KR20230159604A - Method and mechanism for predicting seasonal classification of flights, and machine-readable media - Google Patents

Abstract

To provide a method and related device for predicting seasonal classification of flights that can increase the accuracy of seasonal classification of flights and prevent bias caused by manual seasonal classification of flights, wherein the method determines the target departure date of the target flight locally. Obtaining from a database; Building N data pools corresponding to target flights; determining a first set of historical flights corresponding to the target flights based on the target departure date, wherein each of the first historical flights in the first set of historical flights is a flight that is not classified by season; acquiring flight data for each first historical flight in the first historical flight set; determining a weighted value of each first historical flight based on the N data pools and the flight data of each first historical flight; and determining seasonal classification of the target flights based on the weight of each of the first historical flights and the number of departure days of each of the first historical flights.

Description

Method and mechanism for predicting seasonal classification of flights, and machine-readable media

This application claims the priority of Chinese Patent Application No. 202110604811.2, filed with the State Intellectual Property Administration of China on May 31, 2021, entitled "Method and Apparatus for Predicting Seasonal Classification of Flights, and Machine-Readable Media"; , which is incorporated herein by reference in its entirety.

Field

The present disclosure relates to the field of aviation, and more particularly to a method for predicting seasonal classification of flights, a seasonal classification prediction mechanism for flights, and machine-readable media.

Periodically affected by factors such as climatic conditions, unexpected events, industrial and agricultural production and life, festivals and holidays, and other social customs and habits, and national economic development, the number of passengers and cargo volume in the civil air transport industry varies depending on the season. It fluctuates accordingly. In the air transportation field, historical departure flights are classified by season, such as off-season and peak season.

Currently, seasonal classification of flights is performed by humans in a quantitative or qualitative manner. However, biased results are obtained by artificially determining the seasonal classification of flights, which may result in low accuracy of seasonal classification of flights.

summary

A method for predicting seasonal classification of flights, a seasonal classification prediction device for flights, and machine-readable media are provided to improve the accuracy of seasonal classification in flights and prevent biased results due to artificial seasonal classification.

In a first aspect, a method for predicting seasonal classification of flights according to embodiments of the present disclosure is provided. The method includes obtaining a target departure date of a target flight, which is a non-departure flight classified by season at the target airline, from a local database; Building N data pools corresponding to target flights, where N represents a positive integer of 2 or more; determining a first set of historical flights corresponding to the target flights based on the target departure date, wherein each of the first historical flights in the first set of historical flights is not classified by season; acquiring flight data for each first historical flight in the first historical flight set; determining a weighted value of each first historical flight based on the N data pools and the flight data of each first historical flight; and determining seasonal classification of the target flights based on the weight of each of the first historical flights and the number of departure days of each of the first historical flights.

In a second aspect, a seasonal classification prediction mechanism for flights according to embodiments of the present disclosure is provided. The mechanism includes an acquisition unit, a construction unit, a first decision unit, a second decision unit, and a third decision unit. The acquisition unit is configured to obtain the target departure date of the target flight, which is a non-departure flight classified by season on the target airline, from the local database. The construction unit is configured to build N data pools corresponding to target flights, where N represents a positive integer of 2 or more. The first determination unit is configured to determine a first set of historical flights corresponding to the target flight based on the target departure date, where each of the first historical flights in the first set of historical flights is not classified by season. The acquisition unit is further configured to acquire flight data of each of the first historical flights in the first historical flight set. The second decision unit is configured to determine a weighted value of each of the first historical flights based on the N data pools and the flight data of each of the first historical flights. The third determination unit is configured to determine the seasonal classification of the target flights based on the weight of each of the first historical flights and the number of departure days of each of the first historical flights.

In one embodiment, the third determination unit determines a weighted sum of the set of first historical flights based on the weight of each of the first historical flights; determine the total number of departure days of the first set of historical flights based on the number of departure days for each of the first historical flights; It is configured to determine the seasonal classification of target flights based on the weighted sum and total number of departure days.

In one embodiment, the third decision unit compares the weight sum and the total number of departure days to obtain a comparison result, wherein the comparison result represents a relationship between the weight sum and the total number of departure days; By determining the seasonal classification of the target flight based on the comparison results, the seasonal classification of the target flight is determined based on the sum of weights and the total number of departure days.

In one embodiment, the second determination unit determines, for each of the first historical flights, revenue data of the first historical flight based on flight data of the first historical flight; Sort the first revenue data into a first data pool, where the first revenue data is revenue data of the first historical flight, and the first data pool is the central data with the minimum distance from the first revenue data in the N data pools. It is a data pool with; determine a preset weight of the first data pool; It is configured to determine a preset weight of the first data pool as the weight of the flight corresponding to the first revenue data.

In one embodiment, the first determination unit calculates the first date, second date, third date, and fourth date using the following equation:

1st date = target departure date-52*7*i-1

2nd date = target departure date-52*7*i+1

3rd date = target departure date-51*7*i

4th date = target departure date-53*7*i

Here, i=(1, 2, 3..., n) and i represents the year before the current year; and

and determine flights corresponding to target flights of the first date, second date, third date and fourth date as the first set of historical flights.

In one embodiment, the first determination unit calculates the target history reference date using the following equation:

Target history base date = Target departure date-52*7*i

Here, i=(1, 2, 3..., n) and i represents the year before the current year;

A first date, a second date, a third date, and a fourth date are determined based on the target history reference date, where the first date is the day before the target history reference date, the second date is the day after the target history reference date, and the first date is the day after the target history reference date. The third date is the same day as one week before the target history reference date, and the fourth date is the same day as one week after the target history reference date; and

and determine flights corresponding to the target flights of the first date, the second date, the third date and the fourth date as the first set of historical flights.

