of the target flight associated airport in the embodiment of the disclosure is used for indicating the delay information related to the target flight departure and landing airport, and the associated delay information can visually display direct factors of the target flight delay, so that the accuracy of final flight delay prediction can be improved. For example, if there is a flight delay at the transit airport or the landing airport of the target flight, there is a high probability that the target flight will also be delayed.
Illustratively, the flight operation information of the embodiment of the disclosure may include historical delayed flight information of the target flight, historical departure and landing airports of the target flight, historical departure and landing time of the target flight, flight number of the target flight, and the like;
the weather information of the disclosed embodiment may include historical weather information of the take-off and landing airport, such as minimum temperature, maximum temperature, observation station air pressure, sea level air pressure, height of the observation station, relative humidity, wind direction, wind speed, wind power size, visibility, and the like;
the associated delay information of the embodiment of the present disclosure may include a number of the associated airport, a flight operation condition of the associated airport, whether the associated airport is delayed, and in the case that there is a delay, a flight number, a stop position, and air flow information corresponding to the delayed flight, etc.
It should be noted that, the flight operation information, the weather information, and the associated delay information in the embodiments of the present disclosure are only exemplary illustrations, and do not limit the types and the numbers of the flight operation information, the weather information, and the associated delay information.
Through flight operation information, weather information and associated delay information, the most main factors influencing flight delay can be summarized, the accuracy of flight delay prediction is improved, and the fact that data to be processed by a flight delay prediction model is necessary and small in quantity is guaranteed.
S102, after the flight operation information, the weather information and the associated delay information are converted into feature vectors, predicting whether the target flight will be delayed or not through a pre-constructed flight delay prediction model;
illustratively, the flight delay prediction model of the embodiment of the disclosure is constructed based on a plurality of neural network models, and the output of the preceding neural network model is used as the input of the following neural network model for performing delay prediction on the target flight.
The flight delay model is constructed based on the neural network models, compared with the existing single model for flight delay prediction, the flight delay prediction model is higher in accuracy and higher in stability, and the prediction accuracy after the multiple models are fused is exponentially improved compared with that of the single model, so that the generalization capability is strong.
Before converting flight operation information, weather information and associated delay information into feature vectors, the method further comprises data preprocessing, before inputting the data into the network model, discrete variable data cannot directly enter the model for training, unordered discrete variables can mislead a network content learning mechanism and confuse importance among feature variables, so that the data preprocessing is required, and the method can comprise data cleaning and missing value processing, wherein,
the data washing includes washing dirty data in the data, where the dirty data is error or invalid data, and the data washing in the embodiment of the present disclosure may include missing value washing, format content washing repeated data washing, non-required data and association verification, and the specific type and method of the data washing are not limited in the embodiment of the present disclosure. Data can be standardized through data cleaning, and extra calculation expenses are reduced.
The missing value processing may be implemented by one or more of homogeneous mean interpolation, modeling prediction, mean interpolation, high-dimensional mapping, and multiple interpolation, which is not limited in the embodiments of the present disclosure.
In an alternative embodiment of the method according to the invention,
before inputting the training feature vector into a flight delay prediction model to be trained, the method further comprises:
determining a data preprocessing method corresponding to the training data characteristics according to the training data characteristics of the model training data set, and screening out dirty data in the model training data set to obtain a standard model training data set, wherein the standard model training data set comprises standard flight operation information, standard weather information and standard association delay information;
respectively converting standard flight operation information, standard weather information and standard associated delay information of the standard model training data set into corresponding standard data characteristics;
and scoring the standard flight operation information, the standard weather information and the standard association delay information according to the data type corresponding to the standard data characteristic, and converting the information with the score exceeding a preset threshold value into a training characteristic vector.
Optionally, the method for scoring the data of the standard model training data set according to the data type in the embodiment of the present disclosure may be shown as follows:
wherein the content of the first and second substances,
the amount of data representing the training data set,
、
coordinate values corresponding to data features representing a standard model training data set,
、
the coordinate value corresponding to the standard data characteristic is shown,
、
and coordinate values corresponding to standard deviations of the standard data features are respectively represented.
