Flight seasonal classification prediction method, apparatus, and machine-readable medium

Abstract

The present application provides a method for predicting flight seasonal classification and related equipment, which improves the accuracy of seasonal classification and avoids the phenomenon of over-bias caused by seasonal classification for artificial flights. The method includes the steps of: obtaining a target departure date corresponding to a target flight from a local database; constructing N data pools corresponding to the target flight; determining a first set of past flights corresponding to the target flight according to the target departure date, where each first past flight in the first set of past flights is a flight that is not seasonally classified; obtaining flight data of each first past flight in the first set of past flights; determining a weight value of each first past flight according to the N data pools and the flight data of each first past flight; and determining a seasonal classification of the target flight according to the weight value of each first past flight and the number of days until departure of each first past flight.

Description

Technical Field

[0001] This application claims the priority of a Chinese patent application filed with the China National Intellectual Property Administration on May 31, 2021, with an application number of 202110604811.2 and an invention title of "Method, Apparatus, and Machine-readable Medium for Predicting Flight Seasonal Classification", and all of its content is incorporated herein by reference.

[0002] This application relates to the field of aviation, and in particular, to a method, apparatus, and machine-readable medium for predicting flight seasonal classification.

Background Art

[0003] Due to the periodic influence of factors such as climatic conditions, unexpected events, industrial and agricultural production and life, customs such as holidays, and national economic development, the passenger and cargo transportation volume of the civil aviation industry fluctuates seasonally. In the field of air transportation, the classification of peak seasons and off-seasons for past flights departing is also called seasonal classification.

Summary of the Invention

Problems to be Solved by the Invention

[0004] Currently, the seasonal classification of air flights is mainly carried out quantitatively or qualitatively manually. However, the determination of the seasonal classification of air flights by manual means is prone to a result of being too biased, and the accuracy of the seasonal classification of air flights is low.

[0005] This application provides a method, apparatus, and machine-readable medium for predicting flight seasonal classification, which improves the accuracy of seasonal classification and avoids the phenomenon of being too biased in the seasonal classification of manual flights.

Means for Solving the Problems

[0006] The first aspect of the embodiment of this application provides a method for predicting flight seasonal classification, A step of obtaining a target departure date corresponding to a target flight from a local database, wherein the target flight is an undeparted flight of the target airline that is subject to seasonal classification forecasting. A step of constructing N data pools corresponding to the target flight, wherein N is a positive integer of 2 or more, A step of determining a first set of past flights corresponding to the target flight based on the target departure date, wherein each of the first past flights in the first set of past flights is a flight that is not classified seasonally, The steps include obtaining flight data for each first past flight in the first set of past flights, A step of determining the weight value of each first past flight based on the N data pools and the flight data of each first past flight, The process includes the step of determining the seasonal classification of the target flight based on the weighted values ​​of each of the first past flights and the number of days until departure of each of the first past flights.

[0007] A second aspect of this application provides a flight seasonal classification and prediction device. An acquisition unit that obtains a target departure date corresponding to a target flight from a local database, wherein the target flight is an acquisition unit that is an undeparted flight of a target airline that is subject to seasonal classification prediction, A construction unit that constructs N data pools corresponding to the target flight, wherein N is a positive integer of 2 or more, A first decision unit that determines a first set of past flights corresponding to the target flight based on the target departure date, wherein each of the first past flights in the first set of past flights is a flight that is not classified seasonally, A second decision unit that determines the weighted value of each first past flight based on the N data pools and the flight data of each first past flight, A third determination unit that determines the seasonal classification of the target flight based on the weighted values ​​of each of the first past flights and the number of days until departure of each of the first past flights, The acquisition unit further acquires flight data for each of the first past flights in the first set of past flights.

[0008] In a possible design, the third decision unit specifically, Based on the weighted values ​​of each of the aforementioned first past flights, the sum of the weighted values ​​of the first set of past flights is determined. The total number of days for the first set of past flights is determined based on the number of days until departure for each of the first past flights. The seasonal classification of the target flight is determined based on the aforementioned weighted sum and the aforementioned total number of days.

[0009] In a possible design, the third decision unit determines the seasonal classification of the target flight based on the weighted sum and the total number of days, A step of comparing the weighted sum and the total number of days and obtaining a comparison result, wherein the comparison result indicates the relationship between the magnitude of the weighted sum and the total number of days, The procedure includes the step of determining the seasonal classification of the target flight based on the comparison results.

[0010] In a possible design, the second decision unit specifically: Based on the flight data of each of the aforementioned first past flights, revenue data for each of the aforementioned first past flights is determined. The first revenue data is divided into a first data pool, the first revenue data is the revenue data corresponding to any one of the first past flights, and the first data pool is the data pool among the N data pools that has the minimum distance between the central data and the first revenue data. Determine the predetermined weighting value of the first data pool, The predetermined weighting value of the first data pool is determined to be the weighting value of the flight corresponding to the first revenue data.

[0011] In a possible design, the first decision unit specifically, The first, second, third, and fourth due dates are calculated using the following formulas: The aforementioned first date = the aforementioned target departure date - 52 * 7 * i - 1; The second date mentioned above = the target departure date - 52 * 7 * i + 1; The third date mentioned above = the target departure date - 51 * 7 * i; The fourth date mentioned above = the target departure date - 53 * 7 * i; Among them, i = (1, 2, 3..., n), and the aforementioned i is the year before the current year. Of the first, second, third, and fourth dates mentioned above, the flight corresponding to the target flight is determined to be in the first set of past flights.