In one embodiment, the building unit is

Step 1, obtain flight data of each second historical flight in the second historical flight set corresponding to the target flight;

Step 2, for each second historical flight, calculate revenue data of the second historical flight based on the flight data of the second historical flight;

Step 3, determine the second revenue data as the central data of the second data pool, where the second revenue data is the revenue data of the first flight, the first flight is a flight in the second historical flight set, and the second data A pool is a data pool of N data pools;

Step 4, calculate the distance between the third revenue data and the second revenue data, where the third revenue data is the revenue data of the flight in the flight sub-set and the flight sub-set is the first revenue data in the second historical flight set. Set of flights excluding flights;

Step 5, sort the fourth revenue data into a second data pool, where the fourth revenue data is the revenue data of the second flight, and the second flight corresponds to the revenue data closest to the second revenue data in the flight sub-set. It is a flight; and

Step 6, calculate the central data of the second data pool after classification;

Steps 3 to 6 are configured to be performed until the revenue data of each second historical flight in the second historical flight set is sorted into N data pools.

In a third aspect, a computer device according to the present disclosure is provided. A computer device includes memory, processor, and bus systems. The memory is configured to store programs. A bus system is configured to connect a memory and a processor for communication between the memory and the processor. The processor is configured to execute a program in memory and, when executing instructions to the program, perform the method for predicting the seasonal classification of flights according to the first aspect described above.

In a fourth aspect, a machine-readable medium according to embodiments of the present disclosure is provided. The machine-readable medium stores instructions. The instructions, when executed by the machine, cause the machine to perform the seasonal classification prediction method for flights described above.

In summary, according to an embodiment of the present disclosure, in determining the seasonal classification of a target flight, the seasonal classification prediction mechanism of a flight obtains the target departure date of the target flight and creates N data pools corresponding to the target flight. Construct, determine a first set of historical flights based on the target departure date, determine a weighted value of each first historical flight in the first historical flight set based on the flight data of each first historical flight, and 1 It can be seen that the seasonal classification of the target flight can be finally determined based on the weight of each historical flight and the departure date of each first historical flight. According to the present disclosure, compared to artificially performing the existing classification of flights by season, the accuracy of classification by season can be improved and biased results that occur by artificially performing classification of flights by season can be prevented.

These and other features, advantages and aspects of embodiments of the present disclosure will become more apparent with reference to the following embodiments in conjunction with the drawings. Identical or similar reference numbers throughout the drawings indicate identical or similar elements. It should be understood that the drawings are schematic and that components and elements are drawn to scale beyond what is necessary.
1 is a flowchart of a method for predicting seasonal classification of flights according to an embodiment of the present disclosure.
2 is a schematic diagram of a hypothetical structure of a seasonal classification prediction mechanism for flights according to an embodiment of the present disclosure.
3 is a schematic structural diagram of a machine-readable medium according to an embodiment of the present disclosure.
4 is a schematic diagram of the hardware structure of a server according to an embodiment of the present disclosure.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and is not limited to the embodiments presented herein. Embodiments are provided for a clear and complete understanding of the present disclosure. It should be understood that the drawings and examples of the present disclosure are only schematic and are not intended to limit the scope of protection of the present disclosure.

In this disclosure, the term “including” and variations thereof means open-ended inclusion, i.e., “including but not limited to.” The term “based on” means “at least in part based on.” The term “one embodiment” means “at least one embodiment.” The term “other embodiment” means “at least one other embodiment.” The term “some embodiments” means “at least some embodiments.” Definitions of other terms are provided in the description below.

Concepts such as “first” and “second” mentioned in the present disclosure are used to distinguish between different devices, modules, or units, and the order or interdependence of functions performed by the device, module, or unit. It should be noted that this is not intended to limit.

It should be noted that variations such as “one,” and “many” mentioned in this disclosure are illustrative and not limiting. Those skilled in the art should understand that unless the context clearly indicates otherwise, a variant is to be understood as “one or more”.

First, the following terminology related to embodiments of the present disclosure is explained.

The revenue management system is a system that automatically manages the inventory of non-departing flights using forecasting and optimization models based on flight plan, inventory, departure, and fare data.

Market demand represents the requirements that passengers have the ability to purchase and actually have a demand to purchase. The market demand value may or may not result in an actual order, and the market demand value is calculated by the revenue management system.

Day of Departure (Ndo) is the number of days from the system date (i.e. current date) to the departure date of the flight or flight segment. For example, if the current date is April 26, 2021 and the departure date is May 1, 2021, the number of departure days (Ndo) is equal to 5.

The data collection point (DCP) is determined based on the number of departure days and corresponds 1:1 to the number of departure days. For example, the data collection points may consist of 24 (including DCP ₁ , DCP ₂ , ..., DCP ₂₄ ), the number of departure days may consist of 365 days, and the data collection points may consist of 365 days of departure. There is a 1:1 correspondence, so data collection point DCP ₁ corresponds to the 365-day departure date and data collection point DCP ₂₄ corresponds to the 1-day departure date. It should be understood that the number of data collection points, the number of departure days, and the correspondence between the data collection points and the number of departure days are only examples and are not limited to the present specification.

Day of the week (DOW) represents the general week.

Hereinafter, a method for predicting seasonal classification of flights according to the present disclosure will be described from the perspective of a seasonal classification prediction mechanism for flights. Refer to Figure 1, which is a flow chart of a method for predicting seasonal classification of flights according to an embodiment of the present disclosure. The method includes the following steps 101 to 106.