For example, even if after data preprocessing, not all data in the data of the training set are suitable for being converted into feature vectors, in the embodiment of the present disclosure, according to the data features of the standard model training data set, the data of the standard model training data set is scored according to the data types, the data with the score exceeding a preset threshold value is converted into the training feature vectors, features with the best classification effect are screened out from a large amount of data of the training data set, the dimensions of a feature space are compressed, and features with high precision and high error rate are generated.
In an alternative embodiment of the method according to the invention,
the method further comprises training a flight delay prediction model, and the method for training the flight delay prediction model comprises the following steps:
constructing a model training data set, wherein the model training data set comprises historical flight operation information, historical weather information and historical associated delay information in a preset time period;
converting the model training data set into training characteristic vectors, inputting a flight delay prediction model to be trained, updating the model training data set based on a training output result of the flight delay prediction model to be trained until a preset training convergence condition is reached to finish training the flight delay prediction model to be trained, wherein,
the flight delay prediction model to be trained comprises a preceding neural network model and a following neural network model, the preceding neural network model performs preceding training learning according to the training feature vector and outputs a first output result to the following neural network model, and the following neural network performs following training learning according to the first output result and outputs a training prediction value.
Illustratively, the output value of the preceding neural network model can be used as the input value of the following neural network model, so that the flight delay prediction model can perform secondary learning, the overfitting capability of the model is improved, and the accuracy of the model is ensured.
Optionally, the flight delay prediction model according to the embodiment of the present disclosure may allocate a corresponding weight value to an output value of a preceding neural network model, and allocate a corresponding weight value to an output value of a following neural network model, so as to implement reasonable adjustment on data of different dimensions and different types, and update a model training data set with the output value of the following neural network model after the weight value is allocated, thereby further improving the prediction accuracy of the flight delay prediction model.
In an alternative embodiment of the method according to the invention,
fig. 2 is a schematic flow chart of a prior neural network model for training a flight delay prediction model according to an embodiment of the present disclosure, the prior neural network model including an input layer, a convolutional layer, a hidden layer, and an output layer, as shown in fig. 2,
s201, the convolution layer performs batch normalization processing on output data of the input layer, and determines output characteristics of all hidden layers;
s202, determining a predicted value corresponding to the output feature based on a first loss function of the prior neural network model, judging whether the prior neural network model converges according to the predicted value and a loss value of a preset target value,
and S203, if the hidden layer is not converged, determining the gradient value of the hidden layer through a back propagation algorithm according to the loss value, and updating the weight value of each layer of the prior neural network model according to the gradient value until the prior neural network model is converged.
Illustratively, the preceding neural network model of the embodiments of the present disclosure includes an input layer, a convolutional layer, a hidden layer, and an output layer, the preceding neural network model and the following neural network model of the embodiments of the present disclosure may include a deep neural network, a convolutional neural network, a long-short term memory network, and the like, and the embodiments of the present disclosure do not limit the type of the preceding neural network model. The prior and posterior neural network models of the present disclosure structurally maintain the output layer of the prior neural network model connected with the input layer of the posterior neural network model.
Illustratively, the embodiment of the present disclosure may train a prior neural network model through two directions of forward propagation and backward propagation, and optionally, the convolutional layer performs batch normalization processing on the output data of the input layer, so as to reduce covariate transfer inside the network and improve the generalization capability of the network. Further, a predicted value corresponding to the output feature may be determined according to a first loss function, and whether the preceding neural network model converges may be determined according to the predicted value and a loss value of a preset target value.
In this embodiment of the disclosure, the basis for determining whether the previous neural network model converges may include whether a time difference between the estimated arrival time of the flight and the actual arrival time of the flight output by the previous neural network model is within a preset time range, if yes, convergence of the previous neural network may be determined, and if no, non-convergence of the previous neural network may be determined.
In an alternative embodiment of the method according to the invention,
the first loss function of the prior neural network model is shown in the following equation:
wherein, the first and the second end of the pipe are connected with each other,
、
respectively representing the number of hidden neurons and output neurons,
、
respectively representing the weight values corresponding to the hidden neuron and the output neuron,
、
representing the values input to the hidden neurons and the output neurons, respectively.
If not, determining the gradient value of the hidden layer through a back propagation algorithm according to the loss value, and updating the weight value of each layer of the prior neural network model according to the gradient value until the prior neural network model converges.