[0012] In a possible design, the first decision unit more specifically: The target history reference date is calculated using the following formula: The aforementioned target history reference date = the aforementioned target departure date - 52 * 7 * i; where i = (1, 2, 3..., n), i is the year before the current year. Based on the target history reference date, determine the first due date, the second due date, the third due date, and the fourth due date. The first due date is the day before the target history reference date, the second due date is the day after the target history reference date, the third due date is the week before the target history reference date, and the fourth due date is the week after the target history reference date. Determine the flight corresponding to the target flight among the first due date, the second due date, the third due date, and the fourth due date as the first set of past flights.

[0013] In a possible design, the construction unit specifically Step 1: Obtain the flight data of each second past flight in the second set of past flights corresponding to the target flight. Step 2: Calculate the revenue data of each second past flight based on the flight data of each second past flight. Step 3: Determine the second revenue data as the central data of the second data pool. The second revenue data is the revenue data corresponding to the first flight, the first flight is any one flight in the second set of past flights, and the second data pool is any one of the N data pools. Step 4: Calculate the distance between the third revenue data and the second revenue data. The third revenue data is the revenue data corresponding to any one flight in the flight subset, and the flight subset is the flight set obtained by removing the first flight from the second set of past flights. Step 5: Divide the fourth revenue data into the second data pool. The fourth revenue data is the revenue data corresponding to the second flight, and the second flight is the flight corresponding to the revenue data closest to the second revenue data in the flight subset. Execute Step 6: Calculate the central data of the divided second data pool. Repeat steps 3 to 6 until all the revenue data corresponding to each second past flight in the second past flight set are divided into the N data pools.

[0014] The third aspect of the present application provides a computer device, including a memory, a processor and a bus system. The memory stores a program, the bus system connects the memory and the processor and enables communication between the memory and the processor, and the processor executes the program in the memory to execute the flight seasonal classification prediction method described in the first aspect above based on the instructions in the program code.

[0015] The fourth aspect of the embodiment of the present application provides a machine-readable medium including instructions, which when executed by a machine, cause the machine to execute the steps of the flight seasonal classification prediction method described in each of the above aspects.

Advantages of the Invention

[0016] As can be seen from the above, in the embodiments provided by the present application, when determining the seasonal classification of the target flight, the flight seasonal classification prediction device obtains the target departure date of the target flight, constructs and modifies N data pools corresponding to the target flight, determines the first past flight set based on the target departure date, determines the weighted value of each first past flight based on the flight data of each first past flight in the first past flight set, and finally determines the seasonal classification of the target flight based on the weighted value of the first past flight and the number of days until the departure of each first past flight, improving the accuracy of the seasonal classification compared to the conventional manual seasonal classification of flights and avoiding the phenomenon of being too biased by the seasonal classification of manual flights.

Brief Description of the Drawings

[0017] By combining the drawings and referring to the following specific embodiments, the above and other features, advantages, and aspects of each embodiment of this application will become clearer. In the drawings, similar or identical reference numerals indicate similar or identical elements. Herein, the drawings are schematic, and the originals and elements are not necessarily drawn to scale. [Figure 1] This is a schematic flowchart of the flight seasonal classification prediction method provided in the embodiment of this application. [Figure 2] This is a schematic diagram of the virtual structure of a flight seasonal classification and prediction device provided by the embodiment of this application. [Figure 3] This is a schematic diagram of the structure of a machine-readable medium provided in the embodiment of this application. [Figure 4] This is a schematic diagram of the hardware structure of the server provided in the embodiment of this application. [Modes for carrying out the invention]

[0018] The embodiments of this application will be described in more detail below with reference to the drawings. The drawings show several embodiments of this application, but this application is not limited to the embodiments described herein and may be realized in various forms, and these embodiments will provide a more thorough and complete understanding of this application. Herein, the drawings and embodiments of this application are not limiting to the scope of protection of this application and are merely illustrative.

[0019] In this application, the term "inclusion" and its variations are open inclusions, meaning "inclusions, but not limitations." The term "based on" means "based at least partially." The term "one embodiment" refers to "at least one embodiment," the term "another embodiment" refers to "at least one other embodiment," and the term "several embodiments" refers to "at least several embodiments." Relevant definitions of other terms are given below.

[0020] Herein, the concepts of “first,” “second,” etc., as used in this application are merely for distinguishing different devices, modules, or units, and do not limit the order or interdependence of the functions performed by these devices, modules, or units.

[0021] Here, the modifications of “one” and “multiple” as used in this application are not restrictive but general, and as a person skilled in the art would understand, unless otherwise explicitly stated in the specification, they shall be understood as “one or more.”

[0022] First, we will explain the terminology specific to the embodiments of this application.

[0023] The revenue management system uses flight plan, inventory, departure, and fare data to automatically manage the inventory of undeparted flights based on a predictive and optimization model.

[0024] Market demand is the demand where passengers possess both the ability to purchase and actual purchase demand. As an output value in the revenue management system, it may or may not represent actual orders.