In step 101, the target departure date of the target flight is obtained from the local database.

In embodiments, a seasonal classification prediction tool for flights may obtain a target departure date for a target flight from a local database. Target flights are non-departure flights classified by season on the target airline. It should be understood that the flight data for the designated airline (i.e., the target airline) is stored in a local database, and the flight control system stores either full or incremental flight data for the designated airline. It may be configured to obtain full or incremental flight data of a designated airline from the flight control system at predetermined time intervals, for example, to acquire full or incremental flight data every 24 hours. Flight data includes, but is not limited to, the following data: flight number, flight departure date, data collection point (DCP), number of departure days corresponding to the data collection point (DCP), reservation amount for each class, and Shipping amount equivalent to the reservation amount. Accordingly, the seasonal classification prediction mechanism for flights can directly obtain the target departure date of the target flight from the local database.

It should be noted that the above description uses the target flight as an example. Additionally, a flight segment can be used as an example. The target flight includes at least one flight segment. Flight segments form the passenger route. For example, the target flight is the Beijing-Shanghai-San Francisco flight, and there are three possible routes for passengers: Beijing-Shanghai, Shanghai-San Francisco, and Beijing-San Francisco. That is, the target flights correspond to the following three flight segments: Beijing-Shanghai flight segment, Shanghai-San Francisco flight segment, and Beijing-San Francisco flight segment.

In step 102, N data pools corresponding to target flights are constructed.

In one embodiment, a seasonal classification prediction tool for flights may build N data pools corresponding to target flights, where N represents a positive integer of 2 or more. In this disclosure, N is set to 7. The seven data pools are the peak season 1 data pool (denoted as peak), the peak season 2 data pool (denoted as peak 1), the semi-peak season 1 data pool (denoted as peak 2), and the semi-peak season 2 data pool (off-peak). 2), the data pool in off-peak 1 (marked as off-peak1), the data pool in off-peak 2 (marked as off-peak), and the unclassified data pool (marked as default). The number of data pools and classification of data pools in this disclosure are illustrative only and are not limiting.

In one embodiment, a seasonal classification prediction tool for flights builds N data pools corresponding to target flights by performing the following steps:

In Step 1, flight data for each second historical flight in the second historical flight set corresponding to the target flight is acquired.

At this stage, the flight season classification prediction mechanism may obtain flight data for each second flight history in the second history flight set corresponding to the target flight from the local database. Additionally, the second set of historical flights corresponding to the target flights includes departing flights associated with the current date during the last 3 years (including, but not limited to, other periods such as 4 years) from the current date. For example, if the target flight is a flight on April 25, 2021, and April 25 is a Sunday, the second set of historical flights is the set of flights corresponding to the target flight on all Sundays for the past three years, starting from April 25, 2021. am.

In Step 2, for each of the second historical flights, revenue data of the second historical flight is calculated based on the flight data of the second historical flight.

At this stage, the seasonal segment forecasting tool for flights can calculate revenue data using the following equation:

In the above equation, revenueDcp(x) represents the revenue data of the xth flight in each of the second historical flights, i represents the class, k represents the total number of classes, and BKG(i) represents the revenue data of the It represents the number of seats, and the fare (i) represents the fare for the ith class.

In step 3, the second revenue data is determined as the central data of the second data pool. The second revenue data is the revenue data of the first flight, the first flight is a flight in the second historical flight set, and the second data pool is a data pool among N data pools.

At this stage, the seasonal classification forecasting tool for flights may determine the second revenue data as the central data of the second data pool. The second revenue data is the revenue data of the first flight. The first flight is a flight from the second set of historical flights, and the second data pool is a data pool among N data pools. That is, the revenue data of N historical flights can be randomly selected as the central data of the N data pool.

In step 4, a first distance between the third revenue data and the second revenue data is calculated. The third revenue data is revenue data of flights within the flight sub-set. A flight sub-set is a set of flights excluding the first flight in the second historical flight set.

At this stage, the seasonal classification prediction tool for flights can calculate the distance between the third revenue data and the central data of the N data pools using the following equation:

In the above equation, D(i,j) represents the distance between the second revenue data i and the third revenue data j, the third revenue data j is the revenue data of the flight in the flight sub-set, and W(t) is the revenue data of the flight in the flight sub-set. It represents the weight of the third revenue data, X _it represents the market demand value corresponding to the second revenue data, and Y _jt represents the market demand value corresponding to the third revenue data.

In step 5, the fourth revenue data is sorted into the second data pool. The fourth revenue data is the revenue data of the second flight. The second flight is the flight whose revenue data is closest to the second revenue data in the flight sub-set.

At this stage, the flight season classification prediction tool may classify the fourth revenue data into the first data pool. The fourth revenue data is the revenue data of the second flight. The second flight is the flight whose revenue data is closest to the second revenue data in the flight sub-set. That is, each flight within the flight sub-set can be classified into the data pool that is closest to the flight.

In step 6, the median data of the second data pool is calculated after classification.

At this stage, the seasonal classification prediction tool for flights may calculate the central data of the second data pool using the following equation:

In the above equation, the new average (t) represents the central data of the target data pool after the fourth revenue data is classified into the second data pool, and the old average (t) represents the central data of the target data pool after the fourth revenue data is classified into the second data pool. represents the central data of the target data before, m represents the central data number of the target data pool before the fourth revenue data is classified into the second data pool, and Represents market demand.