The weight value of each layer of the prior neural network model is shown as the following formula:
wherein, the first and the second end of the pipe are connected with each other,
the duration of the period is indicated and,
a loss value representing the ith predicted value and a preset target value at time t, L representing an error function,
indicating the ith predicted value at time t.
Fig. 3 is a schematic flow chart of a post-neural network model for training a flight delay prediction model according to an embodiment of the present disclosure, where the post-neural network model includes a convolution layer, a pooling layer, and a classification layer, as shown in fig. 3,
s301, after local connection and weight sharing extraction are carried out on input feature information through the convolution layer of the posterior neural network, the weight value of the input feature information is calibrated in a self-adaptive mode;
and the input characteristic information is the predicted value of the flight arrival time output by the prior neural network model. Illustratively, by adaptively calibrating the weight values of the input feature information, the importance degree of each feature channel can be automatically learned, and features contributing more strongly to the overall model can be selectively emphasized, and features contributing less strongly or not contributing strongly are suppressed.
In an alternative embodiment of the method according to the invention,
the method for adaptively calibrating the weight value of the input characteristic information is shown in the following formula:
wherein the content of the first and second substances,
、
respectively representing a sigmod function and a ReLu function,
indicates the number of the input characteristic information,
represents the weight value corresponding to the kth input characteristic information,
representing the convolved values of the input feature information.
S302, the pooling layer fuses the input feature information and the corresponding weight values into a comprehensive feature value;
the input characteristic information and the corresponding weight values are fused into the comprehensive characteristic value, so that a large number of redundant parameters caused by intensive connection of the convolutional layers can be reduced, the stability of the model in the operation process is enhanced, the global distribution of the whole characteristic response can be expressed, and the global information of the network is utilized by lower layers.
S303, outputting the probability of the corresponding category of the input characteristic information according to the comprehensive characteristic value through a softmax classifier of the classification layer;
fig. 4 is a schematic diagram of a probability of a classification layer outputting a corresponding category of feature information according to an embodiment of the present disclosure. As shown in fig. 4, taking the parameter dimension of the input classification layer as 9 × 30 as an example, the original parameter dimension is adjusted to 2430 × 1 and 2430 × 5 × 1 and 30 × 5 through two fully connected layers, respectively, and the value after the final adjustment dimension is input to the softmax classifier to output the probability of the corresponding class of the input feature information.
S304, training the post-neural network model in a loop iteration mode until the probability of the corresponding category of the output and input characteristic information meets the preset condition.
The input characteristic information corresponding category comprises delay levels. The flight delay model is constructed based on the neural network models, compared with the flight delay prediction of the existing single model, the flight delay prediction model is higher in accuracy and higher in stability, prediction accuracy is exponentially improved after the multiple models are fused compared with that of the single model, and generalization capability is strong.
S103, if the delay of the target flight is predicted, further determining the delay level of the target flight, and determining a processing scheme corresponding to the delay level according to the corresponding relation acquired in advance;
for example, in the case of predicting the delay of the target flight, the disclosed embodiment may further determine the expected delay time of the target flight, set a corresponding delay level according to the flight delay time,
in this embodiment of the present disclosure, setting a corresponding delay level according to the flight delay time may include:
if the delay time is within the time floating range acceptable by the preset arrival time, determining that the delay is not delayed;
if the delay time is within a range of a first delay time and a second delay time, the delay time can be regarded as a first delay level, wherein the first delay time can comprise 30 minutes, and the second delay time can comprise 60 minutes;
if the delay time is within the range of the second delay time and a third delay time, the delay time can be regarded as a second delay level, wherein the third delay time can comprise 120 minutes;
if the delay time exceeds a third delay time, a third delay level may be identified.
And determining a processing scheme corresponding to the delay level according to the delay level and a pre-acquired corresponding relationship, wherein the pre-acquired corresponding relationship can be used for indicating the corresponding relationship between the delay level and the processing scheme. By way of example, it is possible to provide,
if the delay level is the first delay level, the mileage or the membership points corresponding to the airline company can be given to the passenger;
if the delay level is the second delay level, corresponding rest environments, catering services for passengers and the like can be arranged for the passengers;
if the delay level is the third delay level, the passenger can be arranged with a lodging room, and the passenger who has other flights in the following can be carried out with the flight refund and the like.