[0025] Number of department days (NDO): The system date (i.e., current date) is the number of days until departure of the flight segment of the flight. For example, if the current date is April 26, 2021, and the departure date is May 1, 2021, then the NDO is 5 days.

[0026] Data collection points (Dcp): Determined by the number of days until departure, each point corresponds to a specific number of days until departure. For example, there are 24 such data collection points (Dcp1, Dcp2, ..., Dcp 24 ) is set up, the number of days until departure is set to 365, and each data acquisition point corresponds to the number of days until departure, and data acquisition point Dcp1 corresponds to the number of days until departure of 365, Dcp 24This corresponds to the number of days until departure, and here, the number of data acquisition points, the number of days until departure, and the correspondence between the data acquisition points and the number of days until departure are merely illustrative and not specifically limited.

[0027] DOW: Day of Week, broadly speaking, refers to the days of the week.

[0028] The following describes the flight seasonal classification prediction method provided in this application from the perspective of a flight seasonal classification prediction device, referring to Figure 1, which is a schematic flowchart of the flight seasonal classification prediction method provided in the embodiment of this application, and includes the following steps: 101: Retrieve the target departure date corresponding to the target flight from the local database.

[0029] In this embodiment, the flight seasonal classification prediction device obtains a target departure date corresponding to a target flight from a local database, and the target flight is an undeparted flight of the target airline that is subject to seasonal classification prediction. Here, the flight data of a predetermined airline (i.e., the target airline) is stored in a local database, and the flight control system contains the total or incremental flight data of the predetermined airline and is configured to obtain the total or incremental flight data of the predetermined airline from the flight control system at predetermined intervals, for example, once every 24 hours. The flight data includes, but is not limited to, the flight number, flight departure date, DCP and the number of days until the corresponding flight departure, each cabin reservation value and the corresponding fare value. Therefore, the flight seasonal classification prediction device can directly obtain the target departure date corresponding to the target flight from the local database.

[0030] Here, the above explanation used a target flight as an example, but of course, it would also be possible to explain directly using a flight segment as an example. The target flight includes at least one flight segment, and the flight segment is a flight segment that can constitute a passenger's itinerary. For example, if the target flight corresponds to a Beijing-Shanghai-San Francisco flight, then there are three possible passenger itineraries: Beijing-Shanghai, Shanghai-San Francisco, and Beijing-San Francisco. That is, the target flight includes three flight segments: the Beijing-Shanghai flight segment, the Shanghai-San Francisco flight segment, and the Beijing-San Francisco flight segment.

[0031] 102: Build N data pools corresponding to the target flights.

[0032] In this embodiment, the flight seasonal classification prediction device constructs N data pools corresponding to target flights, where N is a positive integer greater than or equal to 2. In this application, the number of N is set to 7, and the seven data pools are denoted as Peak Season 1 (referred to as peak), Peak Season 2 (referred to as peak1), Off-Peak Season 1 (referred to as peak2), Off-Peak Season 2 (referred to as offpeak2), Off-Season 1 (referred to as offpeak1), Off-Season 2 (referred to as offpeak), and the unclassified data pool default. Hereinafter, the number of data pools and the classification of the data pools are merely illustrative and not specifically limited.

[0033] In one embodiment, the step of a flight seasonal classification predictor constructing N data pools corresponding to target flights includes the following steps: Step 1: Obtain flight data for each second historical flight in the second historical flight set corresponding to the target flight. In this step, the flight seasonal classification predictor first retrieves flight data for each second historical flight in a second historical flight set corresponding to the target flight from the local database. The second historical flight set corresponding to the target flight is a set of flights that departed on the current date during the three years prior to the current date (of course, it may be any other period, such as four years, and is not specifically limited). For example, if the target flight is a flight on April 25, 2021, and April 25 is a Sunday, then the second historical flight set is a set of flights from all Sundays in the past three years that correspond to the target flight. Step 2: Calculate revenue data for each second historical flight based on the flight data for each second historical flight. In this step, the flight seasonal classification predictor calculates using the following formula:

number

number

number

number

number

number

number

number

number

number

number

number

number

[0034] Here, market demand value is the demand where passengers possess purchasing ability and actual purchasing demand, and actual orders may or may not occur. In the revenue management system, the market demand value is output using a restricted algorithm calculation model. This market demand value is applied to market trend determination, market peak season / off-season classification, and is an important input value applied to the revenue management system core algorithm subsystem optimization module. The revenue management system is an automated management system in which airlines automatically manage inventory of undeparted flights using flight plan, inventory, departure, and fare data. Here, the method of obtaining the market demand value for flights is not specifically limited. For example, flight information for a given flight of a target airline is obtained, and inventory data is obtained based on the flight information. This inventory data specifically includes inventory information for departed flights and inventory information for undeparted flights. The inventory data for departed flights is the flight inventory data for the given flight over the past three years based on the current date, and the inventory data for undeparted flights is the flight inventory data for the given flight over the next year based on the current date. Based on the inventory data, the sales status of a given cabin for a given flight is determined. Based on data acquisition points corresponding to the target airline, flight information of the specified airline, and inventory data for the specified flight, the system recognizes the sales status of a specified cabin on a specified flight. The sales status includes cabin lock, cabin open, etc. Finally, the sales status is processed based on a predetermined algorithm to obtain the market demand value for the specified flight.