Steps 3 to 6 are performed until the revenue data for each of the second historical flights in the second historical flight set is sorted into N data pools.

It should be noted that market demand represents the requirements that passengers have the ability to purchase and the actual demand to purchase. Market demand may or may not trigger an actual order. Market demand is calculated by the revenue management system using a limit algorithm. The market demand value is widely used to determine market trends and distinguish market peak or off-peak seasons, and is an important input for the optimization module of the core algorithm subsystem of the revenue management system. The revenue management system is an automated management system in which airlines automatically manage the inventory of non-departure flights based on flight plan, inventory, departure, and fare data. The way to obtain market demand for flights is not limited. For example, after obtaining flight information for a designated flight of a target airline, inventory data is acquired based on the flight information, where the inventory data includes inventory data of departing flights and inventory data of non-departing flights. The inventory data of departing flights is the inventory data of departing flights of designated flights for the past three years from the current date, and the inventory data of non-departing flights is the inventory data of non-departing flights of designated flights for the next year as of the current date. Sales status for a given class on a given flight is determined based on inventory data. The sales status of the designated class on a designated flight is determined based on the data collection point corresponding to the target airline, the designated airline's flight information, and the designated flight's inventory data. Sales status includes locked status, open status, etc. Sales status is processed based on a predetermined algorithm to obtain market demand for a given flight.

Please note that if the number of available seats in a designated class is 0 or less and the availability status is open or closed, it means that the designated class is unsaleable, that is, locked. If the number of available seats in the designated class exceeds 0 and the availability status is closed, it means that the designated class is unsaleable, that is, locked. If the number of available seats in the designated class exceeds 0 and the availability status is open, it means that the designated class is available for sale, that is, in the open state.

Below, we describe the process of processing sales status based on a predetermined algorithm to obtain the market demand for a designated flight.

If the sales status of data collection point DCP (n+1) is open, and the sales status of data collection point DCP (n) is open,

If the number of reserved seats corresponding to data collection point DCP(n) is greater than the number of reserved seats corresponding to data collection point DCP(n+1), the market demand value of data collection point DCP(n+1) is expressed as follows: It is calculated using:

Market demand for DCP(n+1) = Market demand for DCP(n) + increased reserved seats for DCP(n)

In the above equation,

Increased reserved seats = Actual reserved seats in DCP(n+1) - Actual reserved seats in DCP(n); and

If the number of reserved seats corresponding to data collection point DCP(n) is less than the number of reserved seats corresponding to data collection point DCP(n+1), the market demand value of data collection point DCP(n+1) is expressed as follows: It is calculated using:

Market demand for DCP(n+1) = Market demand for DCP(n) + reduced reserved seats for DCP(n)

In the above equation,

Reduced reserved seats = (Actual reserved seats in DCP(n+1)*Market demand (n))/Actual reserved seats in DCP(n)-Market demand (n).

If the sales status of data collection point DCP (n+1) is locked, and the sales status of data collection point DCP (n) is open,

If the number of reserved seats corresponding to data collection point DCP(n) is less than the number of reserved seats corresponding to data collection point DCP(n+1), the market demand value of data collection point DCP(n+1) is expressed as follows: It is calculated using:

Market demand for DCP(n+1) = Market demand for DCP(n) + reduced reserved seats for DCP(n)

In the above equation,

Reduced reserved seats = (Actual reserved seats in DCP(n+1)*Market demand (n))/Actual reserved seats in DCP(n)-Market demand (n).

It should be understood that the above process of calculating the market demand value is an iterative process, that is, the market demand value of DCP(1) is obtained by repeatedly calculating DCP+=1 based on the actual reserved seats of DCP(1).

At step 103, a first set of historical flights corresponding to the target flights is determined based on the target departure date.

In one embodiment, a seasonal classification prediction tool for flights may determine a first set of historical flights corresponding to the target flights based on the target departure date. The first set of historical flights can be determined in two ways:

In a first method, the first date, second date, third date and fourth date are calculated using the following formula:

1st date = target departure date-52*7*i-1

2nd date = target departure date-52*7*i+1

3rd date = target departure date-51*7*i

4th date = target departure date-53*7*i

Here, i=(1, 2, 3..., n) and i represents the year before the current year. For example, if the current year is 2021, i could be a year before 2021, such as 2020, 2019, 2018, etc.;

Flights corresponding to target flights on the first date, second date, third date, and fourth date are determined as the first set of historical flights. Below, an example is provided.

For example, the target departure date is December 30, 2020, the first date is December 31, 2019, the second date is January 2, 2020, the third date, the fourth date, and the target departure date are If they have the same DOW (day of week), the third date is December 25, 2019, and the fourth date is January 8, 2020. The first set of historical flights is the set of flights corresponding to the target flights on December 31, 2019, January 2, 2020, December 25, 2019, and January 8, 2020.

In a second method, the first set of history flights corresponding to the target flights are determined based on the target history reference date. The target history reference date is calculated using the following formula:

Target history reference date=Target departure date-52*7*i

where i=(1, 2, 3..., n) and i represents the year before the current year;

The first date, the second date, the third date, and the fourth date are determined based on the target history reference date, where the first date is the day before the target history reference date, the second date is the day after the target history reference date, and the first date is the day after the target history reference date. The third date is the same day as one week before the target history reference date, and the fourth date is the same day as one week after the target history reference date; and

Flights corresponding to target flights on the first date, second date, third date, and fourth date are determined as the first set of historical flights. Below, an example is provided.

For example, if the target departure date is December 30, 2020, the target history base date is 2020/12/30-52*7*1=2020/1/1, and DOW is obtained as Wednesday. The date of the day before the target history reference date is December 31, 2019 (i.e., the first date is December 31, 2019). The date of the day following the target history reference date is January 2, 2020 (i.e., the second date is January 2, 2020). The date with the same DOW one week prior to the target history reference date is December 25, 2019 (i.e., the third date is December 25, 2019). The date with the same DOW one week after the target history reference date is January 8, 2020 (i.e., the fourth date is January 8, 2020). The first set of historical flights is the set of flights corresponding to the target flights on December 31, 2019, January 2, 2020, December 25, 2019, and January 8, 2020.

At step 104, flight data for each of the first historical flights in the first set of historical flights is obtained.

In an embodiment, the seasonal classification prediction tool for flights may determine the first set of historical flights and then obtain flight data for each of the first historical flights in the first set of historical flights from a local database. A target departure date may be obtained in step 101, N data pools may be constructed in step 102, and flight data for each of the first historical flights in the first set of historical flights may be obtained in steps 103 and 104. Be careful. However, the order in which steps 101, 102, 103 and 104 are performed is not limited. Step 101 may be performed first, step 102 may be performed first, steps 103 and 104 may be performed first, or steps 101, 102, 103, and 104 may be performed simultaneously, but this is not the case. It is not limited to the disclosure content.

In step 105, the weight of each first historical flight is determined based on the N data pools and the flight data of the first historical flight.

In an embodiment, the seasonal classification prediction mechanism for flights may determine a weight for each of the first historical flights based on the N data pools and the flight data of the first historical flights. Specifically, for each first historical flight, revenue data for the first historical flight may be determined based on flight data of the first historical flight, where calculation of the revenue data is described in detail in step 102 and is described in detail in this disclosure. is not repeated), and then the first revenue data is classified into the first data pool. The first revenue data is revenue data corresponding to each of the first historical flights, and the first data pool is a data pool with the central data at the minimum distance from the first revenue data in the N data pools. The preset weight of the first data pool is determined, and the preset weight of the first data pool is determined as the weight of the flight corresponding to the first revenue data. That is, revenue data corresponding to each of the first historical flights may be classified into data pools having a minimum distance from the revenue data, and a preset weight of the data pool may be determined as the weight of the first historical flight.

It should be noted that the local database basically stores the preset weights of each of the N data pools. In this disclosure, N is set to 7 (may be, but is not limited to, other values), and the preset weights of the seven data pools are respectively set as follows: Peak = 3, Peak1 = 2, Peak2 = 1, default = 0, offpeak2 = -1, offpeak1 = -2, and offpeak = -3. After classifying the revenue data corresponding to each of the first historical flights into seven weights, the preset weight of the data pool may be determined as the weight of the first historical flight classified in the data pool. It should be understood that if the first set of historical flights includes flights classified by season, the weight of the flights is determined to be 0.

In step 106, the seasonal classification of the target flights is determined based on the weight of each first historical flight and the number of departure days of each first historical flight.

In an embodiment, the seasonal classification prediction mechanism for flights determines a weighted sum of the set of first historical flights based on the weight of each of the first historical flights, and the first historical flights based on the number of departure days of each of the first historical flights. The total number of departure days in the set is determined, and the seasonal classification of the target flights can be finally determined based on the weighted sum and total number of departure days. In one embodiment, the seasonal classification prediction mechanism for flights compares the sum of weights and the total number of departure days to obtain a comparison result, wherein the comparison result represents a relationship between the sum of weights and the total number of departure days; By determining the seasonal classification of the target flight based on the comparison results, the seasonal classification of the target flight is determined based on the sum of weights and the total number of departure days.

In an embodiment, the weighted sum of the first set _of historical flights is denoted _by X _i , the total number of departure days of the first set of historical flights is denoted by Y _i , and after comparing It is determined based on the relationship between X _i and Y _i . Below, a comparison of X _i and Y _i and classification by season are described.

If X _i >Y _i , the target flight is classified as peak, that is, the seasonal classification of the target flight is peak season 1.

If X _i >(2/3)*Y _i , the target flight is classified as peak 1, that is, the seasonal classification of the target flight is peak season 2.

If X _i >(1/3)*Y _i , the target flight is classified as peak 2, that is, the seasonal classification of the target flight is semi-peak season 1.

If X _i <-Y _i , the target flight is classified as off-peak, that is, the seasonal classification of the target flight is off-peak season 2.

If X _i <(-2/3)*Y _i , the target flight is classified as off-peak 1, that is, the seasonal classification of the target flight is off-peak 1.

_{If _}

In summary, according to an embodiment of the present disclosure, in determining the seasonal classification of a target flight, the seasonal classification prediction mechanism of a flight obtains the target departure date of the target flight and creates N data pools corresponding to the target flight. Construct, determine a first set of historical flights based on the target departure date, determine a weighted value of each first historical flight in the first historical flight set based on the flight data of each first historical flight, and 1 It can be seen that the seasonal classification of the target flight can be finally determined based on the weight of each historical flight and the departure date of each first historical flight. According to the present disclosure, compared to artificially performing the existing classification of flights by season, the accuracy of classification by season can be improved and biased results that occur by artificially performing classification of flights by season can be prevented.

It should be understood that the flow diagrams and block diagrams in the drawings illustratively illustrate possible system architectures, functions, and operations that may be implemented by systems, methods, and computer program products in accordance with various embodiments of the present disclosure. Each block in a flowchart or block diagram may represent part of a module, program section, or code, and each part of a module, program section, or code contains one or more executable instructions to implement a specified logic function. It should be noted that in some alternative implementations, the functions represented by blocks may be performed in an order other than that shown in the figures. For example, two blocks shown in succession may actually be executed in parallel, or sometimes even in reverse order, depending on the functions involved. 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 dedicated hardware-based system that performs the designated function or operation, or may be implemented by a combination of dedicated hardware and computer instructions. It should be noted that there is.