The method disclosed by the embodiment of the disclosure can not only predict whether the flight is delayed, but also further determine the flight delay level, and set a corresponding scheme according to the flight delay level, thereby providing a solution of the system.
The present disclosure provides a flight delay prediction method, comprising
Acquiring flight operation information of a target flight, weather information of a take-off and landing city of the target flight within preset operation time, and associated delay information of an airport associated with the target flight;
through flight operation information, weather information and associated delay information, the most main factors influencing flight delay can be summarized, the accuracy of flight delay prediction is improved, and the fact that data to be processed by a flight delay prediction model is necessary and small in quantity is guaranteed.
After the flight operation information, the weather information and the associated delay information are converted into the characteristic vectors, whether the target flight is delayed or not is predicted through a pre-constructed flight delay prediction model,
the flight delay prediction model is constructed based on a plurality of neural network models, and the output of the prior neural network model is used as the input of the subsequent neural network model and is used for performing delay prediction on the target flight.
The flight delay model is constructed based on the neural network models, compared with the flight delay prediction of the existing single model, the flight delay prediction model is higher in accuracy and higher in stability, prediction accuracy is exponentially improved after the multiple models are fused compared with that of the single model, and generalization capability is strong.
If the target flight will be delayed, further determining the delay level of the target flight, and determining a processing scheme corresponding to the delay level according to a pre-acquired corresponding relation, wherein,
the method disclosed by the embodiment of the disclosure can not only predict whether the flight is delayed, but also further determine the flight delay level, and set a corresponding scheme according to the flight delay level, thereby providing a solution of the system.
Fig. 5 is a schematic structural diagram of a flight delay prediction system according to an embodiment of the disclosure, as shown in fig. 5,
there is provided a flight delay prediction system, the system comprising:
a first unit 51, configured to obtain flight operation information of a target flight, weather information of a city where the target flight takes off and lands within a preset operation time, and associated delay information of an airport associated with the target flight;
a second unit 52, configured to convert the flight operation information, the weather information, and the associated delay information into a feature vector, predict whether the target flight will be delayed through a flight delay prediction model that is constructed in advance,
a third unit 53, configured to, if it is predicted that the target flight will be delayed, further determine a delay level of the target flight, and determine, according to a correspondence relationship obtained in advance, a processing scheme corresponding to the delay level, wherein,
the flight delay prediction model is constructed based on a plurality of neural network models, and the output of the prior neural network model is used as the input of the subsequent neural network model and is used for performing delay prediction on the target flight.
In a third aspect of the embodiments of the present disclosure,
there is provided a flight delay prediction apparatus comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the memory-stored instructions to perform the method of any of the preceding.
In a fourth aspect of an embodiment of the present disclosure,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method of any one of the preceding claims.
The present invention may be methods, apparatus, systems and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therein for carrying out aspects of the present invention.
The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as a punch card or an in-groove protruding structure with instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present invention may be assembler instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing an electronic circuit, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA), with state information of computer-readable program instructions, which can execute the computer-readable program instructions.
Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer-readable program instructions may be provided to a processing unit of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processing unit of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It is noted that, unless expressly stated otherwise, all features disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features. Where used, further, preferably, still further and more preferably is a brief introduction to the description of the other embodiment based on the foregoing embodiment, the combination of the contents of the further, preferably, still further or more preferably back strap with the foregoing embodiment being a complete construction of the other embodiment. Several further, preferred, still further or more preferred arrangements of the back tape of the same embodiment may be combined in any combination to form a further embodiment.
It will be appreciated by persons skilled in the art that the embodiments of the invention described above and shown in the drawings are given by way of example only and are not limiting of the invention. The objects of the invention have been fully and effectively accomplished. The functional and structural principles of the present invention have been shown and described in the examples, and any variations or modifications of the embodiments of the present invention may be made without departing from the principles.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present disclosure, and not for limiting the same; while the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present disclosure.