[0035] Here, if the number of available seats in a designated cabin is 0 or less and the available state is closed, the designated cabin is not for sale, i.e., cabin locked. If the number of available seats in a designated cabin is greater than 0 and the available state is closed, the designated cabin is not for sale, i.e., cabin locked. If the number of available seats in a designated cabin is greater than 0 and the available state is open, the designated cabin is available for sale, i.e., cabin open.

[0036] The following explains how to process the sales status based on a predetermined algorithm and obtain the period demand value for a given flight. If the sales status of data acquisition point DCP(n+1) is cabin open, and the sales status of data acquisition point DCP(n) is cabin open, When the number of reservations for data acquisition point (n) increases relative to data acquisition point (n+1), the market demand value for data acquisition point DCP(n+1) is calculated using the following formula: Market demand value DCP(n+1) = Market demand value DCP(n) + Reserved value increase / change value DCP(n); of which, Reserved value increase / change value = Actual reserved value DCP(n+1) - Actual reserved value DCP(n).

[0037] When the number of reservations for data acquisition point (n) decreases relative to data acquisition point (n+1), the market demand value for data acquisition point DCP(n+1) is calculated using the following formula: Market demand value DCP(n+1) = Market demand value DCP(n) + Reserve reduction change value DCP(n); of which, Reserve reduction change value = (Actual reserved value DCP(n+1) x Market demand value(n)) / Actual reserved value DCP(n) - Market demand value(n).

[0038] If the sales status of data acquisition point DCP(n+1) is cabin locked, and the sales status of data acquisition point DCP(n) is cabin open, When the number of reservations for data acquisition point (n) decreases relative to data acquisition point (n+1), the calculation is performed using the following formula: Market demand value DCP(n+1) = Market demand value DCP(n) + Reserve reduction change value DCP(n); of which, Reserve reduction change value = (Actual reserved value DCP(n+1) * Market demand value(n)) / Actual reserved value DCP(n) - Market demand value(n). Here, the above calculation of market demand value is an iterative process, that is, the market demand value DCP(1) is equal to the actual reserved value, and the market demand value DCP+=1 is calculated iteratively.

[0039] 103: Determine the first set of past flights corresponding to the target flight based on the target departure date.

[0040] In this embodiment, the flight seasonal classification and prediction device determines a first set of past flights corresponding to the target flight based on the target departure date. Specifically, it is determined in the following two ways: 1. Calculate the first, second, third, and fourth due dates using the following formula: First date = Target departure date - 52 * 7 * i - 1; Second date = Target departure date - 52 * 7 * i + 1; Third date = Target departure date - 51 * 7 * i; The fourth date = the aforementioned target departure date - 53 * 7 * i; Among these, i = (1, 2, 3..., n), where i is the year before the current year. For example, if the current year is 2021, i may be any of the years before 2021, such as 2020, 2019, 2018, etc. The flight corresponding to the target flight from the first, second, third, and fourth deadlines is determined as the first set of past flights. An example is given below: For example, if the target departure date is December 30, 2020, then the first date is December 31, 2019, the second date is January 2, 2020, and the third and fourth dates are the same DOW dates as the target departure date, with the third date being December 25, 2019, and the fourth date being January 8, 2020. The first set of past flights is the set of flights corresponding to the target flight from December 31, 2019, January 2, 2020, December 25, 2019, and January 8, 2020.

[0041] 2. Determine the first set of past flights corresponding to the target flight by calculating the target history reference date: The target historical reference date is calculated using the following formula: Target history reference date = Target departure date - 52 * 7 * i; where i = (1, 2, 3..., n), i is the year before the current year; Based on the target history reference date, the first, second, third, and fourth due dates are determined, wherein the first due date is the day before the target history reference date, the second due date is the day after the target history reference date, the third due date is the week before the target history reference date, and the fourth due date is the week after the target history reference date; The flight corresponding to the target flight from the first, second, third, and fourth deadlines is determined to be the first set of past flights. An example is given below to illustrate this. For example, the target departure date is December 30, 2020, the target history reference date = 2020 / 12 / 30 - 52 * 7 * 1 = 2020 / 1 / 1, the corresponding DOW is Wednesday, the date the day before the target history reference date is December 31, 2019 (i.e., the first date is December 31, 2019), and the date the day after the target history reference date is January 2, 2020 (i.e., the second date is January 2, 2020). The date for the same DOW in the week prior to the reference date is December 25, 2019 (i.e., the third date is December 25, 2019), the date for the same DOW in the week following the reference date is January 8, 2020 (i.e., the fourth date is January 8, 2020), and the first past flight set is the set of flights corresponding to the target flight from December 31, 2019, January 2, 2020, December 25, 2019, and January 8, 2020.

[0042] 104: Retrieve the flight data for each of the first past flights in the first set of past flights.

[0043] In this embodiment, after determining a first set of past flights, the flight seasonal classification forecaster obtains flight data for each first past flight in the first set of past flights from the local database. Here, step 101 obtains the target departure date, step 102 constructs N data pools, and steps 103 to 104 obtain flight data for each first past flight in the first set of past flights. However, the execution order between steps 101, 102, and steps 103 to 104 is not limited; step 101 may be executed first, step 102 may be executed first, steps 103 to 104 may be executed first, or simultaneously, and is not specifically limited.