The names of messages or information exchanged between multiple organizations in embodiments of the present disclosure are provided for illustrative purposes only and are not intended to limit the scope of the messages or information.

Although the above tasks are described in a specific order, it should be noted that these tasks should not be understood as necessarily being performed in a specific order or required to be performed sequentially. In certain situations, multitasking and parallel processing can be advantageous.

It should be understood that the steps described in embodiments of the methods of the present disclosure may be performed in different orders and/or may be performed in parallel. Additionally, embodiments of the method may include additional steps and/or illustrated steps that may be omitted. The scope of the present disclosure is not limited thereto.

Additionally, computer program code for performing the tasks of this disclosure may be written in one or more programming languages or a combination thereof. Programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, and C++, and traditional procedural programming languages such as the C language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer, and partially on a remote computer, or entirely on a remote computer or server. When a remote computer is involved, the remote computer may be connected to your computer or an external computer through any type of network, including a local area network (LAN) or a wide area network (WAN) (for example, the remote computer may be connected to your Internet service provider). can be connected to your computer or an external computer through an Internet connection).

Embodiments of the disclosure described above have been described in terms of a method for predicting seasonal classification of flights. In the following, embodiments of the present disclosure are described from the perspective of a seasonal classification prediction mechanism for flights.

Reference is made to Figure 2, which is a schematic diagram of a hypothetical structure of a seasonal classification prediction tool for flights according to an embodiment of the present disclosure. The seasonal classification prediction mechanism for flights 200 includes an acquisition unit 201, a construction unit 202, a first decision unit 203, a second decision unit 204, and a third decision unit 205.

The acquisition unit 201 is configured to obtain the target departure date of the target flight, which is a non-departure flight classified by season on the target airline, from the local database.

The building unit 202 is configured to build N data pools corresponding to target flights, where N represents a positive integer of 2 or more.

The first determination unit 203 is configured to determine a first set of historical flights corresponding to the target flight based on the target departure date, where each of the first historical flights in the first set of historical flights is not classified by season.

The acquisition unit 201 is further configured to acquire flight data of each of the first historical flights in the first historical flight set.

The second determination unit 204 is configured to determine a weighted value of each of the first historical flights based on the N data pools and the flight data of each of the first historical flights.

The third determination unit 205 is configured to determine the seasonal classification of the target flights based on the weight of each first historical flight and the number of departure days of each first historical flight.

In one embodiment, the third determination unit 205 determines a weighted sum of the first set of historical flights based on the weight of each of the first historical flights; determine the total number of departure days of the first set of historical flights based on the number of departure days for each of the first historical flights; It is configured to determine the seasonal classification of target flights based on the weighted sum and total number of departure days.

In one embodiment, the third determination unit 205 compares the weighted sum and the total number of departure days to obtain a comparison result, wherein the comparison result represents a relationship between the weighted sum and the total number of departure days; By determining the seasonal classification of the target flight based on the comparison results, the seasonal classification of the target flight is determined based on the sum of weights and the total number of departure days.

In one embodiment, the second determination unit 204 determines, for each first historical flight, revenue data of the first historical flight based on flight data of the first historical flight; Sort the first revenue data into a first data pool, where the first revenue data is revenue data corresponding to the first historical flight, and the first data pool is the center at the minimum distance from the first revenue data in the N data pools. It is a data pool with data; determine a preset weight of the first data pool; It is configured to determine the preset weight of the first data pool as the weight of the flight corresponding to the first revenue data.

In one embodiment, the first determination unit 203 calculates the first date, second date, third date, and fourth date using the following equation:

1st date = target departure date-52*7*i-1

2nd date = target departure date-52*7*i+1

3rd date = target departure date-51*7*i

4th date = target departure date-53*7*i

where i=(1, 2, 3..., n) and i represents the year before the current year; and

and determine flights corresponding to target flights of the first date, second date, third date and fourth date as the first set of historical flights.

In one embodiment, the first determination unit 203 calculates the target historical reference date using the following equation:

Target history base date = Target departure date-52*7*i

where i=(1, 2, 3..., n) and i represents the year before the current year;

A first date, a second date, a third date, and a fourth date are determined based on the target history reference date, where the first date is the day before the target history reference date, the second date is the day after the target history reference date, and the third date is the day before the target history reference date. The date is the same day as one week before the target history reference date, and the fourth date is the same day as one week after the target history reference date; and

and determine flights corresponding to the target flights of the first date, the second date, the third date and the fourth date as the first set of historical flights.

In one embodiment, construction unit 202 is

Step 1, obtain flight data of each second historical flight in the second historical flight set corresponding to the target flight;

Step 2, for each second historical flight, calculate revenue data of the second historical flight based on the flight data of the second historical flight;

Step 3, determine the second revenue data as the central data of the second data pool, where the second revenue data is the revenue data of the first flight, the first flight is a flight in the second historical flight set, and the second data A pool is a data pool among N data pools;

Step 4, calculate the distance between the third revenue data and the second revenue data, where the third revenue data is the revenue data of the flight in the flight sub-set and the flight sub-set is the first revenue data in the second historical flight set. Set of flights excluding flights;

Step 5, sort the fourth revenue data into a second data pool, where the fourth revenue data is the revenue data of the second flight, and the second flight corresponds to the revenue data closest to the second revenue data in the flight sub-set. It is a flight; and

Step 6, calculate the central data of the second data pool after classification;

Steps 3 to 6 are configured to be performed until the revenue data of each second historical flight in the second historical flight set is sorted into N data pools.