Claims (8)
1. A flight delay prediction method, the method comprising:
acquiring flight operation information of a target flight, weather information of a take-off and landing city of the target flight within preset operation time, and associated delay information of an airport associated with the target flight;
after the flight operation information, the weather information and the associated delay information are converted into feature vectors, whether the target flight will be delayed or not is predicted through a pre-constructed flight delay prediction model,
if the target flight will be delayed, further determining the delay level of the target flight, and determining a processing scheme corresponding to the delay level according to a pre-acquired corresponding relation, wherein,
the flight delay prediction model is constructed based on a plurality of neural network models, and the output of a preceding neural network model is used as the input of a following neural network model and is used for performing delay prediction on the target flight;
the method further includes training a flight delay prediction model,
the method for training the flight delay prediction model further comprises the step of training a prior neural network model of the flight delay prediction model, wherein the prior neural network model is constructed on the basis of a deep neural network model and comprises an input layer, a convolutional layer, a hidden layer and an output layer,
the convolutional layers perform batch normalization processing on the training feature vectors and determine output features corresponding to the hidden layers;
determining a predicted value corresponding to the output feature based on a first loss function of the prior neural network model, and judging whether the prior neural network model converges according to the predicted value and a loss value of a preset target value,
if not, determining the gradient value of the hidden layer through a back propagation algorithm according to the loss value, and updating the weight value of each layer of the prior neural network model according to the gradient value until the prior neural network model converges, wherein,
the judgment basis of the convergence of the prior neural network model comprises the judgment of whether the time difference value between the estimated arrival time of the flight and the actual arrival time of the flight output by the prior neural network model is within a preset time range;
the method for training the flight delay prediction model further comprises the step of training a post-neural network model of the flight delay prediction model, wherein the post-neural network model is constructed on the basis of a convolutional neural network model, the post-neural network model comprises a convolutional layer, a pooling layer and a classification layer,
after local connection and weight sharing extraction are carried out on input characteristic information through the convolution layer of the posterior neural network, the weight value of the input characteristic information is calibrated in a self-adaptive mode, wherein the input characteristic information is the predicted arrival time of the flight output by the prior neural network model;
the pooling layer fuses the input characteristic information and the corresponding weight values into a comprehensive characteristic value;
outputting the probability of the corresponding category of the input characteristic information according to the comprehensive characteristic value through a softmax classifier of the classification layer;
and (3) carrying out loop iteration training on the post-neural network model until the probability of outputting and inputting the corresponding category of the characteristic information meets a preset condition, wherein,
the input characteristic information corresponding category comprises delay levels.
2. The method of claim 1, further comprising training a flight delay prediction model, the method of training a flight delay prediction model comprising:
constructing a model training data set, wherein the model training data set comprises historical flight operation information, historical weather information and historical associated delay information in a preset time period;
converting the model training data set into training characteristic vectors, inputting a flight delay prediction model to be trained, updating the model training data set based on a training output result of the flight delay prediction model to be trained until a preset training convergence condition is reached to finish training the flight delay prediction model to be trained, wherein,
the flight delay prediction model to be trained comprises a preceding neural network model and a following neural network model, the preceding neural network model performs preceding training learning according to the training feature vector and outputs a first output result to the following neural network model, and the following neural network performs following training learning according to the first output result and outputs a training prediction value.
3. The method of claim 2, wherein prior to inputting the training feature vectors into a flight delay prediction model to be trained, the method further comprises:
determining a data preprocessing method corresponding to the training data characteristics according to the training data characteristics of the model training data set, and screening dirty data in the model training data set to obtain a standard model training data set, wherein the standard model training data set comprises standard flight operation information, standard weather information and standard association delay information;
respectively converting the standard flight operation information, the standard weather information and the standard associated delay information of the standard model training data set into corresponding standard data characteristics;
and scoring the standard flight operation information, the standard weather information and the standard association delay information according to the data type corresponding to the standard data characteristic, and converting the information with the score exceeding a preset threshold value into a training characteristic vector.
4. The method of claim 1,
the first loss function of the prior neural network model is shown in the following equation:
wherein the content of the first and second substances,
、
respectively representing the number of hidden neurons and output neurons,
、
respectively representing the weight values corresponding to the hidden neuron and the output neuron,
、
representing values input to hidden neurons and output neurons, respectively;
the weight value of each layer of the prior neural network model is shown as the following formula:
wherein the content of the first and second substances,
the duration of the period is indicated and,
a loss value representing the ith predicted value and a preset target value at time t, L representing an error function,
indicating the ith predicted value at time t.