[0044] 105: Determine the weights for each first past flight based on N data pools and the flight data for each first past flight.

[0045] In this embodiment, the flight seasonal classification prediction device determines the weight value of each first past flight based on N data pools and the flight data of each first past flight. Specifically, it first determines the revenue data for each first past flight based on the flight data of each first past flight (the calculation of revenue data has already been explained in detail in step 102 above, so it will not be elaborated here), then divides the first revenue data into a first data pool, where the first revenue data corresponds to any one of the first past flights, and the first data pool is the data pool with the smallest distance between the central data and the first revenue data among the N data pools. It then determines a predetermined weight value for the first data pool and sets the predetermined weight value of the first data pool as the weight value of the flight corresponding to the first revenue data. In other words, it divides the revenue data corresponding to each first past flight into the data pool with the closest distance and sets the predetermined weight value of the corresponding data pool as the weight value of each first past flight.

[0046] Here, the local database stores a predetermined weight value for each of the N data pools by default. In this application, N is set to 7 (of course, it may be any other number and is not specifically limited), and the predetermined weight values ​​for the seven data pools are set to peak=3, peak1=2, peak2=1, default=0, off-peak2=-1, off-peak1=-2, and off-peak=-3, respectively. After dividing the revenue data corresponding to each first past flight into seven numerical values, the predetermined weight value of the data pool is determined to be the weight value of the first past flight divided into that data pool. Here, if there is a seasonally classified flight in the first set of past flights, the weight value of that flight is determined to be 0.

[0047] 106: Determine the seasonal classification of the target flight based on the weighted value of each first past flight and the number of days until departure of each first past flight.

[0048] In this embodiment, the flight seasonal classification predictor determines a first set of weighted past flights based on the weighted values ​​of each first past flight, determines the total number of days in the first set of past flights based on the number of days until departure of each first past flight, and finally determines the seasonal classification of the target flight based on the weighted sum and the total number of days. In one embodiment, the step of the flight seasonal classification predictor determining the seasonal classification of the target flight based on the weighted sum and the total number of days includes the step of comparing the weighted sum and the total number of days to obtain a comparison result, wherein the comparison result indicates the relationship between the magnitude of the weighted sum and the total number of days, and the step of determining the seasonal classification of the target flight based on the comparison result.

[0049] In this embodiment, the weighted sum of the first past flight set is used as the field.

number

number

number

number

number

number

number

number

number

number

[0050] As can be seen from the above, in the embodiment provided in this application, when determining the seasonal classification of a target flight, the flight seasonal classification prediction device obtains the target departure date of the target flight, constructs and modifies N data pools corresponding to the target flight, determines a first set of past flights based on the target departure date, determines the weight value of each first past flight based on the flight data of each first past flight in the first set of past flights, and finally determines the seasonal classification of the target flight based on the weight value of the first past flight and the number of days until departure of each first past flight, thereby improving the accuracy of seasonal classification compared to conventional seasonal classifications for artificial flights and avoiding the phenomenon of excessive bias due to seasonal classifications for artificial flights.

[0051] The flowcharts and block diagrams in the drawings illustrate the system architectures, functions, and operations that can be implemented by the systems, methods, and computer programs of various embodiments of this application. In this regard, each block in the flowchart or block diagram represents a module, program segment, or part of code, and such module, program segment, or part of code contains one or more executable instructions that realize a predetermined logical function. In some alternative implementations, the functions described in the blocks may occur in a different order than in the drawings. For example, two blocks that are actually shown consecutively may be executed basically in parallel, or in some cases in reverse order, which is determined by the functions involved. Furthermore, each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a specialized system of hardware that performs a predetermined function or operation, or by a combination of specialized hardware and computer instructions.

[0052] The names of messages or information that interact between multiple devices in the embodiments of this application are for illustrative purposes only and do not limit the scope of such messages or information.

[0053] Furthermore, while each operation is described in a specific sequence, these operations are not necessarily required to be executed in the specific order or sequence indicated. In certain environments, multitasking and parallel processing are advantageous.

[0054] Herein, the steps described in the method embodiments of this application may be performed in a different order and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the performance of the indicated steps. The scope of this application is not limited in this respect.

[0055] Computer program code that performs the operations of this application is written in one or more program design languages, or a combination thereof, and the program design languages ​​include, but are not limited to, object-oriented program design languages ​​such as Java, Smalltalk, C++, or conventional procedural program design languages ​​such as the "C" language or similar programs. The program code may run entirely on the user's computer, partially on the user's computer, as a separate package of software, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer or to an external computer via any type of network, including a local area network (LAN) or a wide area network (WAN) (for example, connected via the Internet using an Internet service provider).

[0056] The embodiments of this application have been described above from the perspective of a flight seasonal classification prediction method, and the embodiments of this application will now be described below from the perspective of a flight seasonal classification prediction device.