In summary, according to an embodiment of the present disclosure, in determining the seasonal classification of a target flight, the seasonal classification prediction mechanism of a flight obtains the target departure date of the target flight and creates N data pools corresponding to the target flight. Construct, determine a first set of historical flights based on the target departure date, determine a weighted value of each first historical flight in the first historical flight set based on the flight data of each first historical flight, and 1 It can be seen that the seasonal classification of the target flight can be finally determined based on the weight of each historical flight and the departure date of each first historical flight. According to the present disclosure, compared to artificially performing the existing classification of flights by season, the accuracy of classification by season can be improved and biased results that occur by artificially performing classification of flights by season can be prevented.

It should be noted that the units described in embodiments of the present disclosure may be implemented in software or hardware. In some cases, the unit name does not limit the unit itself. For example, the acquisition unit may be further described as “a unit that acquires identification information of a target user.”

The functions described above may be performed at least in part by one or more hardware logic components. Example, but not limited, types of hardware logic components that may be used include, but are not limited to, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOC), Includes CPLD (complex programmable logic device), etc.

See Figure 3, which is a schematic structural diagram of a machine-readable medium according to an embodiment of the present disclosure.

As shown in FIG. 3 , a machine-readable medium 300 is provided in accordance with an embodiment of the present disclosure. Machine-readable medium 300 stores computer program 311. When executed by a processor, the computer program 311 causes the processor to perform a method for predicting the classification of flights by season, as shown in FIG. 1 .

It should be noted that, in the context of this disclosure, a machine-readable medium may be a tangible medium that can contain or store a program used by or in combination with an instruction execution system, instrument or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media may include, but are not limited to, systems, instruments or devices in electronic, magnetic, optical, electromagnetic, infrared or semiconductor form, or combinations thereof. Specific examples of machine-readable storage media include electrical connections using one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory). , optical fiber, portable compact disk read-only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination thereof.

It should be noted that the computer-readable medium of the present disclosure may be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. Computer-readable storage media may include, but are not limited to, systems, instruments or devices in electronic, magnetic, optical, electromagnetic, infrared or semiconductor form, or combinations thereof. Specific examples of computer-readable storage media include electrical connections using one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory). , optical fiber, portable compact disk read-only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any type of medium that contains or stores a program, and the program may be used by, or in combination with, an instruction execution system, mechanism, or device. In the present disclosure, a computer-readable signal medium may include a baseband data signal or may be transmitted as part of a carrier wave to carry computer-readable program code. The transmitted data signals may be in a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may alternatively be any computer-readable medium other than a computer-readable storage medium. A computer-readable signal medium may transmit, propagate, or transmit a program used by or in combination with an instruction execution system, apparatus, or component. Program code stored on a computer-readable medium may be transmitted via any suitable medium, including but not limited to wired, optical cable, radio frequency (RF), etc., or any suitable combination thereof.

A computer-readable medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

Reference is made to Figure 4, which is a schematic diagram of the hardware structure of a server according to an embodiment of the present disclosure. Server 400 may vary significantly due to different configurations or performance. Server 400 includes one or more central processing units (CPUs) 422 (e.g., one or more processors), memory 432, and one or more storage media ( 430) (e.g., one or more mass storage devices). The memory 432 and storage medium 430 may perform temporary storage or permanent storage. A program stored in the storage medium 430 may include one or more modules (not shown in FIG. 4), and each module may include a series of command operations in the server. Additionally, the CPU 422 may be configured to communicate with the storage medium 430 and perform a series of command operations on the storage medium 430 in the server 400 .

Server 400 may include one or more power supplies 426, one or more wired or wireless network interfaces 450, one or more input/output interfaces 458, and/or Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, or It may further include one or more operating systems 441, such as FreeBSDTM.

The steps performed by the seasonal classification prediction tool for flights in the embodiment described above may be performed based on the structure of the server shown in FIG. 4.

Additionally, it should be further noted that, according to one embodiment of the present disclosure, the method for predicting seasonal classification of flights shown in FIG. 1 may be implemented as a computer software program. For example, computer program products according to embodiments of the present disclosure are further provided. A computer program product includes a computer program delivered on a non-transitory computer-readable medium. The computer program includes program code for performing the method as shown in the flowchart of FIG. 1.

Although subject matter is described in language specific to the structural features and/or logical operations of the method, it should be understood that the subject matter defined in the claims is unnecessarily limited to the specific features or operations described above. The specific features and operations described above are merely example forms of implementing the claims.

Although numerous implementation details are included in the above description, the details should not be construed as limitations on the scope of the disclosure. Some features described in the context of an embodiment can be implemented in combination in other embodiments. Alternatively, some features described in the context of the examples may be implemented individually or in any suitable subcombination in multiple embodiments.

The above description merely includes a description of preferred embodiments of the present disclosure and the technical principles used. Those skilled in the art will understand that the scope of the present disclosure is not limited to technical solutions formed by a combination of the technical features described above, and other solutions formed by any combination of technical features or equivalents thereof without departing from the concept of the present disclosure. It must be understood that it also includes technical measures. For example, the scope of the present disclosure may include technical solutions formed by replacing the above-described features with technical features having similar functions as (but not limited to) described in the present disclosure.

Claims (10)

Obtaining the target departure date of the target flight, which is a non-departure flight classified by season at the target airline, from a local database;
Building N data pools corresponding to the target flights, where N represents a positive integer of 2 or more;
determining a first set of historical flights corresponding to the target flights based on the target departure date, wherein each of the first historical flights in the first set of historical flights is not classified by season;
acquiring flight data for each of the first historical flights in the first historical flight set;
determining a weighted value of each of the first historical flights based on the N data pools and flight data of each of the first historical flights; and
A method for predicting seasonal classification of flights, comprising determining a seasonal classification of the target flight based on a weight of each of the first historical flights and a departure date of each of the first historical flights. The method of claim 1, wherein the step of determining the seasonal classification of the target flight based on the weight of each of the first historical flights and the number of departure days of each of the first historical flights includes,
determining a weighted sum of the first set of historical flights based on the weight of each of the first historical flights;
determining a total number of departure days for the first set of historical flights based on the number of departure days for each of the first historical flights; and
Method comprising determining a seasonal classification of the target flight based on the weighted sum and the total number of departure days. The method of claim 2, wherein the step of determining the seasonal classification of the target flight based on the weight sum and the total number of departure days includes:
Comparing the weighted sum and the total number of departure days to obtain a comparison result, wherein the comparison result represents a relationship between the weighted sum and the total number of departure days; and
A method comprising determining a seasonal classification of the target flight based on the comparison result. The method of claim 1, wherein determining a weight for each of the first historical flights based on the N data pools and flight data for each of the first historical flights comprises:
For each of the first historical flights, determining revenue data of the first historical flight based on flight data of the first historical flight;
Sorting the first revenue data into a first data pool, wherein the first revenue data is revenue data of the first historical flight, and the first data pool is the central data at the minimum distance from the first revenue data in the N data pools. Data pool with;
determining a preset weight of the first data pool; and
A method comprising determining a preset weight of the first data pool as the weight of a flight corresponding to the first revenue data. The method of any one of claims 1 to 4, wherein determining a first set of historical flights corresponding to the target flight based on the target departure date comprises:
Calculating the first date, second date, third date and fourth date using the following equations:
1st date = target departure date-52*7*i-1
2nd date = target departure date-52*7*i+1
3rd date = target departure date-51*7*i
4th date = target departure date-53*7*i
Here, i=(1, 2, 3..., n) and i represents the year before the current year; and
A method comprising determining flights corresponding to target flights on a first date, a second date, a third date, and a fourth date as a first set of historical flights. The method of any one of claims 1 to 4, wherein determining a first set of historical flights corresponding to the target flight based on the target departure date comprises:
Steps to calculate the target history base date using the following equation:
Target history base date = Target departure date-52*7*i
Here, i=(1, 2, 3..., n) and i represents the year before the current year;
Determining a first date, a second date, a third date, and a fourth date based on the target history reference date, where the first date is the day before the target history reference date, and the second date is the day after the target history reference date. , the third date is the same day as one week before the target history reference date, and the fourth date is the same day as one week after the target history reference date; and
A method comprising determining flights corresponding to target flights on a first date, a second date, a third date, and a fourth date as a first set of historical flights. The method of any one of claims 1 to 4, wherein the step of constructing N data pools corresponding to the target flights comprises:
Step 1, acquiring flight data for each second historical flight in a second historical flight set corresponding to the target flight;
Step 2, for each of the second historical flights, calculating revenue data of the second historical flight based on flight data of the second historical flight;
Step 3, determining the second revenue data as the central data of the second data pool, wherein the second revenue data is the revenue data of the first flight, the first flight is a flight in the second set of historical flights, and the second revenue data is the revenue data of the first flight, the first flight is a flight in the second set of historical flights, and A data pool is a data pool of N data pools;
Step 4, calculating the distance between the third revenue data and the second revenue data, where the third revenue data is the revenue data of the flights in the flights sub-set, and the flight sub-set is the revenue data of the flights in the second historical flight set. Set of flights excluding flight 1;
Step 5, sorting the fourth revenue data into a second data pool, where the fourth revenue data is the revenue data of the second flight, and the second flight is the revenue data closest to the second revenue data in the flight sub-population. Eligible flights; and
Step 6, calculating the central data of the second data pool after classification; and
The method comprising performing steps 3 to 6 until the revenue data for each of the second historical flights in the second set of historical flights is sorted into N data pools. an acquisition unit configured to obtain, from a local database, a target departure date of a target flight, which is a non-departure flight classified by season on the target airline;
a building unit configured to build N data pools corresponding to the target flights, where N represents a positive integer of 2 or more;
a first determination unit configured to determine a first set of historical flights corresponding to the target flight based on the target departure date, wherein each of the first historical flights in the first set of historical flights is not classified by season, and the acquisition unit is further configured to obtain flight data for each of the first historical flights in the first historical flight set;
a second determination unit configured to determine a weighted value of each of the first historical flights based on the N data pools and flight data of each of the first historical flights; and
A seasonal classification prediction mechanism for flights, comprising a third determination unit configured to determine a seasonal classification of the target flight based on the weight of each of the first historical flights and the number of departure days of each of the first historical flights. a memory configured to store a program;
a processor configured to execute the program in the memory and perform a method for predicting the classification of flights by season according to any one of claims 1 to 7 when executing a command to the program; and
A computer device comprising a bus system configured to couple a memory and a processor for communication between the memory and the processor. A machine-readable medium storing instructions, wherein, when executed by a machine, the instructions cause the machine to perform the method for predicting seasonal classification of flights according to any one of claims 1 to 7.