5. The method of claim 1,
the method for adaptively calibrating the weight value of the input characteristic information is shown by the following formula:
wherein the content of the first and second substances,
、
respectively representing a sigmod function and a ReLu function,
indicates the number of input characteristic information,
represents the weight value corresponding to the kth input characteristic information,
representing the convolved values of the input feature information.
6. A flight delay prediction system, the system comprising:
the system comprises a first unit, a second unit and a third unit, wherein the first unit is used for acquiring flight operation information of a target flight, weather information of a take-off and landing city of the target flight in a preset operation time and association delay information of an airport associated with the target flight;
a second unit, configured to convert the flight operation information, the weather information, and the associated delay information into feature vectors, predict whether the target flight will be delayed through a pre-constructed flight delay prediction model,
a third unit configured to further determine a delay level of the target flight if it is predicted that the target flight will be delayed, and determine a processing scenario corresponding to the delay level according to a correspondence relationship obtained in advance, wherein,
the flight delay prediction model is constructed based on a plurality of neural network models, and the output of a preceding neural network model is used as the input of a following neural network model and is used for performing delay prediction on the target flight;
the system also comprises a fourth unit for training the flight delay prediction model,
the fourth unit is further for training a prior neural network model of the flight delay prediction model,
wherein the prior neural network model is constructed based on a deep neural network model and comprises an input layer, a convolutional layer, a hidden layer and an output layer,
the convolutional layers perform batch normalization processing on the training feature vectors and determine output features corresponding to the hidden layers;
determining a predicted value corresponding to the output feature based on a first loss function of the preceding neural network model, and judging whether the preceding neural network model converges according to the predicted value and a loss value of a preset target value,
if not, determining the gradient value of the hidden layer through a back propagation algorithm according to the loss value, and updating the weight value of each layer of the prior neural network model according to the gradient value until the prior neural network model converges, wherein,
the judgment basis of the convergence of the prior neural network model comprises the steps of judging whether the time difference value between the estimated arrival time of the flight and the actual arrival time of the flight output by the prior neural network model is within a preset time range or not;
the fourth unit is further for training a post neural network model of the flight delay prediction model,
wherein the post-neural network model is constructed based on a convolutional neural network model, the post-neural network model comprises a convolutional layer, a pooling layer and a classification layer,
after local connection and weight sharing extraction are carried out on input characteristic information through the convolution layer of the posterior neural network, the weight value of the input characteristic information is calibrated in a self-adaptive mode, wherein the input characteristic information is the predicted arrival time of the flight output by the prior neural network model;
the pooling layer fuses the input characteristic information and the corresponding weight values into a comprehensive characteristic value;
outputting the probability of the corresponding category of the input characteristic information according to the comprehensive characteristic value through a softmax classifier of the classification layer;
and (3) carrying out loop iteration training on the post-neural network model until the probability of the corresponding category of the output and input characteristic information meets a preset condition, wherein,
the input characteristic information corresponding category comprises delay levels.
7. A flight delay prediction apparatus, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the memory-stored instructions to perform the method of any of claims 1 to 5.