[0057] Referring to Figure 2, which is a schematic diagram of a virtual structure of a flight seasonal classification prediction device provided by an embodiment of this application, the flight seasonal classification prediction device 200 is, Acquisition unit 201 that obtains a target departure date corresponding to a target flight from a local database, wherein the target flight is an undeparted flight of a target airline that is subject to seasonal classification prediction, A construction unit 202 that constructs N data pools corresponding to the target flight, wherein N is a positive integer of 2 or more, A first decision unit 203 that determines a first set of past flights corresponding to the target flight based on the target departure date, wherein each of the first past flights in the first set of past flights is a flight that is not classified seasonally, A second determination unit 204 determines the weighted value of each of the first past flights based on the N data pools and the flight data of each of the first past flights, A third determination unit 205 that determines the seasonal classification of the target flight based on the weighted values ​​of each of the first past flights and the number of days until departure of each of the first past flights, The acquisition unit 201 further acquires flight data for each of the first past flights in the first set of past flights.

[0058] In a possible design, the third decision unit 205 specifically: Based on the weighted values ​​of each of the aforementioned first past flights, the sum of the weighted values ​​of the first set of past flights is determined. The total number of days for the first set of past flights is determined based on the number of days until departure for each of the first past flights. The seasonal classification of the target flight is determined based on the aforementioned weighted sum and the aforementioned total number of days.

[0059] In a possible design, the third decision unit 205 determines the seasonal classification of the target flight based on the weighted sum and the total number of days, A step of comparing the weighted sum and the total number of days and obtaining a comparison result, wherein the comparison result indicates the relationship between the magnitude of the weighted sum and the total number of days, The procedure includes the step of determining the seasonal classification of the target flight based on the comparison results.

[0060] In a possible design, the second decision unit 204 specifically, Based on the flight data of each of the aforementioned first past flights, revenue data for each of the aforementioned first past flights is determined. The first revenue data is divided into a first data pool, the first revenue data is the revenue data corresponding to any one of the first past flights, and the first data pool is the data pool among the N data pools that has the minimum distance between the central data and the first revenue data. Determine the predetermined weighting value of the first data pool, The predetermined weighting value of the first data pool is determined to be the weighting value of the flight corresponding to the first revenue data.

[0061] In a possible design, the first decision unit 203 specifically, The first, second, third, and fourth due dates are calculated using the following formulas: The aforementioned first date = the aforementioned target departure date - 52 * 7 * i - 1; The second date mentioned above = the target departure date - 52 * 7 * i + 1; The third date mentioned above = the target departure date - 51 * 7 * i; The fourth date mentioned above = the target departure date - 53 * 7 * i; Among them, i = (1, 2, 3..., n), and the aforementioned i is the year before the current year. Of the first, second, third, and fourth dates mentioned above, the flight corresponding to the target flight is determined to be in the first set of past flights.

[0062] In a possible design, the first decision unit 203 more specifically, The target history reference date is calculated using the following formula: The aforementioned target history reference date = the aforementioned target departure date - 52 * 7 * i; where i = (1, 2, 3..., n), i is the year before the current year. Based on the aforementioned target history reference date, the first, second, third, and fourth due dates are determined, wherein the first due date is the day before the aforementioned target history reference date, the second due date is the day after the aforementioned target history reference date, the third due date is the week before the aforementioned target history reference date, and the fourth due date is the week after the aforementioned target history reference date. Of the first, second, third, and fourth dates mentioned above, the flight corresponding to the target flight is determined to be in the first set of past flights.

[0063] In a possible design, the construction unit 202 specifically, Step 1 involves obtaining flight data for each second past flight in a second set of past flights corresponding to the target flight, Step 2 involves calculating revenue data for each of the second past flights based on the flight data for each of the second past flights, Step 3 is to determine the second revenue data to be the central data of the second data pool, wherein the second revenue data is revenue data corresponding to the first flight, the first flight is any one flight in the second set of past flights, and the second data pool is any one of the N data pools. Step 4, which calculates the distance between the third revenue data and the second revenue data, wherein the third revenue data is revenue data corresponding to any one flight in the flight subset, and the flight subset is the set of flights from the second set of past flights excluding the first flight. Step 5, which involves dividing the fourth revenue data into the second data pool, wherein the fourth revenue data is revenue data corresponding to a second flight, and the second flight is the flight in the flight subset that corresponds to the revenue data closest to the second revenue data. Step 6 is performed to calculate the central data of the divided second data pool, Steps 3 to 6 are repeated until the revenue data corresponding to each second past flight in the second set of past flights is divided into the N data pools.

[0064] As can be seen from the above, in the embodiment provided in this application, when determining the seasonal classification of a target flight, the flight seasonal classification prediction device obtains the target departure date of the target flight, constructs and modifies N data pools corresponding to the target flight, determines a first set of past flights based on the target departure date, determines the weight value of each first past flight based on the flight data of each first past flight in the first set of past flights, and finally determines the seasonal classification of the target flight based on the weight value of the first past flight and the number of days until departure of each first past flight, thereby improving the accuracy of seasonal classification compared to conventional seasonal classifications for artificial flights and avoiding the phenomenon of excessive bias due to seasonal classifications for artificial flights.

[0065] Here, the units described in the embodiments of this application may be implemented in software form or in hardware form. In some cases, the name of the unit is not limited to the unit itself; for example, the acquisition unit may be further described as "a unit for acquiring the certificate information of a target user."

[0066] In this specification, the functions described above may be performed by at least partially one or more hardware logic components. For example, non-limiting types of hardware logic components that can be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), and complex-programmable logic devices (CPLDs).

[0067] Referring to Figure 3, Figure 3 is a schematic diagram of an embodiment of a machine-readable medium provided in the embodiment of this application.

[0068] As shown in Figure 3, this embodiment provides a machine-readable medium 300 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it realizes the steps of the flight seasonal classification prediction method described in Figure 1.

[0069] Herein, in this application, a machine-readable medium is a tangible medium that contains or stores a program used by or in conjunction with a command execution system, device, or apparatus. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or apparatus, or any suitable combination of the above. More specific examples of machine-readable storage media include one or more wired electrical connections, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0070] Here, the computer-readable medium in this application may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the above. The computer-readable storage medium may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage mediums may include, but are not limited to, electrical connections having one or more leads, portable computer magnetic disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, compact disk read-only memory (CD-ROM), optical memory, magnetic memory, or any suitable combination of the above. In this application, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or used in conjunction with a command execution system, apparatus, or device. In this application, the computer-readable signal medium contains data signals propagated in the baseband or as part of a carrier, and contains computer-readable program code. The data signals propagated in this manner may take many forms and may include, but are not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium may further be any computer-readable medium other than a computer-readable storage medium, which transmits, propagates, or transmits a program used by or combined with a command execution system, apparatus, or device. The program code contained in the computer-readable medium may be transmitted by any suitable medium, including, but are not limited to, wires, optical cables, RF (radio frequency), or any suitable combination thereof.

[0071] The computer-readable media described above may be included in the electronic device described above, or it may exist separately and not be attached to the electronic device.

[0072] Referring to Figure 4, which is a schematic diagram of the hardware structure of a server provided in an embodiment of the present application, the server 400 varies greatly in configuration or performance and includes one or more central processing units (CPUs) 422 (e.g., one or more processors) and memory 432, and one or more storage media 430 (e.g., one or more mass storage devices) for storing application programs 442 or data 444. The memory 432 and storage media 430 may be temporary or permanent storage. The programs stored in the storage media 430 include one or more modules (not shown), each module including a set of command operations in the server. Furthermore, the central processing unit 422 is configured to communicate with the storage media 430 so that the server 400 can execute a set of command operations in the storage media 430.

[0073] Server 400 has one or more power supplies 426, one or more wired or wireless network interfaces 450, one or more input / output interfaces 458, and / or one or more 441, for example, Windows Server TM Mac OS X TM Unix TM Linux TM FreeBSD TM This includes, among others.

[0074] In the above embodiment, the steps performed by the flight seasonal classification prediction device may be based on the server structure shown in Figure 4.

[0075] Furthermore, based on embodiments of this application, the process of the flight seasonal classification prediction method described in the flowchart of Figure 1 above may be implemented as a computer software program. For example, embodiments of this application include a computer program, including a computer program mounted on a non-temporary computer-readable medium, the computer program including program code that performs the method of the flowchart of Figure 1 above.

[0076] Although this topic has been described in a specific language of structural features and / or methodological logic, the topic limited to the claims is not necessarily limited to the specific features or operations described above. Furthermore, the specific features and operations described above are merely exemplary forms of realizing the claims.

[0077] The above discussion includes several specific implementation details, but these should not be understood as limitations on the scope of this application. Some features described in individual embodiments may be combined to be implemented in a single embodiment. Furthermore, various features described in a single embodiment may be implemented in multiple embodiments individually or in any appropriate sub-combination.

[0078] The above description is a description of preferred embodiments and the technical principles used in this application. As those skilled in the art will understand, the scope of disclosure of this application is not limited to technical solutions consisting of specific combinations of the above technical features, and also covers other technical solutions formed by arbitrarily combining the above technical features or their equivalent features, without departing from the concept of the above disclosure. For example, technical solutions formed by substituting the above features with similar functional technical features disclosed (but not limited to) in this application.

Claims

1. A method for predicting the seasonal classification of flights, A step of obtaining a target departure date corresponding to a target flight from a local database, wherein the target flight is an undeparted flight of the target airline that is subject to seasonal classification forecasting. A step of constructing N data pools corresponding to the target flight, separated by seasonal classification, based on a set of departing flights associated with the target flight, wherein N is a positive integer of 2 or more. A step of determining a first set of past flights corresponding to the target flight based on the target departure date, wherein each of the first past flights in the first set of past flights is a flight that is not classified seasonally, The steps include obtaining flight data for each first past flight in the first set of past flights, The steps include determining each of the data pools corresponding to each of the first past flights based on the N data pools and the flight data of each of the first past flights, and determining the weighted value of each of the first past flights as the weighted value corresponding to each of the corresponding data pools, A step of determining the seasonal classification of the target flight based on the weighted value of each of the first past flights and the number of days until departure of each of the first past flights, wherein the number of days until departure of each of the first past flights is the number of days until departure corresponding to the data acquisition point of the flight data of each of the first past flights, The method is characterized in that each step of the flight seasonal classification prediction method is performed by a computer device.

2. The step of determining the seasonal classification of the target flight based on the weighted values ​​of each of the first past flights and the number of days until departure of each of the first past flights is: A step of determining the sum of the weights of the first set of past flights based on the weights of each of the first past flights, A step of determining the total number of days for the first set of past flights based on the number of days until departure for each of the first past flights, The method according to claim 1, further comprising the step of determining the seasonal classification of the target flight based on the sum of the weighted values ​​and the total number of days.

3. The step of determining the seasonal classification of the target flight based on the sum of the weighted values ​​and the total number of days is: A step of comparing the sum of the weighted values ​​with the total number of days and obtaining a comparison result, wherein the comparison result indicates the relationship between the magnitude of the sum of the weighted values ​​and the total number of days, The method according to the previous version, comprising the step of determining the seasonal classification of the target flight based on the comparison results.

4. The steps of determining each of the data pools corresponding to each of the first past flights based on the N data pools and the flight data of each of the first past flights, and determining the weighted value of each of the first past flights in the set of past flights as the weighted value corresponding to each of the corresponding data pools, A step of determining revenue data for each of the first past flights based on the flight data for each of the first past flights, A step of dividing first revenue data into a first data pool, wherein the first revenue data is revenue data corresponding to any one of the first past flights, and the first data pool is the data pool among the N data pools that has the smallest distance between the central data and the first revenue data. The steps include determining a predetermined weight value for the first data pool, The method according to claim 1, comprising the step of determining a predetermined weighting value for the first data pool to be the weighting value for the flight corresponding to the first revenue data.

5. The step of determining a first set of past flights corresponding to the target flight based on the target departure date is: The following is a step in calculating the first due date, the second due date, the third due date, and the fourth due date using the following formula: The aforementioned first date = the aforementioned target departure date - 52 * 7 * i - 1; The second date mentioned above = the target departure date - 52 * 7 * i + 1; The third date mentioned above = the target departure date - 51 * 7 * i; The fourth date mentioned above = the target departure date - 53 * 7 * i; Among them, i = (1, 2, 3..., n), and the i is the step that is the year before the current year, The method according to any one of claims 1 to 4, comprising the step of determining a flight corresponding to the target flight among the first date, the second date, the third date, and the fourth date in the first set of past flights.

6. The step of determining a first set of past flights corresponding to the target flight based on the target departure date is: The step of calculating the target history reference date using the following formula, The aforementioned target history reference date = the aforementioned target departure date - 52 * 7 * i; where i = (1, 2, 3..., n), i is the step of the year preceding the current year, A step of determining a first due date, a second due date, a third due date, and a fourth due date based on the target history reference date, wherein the first due date is the day before the target history reference date, the second due date is the day after the target history reference date, the third due date is the week before the target history reference date, and the fourth due date is the week after the target history reference date. The method according to any one of claims 1 to 4, comprising the step of determining a flight corresponding to the target flight among the first date, the second date, the third date, and the fourth date in the first set of past flights.

7. The step of constructing N data pools, separated by seasonal classification, that correspond to the aforementioned target flights, is: Step 1 involves obtaining flight data for each second past flight in a second set of past flights corresponding to the target flight, Step 2 involves calculating revenue data for each of the second past flights based on the flight data for each of the second past flights, Step 3 is to determine the second revenue data to be the central data of the second data pool, wherein the second revenue data is revenue data corresponding to the first flight, the first flight is any one flight in the second set of past flights, and the second data pool is any one of the N data pools. Step 4, which calculates the distance between the third revenue data and the second revenue data, wherein the third revenue data is revenue data corresponding to any one flight in the flight subset, and the flight subset is the set of flights from the second set of past flights excluding the first flight. Step 5, which involves dividing the fourth revenue data into the second data pool, wherein the fourth revenue data is revenue data corresponding to a second flight, and the second flight is the flight in the flight subset that corresponds to the revenue data closest in distance to the second revenue data. Step 6 includes calculating the central data of the divided second data pool, The method according to any one of claims 1 to 4, characterized in that steps 3 to 6 are repeated until the revenue data corresponding to each second past flight in the second set of past flights is divided into the N data pools.

8. A flight seasonal classification and prediction device, An acquisition unit that obtains a target departure date corresponding to a target flight from a local database, wherein the target flight is an acquisition unit that is an undeparted flight of a target airline that is subject to seasonal classification prediction, A construction unit that constructs N data pools corresponding to the target flight, separated by seasonal classification, based on the flight set of departing flights associated with the target flight, wherein N is a positive integer of 2 or more, A first decision unit that determines a first set of past flights corresponding to the target flight based on the target departure date, wherein each of the first past flights in the first set of past flights is a flight that is not classified seasonally, A second determination unit determines each of the data pools corresponding to each of the first past flights based on the N data pools and the flight data of each of the first past flights, and determines the weighted value of each of the first past flights as the weighted value corresponding to each of the corresponding data pools, The seasonal classification of the target flight is determined based on the weighted value of each of the first past flights and the number of days until departure of each of the first past flights, and the number of days until departure of each of the first past flights is the number of days until departure corresponding to the data acquisition point of the flight data of each of the first past flights, and includes a third determination unit, The acquisition unit is further characterized by acquiring flight data for each of the first past flights in the first set of past flights.

9. Computer equipment, including memory, processor and bus system, The aforementioned memory stores the program, The bus system connects the memory and the processor, and enables communication between the memory and the processor. The computer device is characterized in that the processor executes a program in the memory and performs the flight seasonal classification prediction method according to any one of claims 1 to 4 based on instructions in the program code.

10. A computer-readable medium comprising a command, wherein, when executed by a computer device, the command causes the computer device to execute the flight seasonal classification prediction method described in any one of claims 1 to 4 above.

Images