8. A computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211243543.7A CN115310732B (en) | 2022-10-12 | 2022-10-12 | Flight delay prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211243543.7A CN115310732B (en) | 2022-10-12 | 2022-10-12 | Flight delay prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115310732A CN115310732A (en) | 2022-11-08 |
| CN115310732B true CN115310732B (en) | 2022-12-20 |
Family
ID=83867634
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211243543.7A Active CN115310732B (en) | 2022-10-12 | 2022-10-12 | Flight delay prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115310732B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502776B (en) * | 2023-06-27 | 2023-08-25 | 中国民航大学 | Flight recovery modeling method, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109448445A (en) * | 2018-09-05 | 2019-03-08 | 南京航空航天大学 | Flight based on shot and long term Memory Neural Networks is delayed grading forewarning system method |
| CN110503245A (en) * | 2019-07-30 | 2019-11-26 | 南京航空航天大学 | A kind of prediction technique of air station flight large area risk of time delay |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105956621B (en) * | 2016-04-29 | 2019-05-24 | 南京航空航天大学 | A kind of flight delay method for early warning based on evolution sub- sampling integrated study |
| AU2018241119A1 (en) * | 2017-10-06 | 2019-05-02 | Tata Consultancy Services Limited | System and method for flight delay prediction |
| GB2584625A (en) * | 2019-05-28 | 2020-12-16 | Sita Information Networking Computing Uk Ltd | System and method for flight arrival time prediction |
| CN111950791A (en) * | 2020-08-14 | 2020-11-17 | 中国民航信息网络股份有限公司 | Flight delay prediction method, device, server and storage medium |
| CN112308285B (en) * | 2020-09-16 | 2022-05-10 | 北京中兵数字科技集团有限公司 | Information processing method, information processing device, electronic equipment and computer readable storage medium |
| CN113657671B (en) * | 2021-08-18 | 2024-02-09 | 北京航空航天大学 | Flight delay prediction method based on ensemble learning |
-
2022
- 2022-10-12 CN CN202211243543.7A patent/CN115310732B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109448445A (en) * | 2018-09-05 | 2019-03-08 | 南京航空航天大学 | Flight based on shot and long term Memory Neural Networks is delayed grading forewarning system method |
| CN110503245A (en) * | 2019-07-30 | 2019-11-26 | 南京航空航天大学 | A kind of prediction technique of air station flight large area risk of time delay |
Non-Patent Citations (2)
| Title |
|---|
| Generalized Flight Delay Prediction Method Using Gradient Boosting Decision Tree;Fan Liu 等;《2020 IEEE 91st Vehicular Technology Conference》;20200630;第1-5页 * |
| 基于VMD-MD-Clustering方法的航班延误等级分类;王兴隆 等;《交通信息与安全》;20220630(第3期);第171-178页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115310732A (en) | 2022-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11119250B2 (en) | Dynamic adaption of vessel trajectory using machine learning models | |
| US10636293B2 (en) | Uncertainty modeling in traffic demand prediction | |
| JP2023510879A (en) | Route planning method, device, equipment, and computer storage medium | |
| WO2018209576A1 (en) | Systems and methods for digital route planning | |
| CN107818396A (en) | For changing the decision assistant of flight plan | |
| Choi et al. | Artificial neural network models for airport capacity prediction | |
| CN115239026B (en) | Method, system, device and medium for optimizing parking space allocation | |
| WO2022193497A1 (en) | Method and system for graph-based panoptic segmentation | |
| US20150253463A1 (en) | Generating an Ensemble of Forecasting Models | |
| CN113177968A (en) | Target tracking method and device, electronic equipment and storage medium | |
| US10599976B2 (en) | Update of attenuation coefficient for a model corresponding to time-series input data | |
| CN115310732B (en) | Flight delay prediction method and system | |
| CN112862171B (en) | Flight arrival time prediction method based on space-time neural network | |
| CN114519932A (en) | Regional traffic condition integrated prediction method based on space-time relation extraction | |
| CN114254840B (en) | Data processing method, electronic equipment and storage medium | |
| CN115392752B (en) | Subway short-time passenger flow prediction method, system, electronic equipment and storage medium | |
| Soares et al. | Incremental gaussian granular fuzzy modeling applied to hurricane track forecasting | |
| US20200349416A1 (en) | Determining computer-executed ensemble model | |
| CN116434538A (en) | Urban traffic flow prediction model construction method based on heterogeneous data fusion | |
| US20200073014A1 (en) | Generation of weather analytical scenarios translating likely airport capacity impact from probabilistic weather forecast | |
| CN112365037A (en) | Airport airspace flow prediction method based on long-term and short-term data prediction model | |
| CN115359062B (en) | Method and system for dividing and calibrating monitoring target through semi-supervised example | |
| CN114972429B (en) | Target tracking method and system for cloud edge cooperative self-adaptive reasoning path planning | |
| CN115423038A (en) | Method, apparatus, electronic device and storage medium for determining fairness | |
| Ayaydın et al. | Deep learning based forecasting of delay on flights |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |