Flight seat sales condition evaluation method and device and electronic equipment

Abstract

The invention discloses a flight seat sales condition assessment method and device and electronic equipment, relates to the technical field of data processing, and aims to solve the problem of accurately assessing the sales condition of flight seats. The method comprises the steps of obtaining a current seat sales volume of a target flight, an estimated passenger volume corresponding to current time, and flight seat sales data of a plurality of reference flights; based on the estimated passenger quantity of the target flight and the flight seat sales data of the plurality of reference flights, determining a seat sales condition range of the target flight at the current time, the seat sales condition range being a range of seat sales quantity of the target flight at the current time; and determining whether the seat sales condition of the target flight at the current time is abnormal or not based on the current seat sales volume and the seat sales condition range of the target flight. Therefore, the sales condition of the flight seats is accurately evaluated, so that the operation benefit of an airline company is increased, and the customer experience is improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, and electronic device for evaluating flight seat sales. Background Technology

[0002] With the rapid development of the civil aviation industry, the management requirements for the flight seat sales process are constantly increasing. The rationality of flight seat sales is directly related to the airline's resource allocation, operational efficiency, and customer experience.

[0003] Currently, forecasting flight seat sales can only be done by processing and filtering historical data to analyze the occurrence of abnormal flights, but it lacks a dynamic response to market fluctuations. Furthermore, it often relies on a single indicator, making it difficult to accurately assess abnormal changes in flight seat sales.

[0004] Therefore, how to accurately assess the sales of flight seats in order to increase the operational efficiency of airlines and improve customer experience has become an urgent technical problem to be solved. Summary of the Invention

[0005] The purpose of this application is to provide a method, apparatus, and electronic device for evaluating flight seat sales, aiming to solve the problem of how to accurately evaluate flight seat sales in order to increase airline operating efficiency and improve customer experience.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, this application provides a method for assessing flight seat sales, including obtaining the current seat sales volume of a target flight, the estimated passenger volume at the current time, and flight seat sales data of multiple reference flights. Reference flights are those in historical flight data where the similarity between external influencing factors and the target flight's external influencing factors is greater than a similarity threshold. External influencing factors are used to indicate the influencing factors affecting the seat sales of the target flight. Based on the estimated passenger volume of the target flight and the flight seat sales data of multiple reference flights, the range of seat sales for the target flight at the current time is determined, where the range is the range of seat sales for the target flight at the current time. Based on the current seat sales volume and the range of seat sales for the target flight, it is determined whether the seat sales of the target flight at the current time are abnormal.

[0008] Based on the aforementioned technical means, this application identifies historical flights with a similarity greater than a similarity threshold as reference flights by comparing the external influencing factors of historical flights with those of the target flight. It then obtains seat sales data for multiple reference flights, providing a reference object and data basis for determining the range of seat sales for the target flight. This application obtains the estimated passenger volume for the target flight at the current time and, based on this estimated passenger volume and seat sales data from multiple reference flights, determines the range of seat sales at the current time, thus providing data support for evaluating the seat sales situation of the target flight. Finally, based on the current seat sales volume and the range of seat sales for the target flight, this application determines whether the seat sales situation of the target flight is abnormal at the current time, thereby accurately evaluating the seat sales situation, increasing the airline's operational efficiency and improving customer experience.

[0009] In one possible approach, flight seat sales data is used to indicate the difference between the number of seats sold for a reference flight at a reference time and the number of seats sold for a reference flight at its departure time, where the time between the reference time and the departure time of the reference flight is equal to the time between the current time and the departure time of the target flight. Based on the estimated passenger volume of the target flight and the flight seat sales data of multiple reference flights, the range of seat sales for the target flight at the current time is determined, including: determining statistical characteristic data of the reference flights based on the flight seat sales data of multiple reference flights, whereby the statistical characteristic data of the reference flights is used to reflect the distribution pattern of the difference in seat sales; and determining the range of seat sales for the target flight at the current time based on the statistical characteristic data of the reference flights and the estimated passenger volume of the target flight.

[0010] Based on the aforementioned technical means, this application determines the statistical characteristics of multiple reference flights based on their sales data, thereby identifying the distribution pattern of the differences in seat sales volume among the reference flights. This provides data support for subsequently determining the range of seat sales for the target flight at the current time. Furthermore, based on the statistical characteristics of the reference flights and the estimated passenger volume of the target flight, the range of seat sales for the target flight at the current time is accurately determined, providing a precise judgment standard for subsequent sales assessment.

[0011] In one possible approach, the reference flight statistical characteristic data includes: a reference mean for seat sales and a reference standard deviation for seat sales. The reference mean for seat sales is the average of seat sales data for multiple flights, and the reference standard deviation for seat sales is the standard deviation of seat sales data for multiple flights. Based on the seat sales data for multiple flights and the estimated passenger volume of the target flight, the range of seat sales for the target flight at the current time is determined. This includes: determining the range of seat sales for the target flight at the current time based on the estimated passenger volume of the target flight, the reference mean for seat sales, the reference standard deviation for seat sales, and a range correction factor. The range correction factor is used to indicate the impact of the time length from the current time to the departure time of the target flight on the range of seat sales.

[0012] Based on the aforementioned technical means, this application uses the average of seat sales data from multiple reference flights as the reference mean for seat sales, and the standard deviation of seat sales data from multiple reference flights as the reference standard deviation for seat sales, thereby providing data support for subsequent analysis. Based on the estimated passenger volume of the target flight, the reference mean for seat sales, the reference standard deviation for seat sales, and a range correction coefficient, this application comprehensively considers multiple parameters to accurately determine the range of seat sales for the target flight at the current time.

[0013] In one possible approach, the range of seat sales includes a preset upper limit and / or a preset lower limit; the preset upper limit satisfies the following first preset formula:

[0014] Q1 = Z - (m - ke);

[0015] Where Q1 represents the preset upper limit value, Z represents the estimated passenger volume, m represents the reference average value of seat sales, e represents the reference standard deviation value of seat sales, and k represents the correction factor.

[0016] The preset lower limit value satisfies the following second preset formula:

[0017] A2 = Z - (m + ke);

[0018] Where Q2 represents the preset lower limit value, Z represents the estimated passenger volume, m represents the reference average value for seat sales, e represents the reference standard deviation value for seat sales, and k represents the correction coefficient.

[0019] In one possible approach, if the difference between the departure time of the target flight and the current time is greater than a first preset threshold, the range correction coefficient is the first correction coefficient; if the difference between the departure time of the target flight and the current time is less than or equal to the first preset threshold and greater than or equal to a second preset threshold, the range correction coefficient is the second correction coefficient; if the difference between the departure time of the target flight and the current time is less than the second preset threshold, the range correction coefficient is the third correction coefficient; the first correction coefficient is greater than the second correction coefficient, and the second correction coefficient is greater than the third correction coefficient.

[0020] Based on the aforementioned technical means, this application defines the following scenarios: When the difference between the departure time of the target flight and the current time is greater than a first preset threshold, the flight's sales are considered to be in the ticketing stage, and the range correction coefficient is the first correction coefficient, thus relaxing the evaluation criteria. When the difference between the departure time of the target flight and the current time is less than or equal to the first preset threshold, but greater than or equal to a second preset threshold, the flight's sales are considered to be in the middle of the ticketing stage, and the range correction coefficient is the second correction coefficient, thus tightening the evaluation criteria. When the difference between the departure time of the target flight and the current time is less than the second preset threshold, the range correction coefficient is the third correction coefficient, indicating the end of the ticketing stage, thus strictly controlling the evaluation criteria. The first correction coefficient is greater than the second correction coefficient, which is greater than the third correction coefficient. This achieves dynamic evaluation of the ticketing situation of the target flight at different stages.

[0021] In one possible approach, the seat sales situation of the target flight is determined based on the current seat sales volume and the range of seat sales, including: if the current seat sales volume is less than or equal to a preset upper limit and greater than or equal to a preset lower limit, the seat sales situation of the target flight is considered normal; if the current seat sales volume is greater than the preset upper limit or less than the preset lower limit, the seat sales situation of the target flight is considered abnormal.

[0022] Based on the aforementioned technical means, this application accurately assesses the current seat sales volume of a target flight by comparing the current seat sales volume with the upper and lower limits of the seat sales range.

[0023] One possible approach involves obtaining the estimated passenger volume at the current time, including: obtaining the target flight's preset route, the market share of multiple preset itineraries, and the flight segment information corresponding to each preset itinerary. The preset route includes multiple preset itineraries, and a flight segment indicates the smallest itinerary unit within a preset itinerary; one flight segment corresponds to one preset flight. Based on the preset route, determine the time-series characteristics of the route relationship graph and the total number of passengers at the current time. The route relationship graph indicates the influence relationship between the preset route and adjacent routes, and the time-series characteristics of the total number of passengers at the current time indicate the trend of the total number of passengers at the current time changing over time. Input the route relationship graph and the time-series characteristics of the total number of passengers at the current time into a trained passenger prediction model to determine the estimated passenger volume of the preset route. Based on the estimated passenger volume of the preset route and the market share of multiple preset itineraries, determine the estimated passenger volume of each preset itinerary. Based on the estimated passenger volume of each preset itinerary and the flight segment information corresponding to each preset itinerary, determine the estimated passenger volume of the target flight.

[0024] Based on the aforementioned technical means, this application obtains the preset route and flight competitiveness of the target flight, and determines the time series characteristics of the route relationship diagram and the total number of passengers at the current time based on the preset route, thereby providing data support for subsequent passenger volume prediction of the target flight. Furthermore, the time series characteristics of the route relationship diagram and the total number of passengers at the current time are input into the trained passenger prediction model, thereby accurately determining the total number of passengers on the preset route. And based on the total number of passengers and flight competitiveness, the estimated passenger volume of the target flight is accurately determined.

[0025] In one possible approach, the trained passenger prediction model includes graph convolutional layers and fully connected network layers. By using the trained passenger prediction model, the route relationship graph, and the time series features of passenger volume on the preset route, the total number of passengers on the preset route is determined. This includes: extracting features from the route relationship graph using graph convolutional layers to determine the relationship feature matrix; and processing the relationship feature matrix and the time series features of the total number of passengers at the current time using fully connected network layers to determine the total number of passengers on the preset route.

[0026] Based on the aforementioned technical means, this application extracts features from the route relationship graph using graph convolutional layers to determine the relationship feature matrix, thereby accurately identifying the influence features between various routes in the route relationship graph. Furthermore, a fully connected network layer processes the relationship feature matrix and the time-series features of the total number of passengers at the current time, thereby accurately predicting the total number of passengers on the preset routes.

[0027] Secondly, this application provides an assessment device for flight seat sales, comprising: an acquisition module for acquiring the current seat sales volume of a target flight, the estimated passenger volume at the current time, external influencing factors, and flight seat sales data of multiple reference flights, wherein the external influencing factors are used to indicate the influencing factors affecting the seat sales of the target flight, and the reference flights are flights in historical flights whose external influencing factors have a similarity greater than a similarity threshold with the external influencing factors of the target flight; a processing module for determining the range of seat sales of the target flight at the current time based on the estimated passenger volume of the target flight, the external influencing factors, and the flight seat sales data of multiple reference flights, wherein the range of seat sales is the range of the number of seats sold for the target flight at the current time; and the processing module is further configured to determine whether the seat sales of the target flight at the current time are abnormal based on the current seat sales volume and the range of seat sales.

[0028] In one possible approach, flight seat sales data is used to indicate the difference between the number of seats sold for a reference flight at a reference time and the number of seats sold for a reference flight at its departure time. The time length between the reference time and the departure time of the reference flight is equal to the time length between the current time and the departure time of the target flight. The processing module is specifically used to determine the statistical characteristic data of the reference flights based on the flight seat sales data of multiple reference flights. The statistical characteristic data of the reference flights is used to reflect the distribution pattern of the difference in the number of seats sold. Based on the statistical characteristic data of the reference flights and the estimated passenger volume of the target flight, the range of seat sales for the target flight at the current time is determined.

[0029] In one possible approach, the reference flight statistical characteristic data includes: the reference mean seat sales and the reference standard deviation of seat sales. The reference mean seat sales is the average of seat sales data for multiple flights, and the reference standard deviation of seat sales is the standard deviation of seat sales data for multiple flights. The processing module is specifically used to determine the range of seat sales for the target flight at the current time based on the estimated passenger volume of the target flight, the reference mean seat sales, the reference standard deviation of seat sales, and the range correction coefficient. The range correction coefficient is used to indicate the impact of the time length from the current time to the departure time of the target flight on the range of seat sales.

[0030] In one possible approach, the range of seat sales includes a preset upper limit and / or a preset lower limit; the preset upper limit satisfies the following first preset formula:

[0031] Q1 = Z - (m - ke);

[0032] Where Q1 represents the preset upper limit value, Z represents the estimated passenger volume, m represents the reference average value of seat sales, e represents the reference standard deviation value of seat sales, and k represents the correction factor.

[0033] The preset lower limit value satisfies the following second preset formula:

[0034] Q2 = Z - (m + ke);

[0035] Where Q2 represents the preset lower limit value, Z represents the estimated passenger volume, m represents the reference average value for seat sales, e represents the reference standard deviation value for seat sales, and k represents the correction coefficient.

[0036] In one possible approach, the processing module is specifically configured to: use a first correction coefficient when the difference between the departure time of the target flight and the current time is greater than a first preset threshold; use a second correction coefficient when the difference between the departure time of the target flight and the current time is less than or equal to the first preset threshold and greater than or equal to a second preset threshold; and use a third correction coefficient when the difference between the departure time of the target flight and the current time is less than the second preset threshold. The first correction coefficient is greater than the second correction coefficient, and the second correction coefficient is greater than the third correction coefficient.

[0037] In one possible approach, the processing module is specifically configured to consider the seat sales situation of the target flight as normal when the current seat sales volume is less than or equal to a preset upper limit and greater than or equal to a preset lower limit; and to consider the seat sales situation of the target flight as abnormal when the current seat sales volume is greater than the preset upper limit or less than the preset lower limit.

[0038] In one possible approach, the acquisition module is specifically used to acquire the target flight's preset route, the market share of multiple preset itineraries, and the flight segment information corresponding to each preset itinerary. The preset route includes multiple preset itineraries, and the flight segment indicates the smallest itinerary unit in the preset itinerary, with one flight segment corresponding to one preset flight. Based on the preset route, the module determines the time series characteristics of the route relationship graph and the total number of passengers at the current time. The route relationship graph indicates the influence relationship between the preset route and adjacent routes, and the time series characteristics of the total number of passengers at the current time indicate the trend of the total number of passengers at the current time changing over time. The module inputs the route relationship graph and the time series characteristics of the total number of passengers at the current time into a trained passenger prediction model to determine the estimated passenger volume of the preset route. Based on the estimated passenger volume of the preset route and the market share of multiple preset itineraries, the module determines the estimated passenger volume of each preset itinerary. Based on the estimated passenger volume of each preset itinerary and the flight segment information corresponding to each preset itinerary, the module determines the estimated passenger volume of the target flight.

[0039] In one possible approach, the trained passenger prediction model includes graph convolutional layers and fully connected network layers. The processing module is specifically used to extract features from the route relationship graph through the graph convolutional layers to determine the relationship feature matrix; and to process the relationship feature matrix and the time series features of the current time preset route passenger volume through the fully connected network layers to determine the total number of passengers on the preset route.

[0040] Thirdly, this application provides an electronic device including a memory and a processor; the memory and the processor are coupled; the memory is used to store instructions executable by the processor; when the processor executes the instructions, it performs the methods described in the first aspect and any possible implementation thereof.

[0041] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the methods described in the first aspect and any possible implementation thereof.

[0042] Fifthly, this application provides a chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run computer programs or instructions to implement the methods described in the first aspect and any possible implementation thereof.

[0043] Sixthly, this application provides a computer program product containing instructions that, when executed by a computer, cause the computer to perform the methods described in the first aspect and any possible implementation thereof.

[0044] The technical problems that can be solved and the technical effects that can be achieved by the assessment device, electronic device, computer storage medium, chip or computer program product for flight seat sales in the above solution can be referred to the technical problems and technical effects solved in the first aspect above, and will not be repeated here. Attached Figure Description

[0045] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 A flowchart illustrating a method for evaluating flight seat sales, provided as an embodiment of this application;

[0047] Figure 2 A flowchart illustrating yet another method for evaluating flight seat sales, provided as an embodiment of this application;

[0048] Figure 3 A flowchart illustrating yet another method for evaluating flight seat sales, provided as an embodiment of this application;

[0049] Figure 4 This is a structural diagram of an assessment device for flight seat sales provided in an embodiment of this application. Detailed Implementation

[0050] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] In the description of this application, it should be understood that the terms "upper," "lower," "left," "right," "front," "rear," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or relative positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and for simplification, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Unless otherwise specified, the above-mentioned orientational descriptions can be flexibly set in practical applications, provided that the relative positional relationships shown in the accompanying drawings are satisfied.

[0052] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0053] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "communication" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection. They can refer to a direct connection or an indirect connection through an intermediate medium, or a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0054] In embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, article, or apparatus that includes that element.

[0055] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0056] In the description of this specification, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0057] In some embodiments, such as Figure 1 As shown, this application provides a method for evaluating the sales of flight seats, specifically including: S101-S103.

[0058] S101. Obtain the current seat sales volume of the target flight, the estimated passenger volume at the current time, and the seat sales data of multiple reference flights.

[0059] The reference flight refers to a flight in the historical flight data whose external influencing factors are more similar to those of the target flight than a similarity threshold. External influencing factors are used to indicate the factors that affect seat sales on the target flight.

[0060] For example, external influencing factors may include flight attributes, such as route type, for instance, the number of seats sold for flights from first-tier cities to second-tier cities being greater than the number of seats sold for flights from first-tier cities to second-tier cities. Temporal context, such as holidays or peak / off-peak seasons, and market context, such as the impact of relevant policies, may also be considered.

[0061] In one possible implementation, preset external influencing factors are set, and data on these factors for both the target and candidate flights are acquired. The similarity between the target and candidate flights is determined using a cosine similarity calculation formula and the external influencing factor data. Candidate flights with similarity scores greater than a preset similarity threshold are designated as reference flights. Based on multiple reference flights, seat availability data for each reference flight is determined.

[0062] Among them, candidate flights are historical flights with the same route type as the target flight and whose time difference is within a preset range.

[0063] It should be noted that the aforementioned external factors are merely illustrative, and this application uses only these influencing factors as examples, without imposing specific limitations on the types of influencing factors. In actual use, influencing factors can be set according to specific circumstances. The cosine similarity calculation formula treats the external influencing factors of the target flight as a vector and the external influencing factors of the candidate flight as a vector. By calculating the angle between the two vectors, the similarity between the target flight and the candidate flight is determined.

[0064] In another possible implementation, the device for assessing flight seat sales stores seat sales data for multiple flights. Based on the identifier of the target flight, the current seat sales volume of the target flight is determined.

[0065] S102. Based on the estimated passenger volume of the target flight and the seat sales data of multiple reference flights, determine the range of seat sales for the target flight at the current time.

[0066] The seat sales range refers to the range of seat sales for the target flight at the current time. Flight seat sales data is used to indicate the difference between the number of seats sold for the reference flight at the reference time and the number of seats sold for the reference flight at the departure time.

[0067] It should be noted that the time between the reference time and the departure time of the reference flight is equal to the time between the current time and the departure time of the target flight. The time between the reference time and the departure time of the reference flight can also be expressed as the time remaining between the reference time and the departure time of the reference flight. Similarly, the time between the current time and the departure time of the target flight can be expressed as the time remaining between the departure time of the target flight.

[0068] One possible implementation involves determining statistical characteristics of the reference flights based on seat sales data from multiple reference flights. Then, based on these statistical characteristics and the estimated passenger volume of the target flight, the range of seat sales for the target flight at the current time is determined.

[0069] For example, the maximum difference between the number of seats sold for a reference flight during a reference time period and the number of seats sold for a reference flight at the departure time can be obtained from the seat sales data of multiple reference flights as the statistical feature data of the reference flight.

[0070] It should be noted that the above examples of statistical characteristic data are only one possible implementation method.

[0071] The range of seat sales includes a preset upper limit and / or a preset lower limit.

[0072] S103. Based on the current seat sales volume and the range of seat sales for the target flight, determine whether the seat sales situation of the target flight at the current time is abnormal.

[0073] In one possible implementation, if the current seat sales volume is less than or equal to a preset upper limit and greater than or equal to a preset lower limit, the seat sales situation of the target flight is considered normal.

[0074] For example, with a preset upper limit of 200, a preset lower limit of 100, and a current seat sales volume of 150, the current seat sales volume is less than or equal to the preset upper limit and greater than or equal to the preset lower limit, indicating that the seat sales situation for the target flight is normal.

[0075] If the current seat sales volume exceeds the preset upper limit, the seat sales situation of the target flight is considered oversold.

[0076] For example, with a preset upper limit of 200 and a current seat sales volume of 280, the current seat sales volume exceeds the preset upper limit, indicating that the target flight is oversold.

[0077] If the current seat sales volume is less than the preset lower limit, the sales situation of the target flight is considered unsold.

[0078] For example, with a preset lower limit of 100 and a current seat sales volume of 50, the current seat sales volume is less than the preset lower limit, indicating that the target flight's seat sales are unsold.

[0079] As can be seen from the above technical solution, this application determines historical flights with a similarity greater than a similarity threshold as reference flights based on the similarity between external influencing factors of historical flights and those of the target flight. It then obtains seat sales data for multiple reference flights, providing a reference object and data basis for determining the range of seat sales for the target flight. This application obtains the estimated passenger volume for the target flight at the current time and, based on the estimated passenger volume and seat sales data of multiple reference flights, determines the range of seat sales at the current time, thus providing data support for evaluating the seat sales of the target flight. Based on the current seat sales volume and the range of seat sales for the target flight, this application determines whether the seat sales of the target flight are abnormal at the current time, thereby accurately evaluating the seat sales situation, increasing the airline's operational efficiency and improving customer experience.

[0080] In some embodiments, such as Figure 2As shown, in the above method S102, the range of seat sales for the target flight at the current time is determined based on the estimated passenger volume of the target flight and the seat sales data of multiple reference flights, specifically including: S201-S202.

[0081] S201. Based on seat sales data of multiple reference flights, determine the statistical characteristics of the reference flights.

[0082] The reference flight statistics include the reference mean and reference standard deviation of seat sales. The reference mean is the average of seat sales data for multiple flights, and the reference standard deviation is the standard deviation of seat sales data for multiple flights.

[0083] In one possible implementation, the average seat sales data of multiple reference flights is determined as the reference mean seat sales. The standard deviation of the seat sales data of the multiple reference flights is then determined as the reference standard deviation seat sales.

[0084] For example, the reference average for seat sales satisfies the following formula:

[0085]

[0086] in, The value is used to represent the average sales reference, where n represents the number of reference flights, and x represents the average number of flights. i Used to represent seat sales data for the i-th flight.

[0087] The unbiased variance of seat sales reference satisfies the following formula:

[0088]

[0089] Among them, s 2 The value used to represent the unbiased variance of seat sales reference, where n represents the number of reference flights, and x represents the unbiased variance. i Used to represent seat sales data for the i-th flight. Used to represent the average sales reference value.

[0090] The reference standard deviation for seat sales satisfies the following formula three:

[0091]

[0092] Where e represents the reference standard deviation of seat sales, s 2 Used to represent the unbiased variance of seat sales reference.

[0093] S202. Based on the statistical characteristics of the reference flight and the estimated passenger volume of the target flight, determine the range of seat sales for the target flight at the current time.

[0094] In one possible implementation, the range of seat sales for the target flight at the current time is determined based on the estimated passenger volume of the target flight, the reference mean of seat sales, the reference standard deviation of seat sales, and the range correction factor.

[0095] It should be noted that when determining the range of seat sales, you can set only one preset upper limit or one preset lower limit, or you can set both preset upper and lower limits at the same time.

[0096] For example, the preset upper limit value satisfies the following formula four:

[0097] Q1 = Z - (m - ke) Formula 4.

[0098] Where Q1 represents the preset upper limit value, Z represents the estimated passenger volume, m represents the reference average value for seat sales, e represents the reference standard deviation value for seat sales, and k represents the range correction coefficient.

[0099] The preset lower limit value satisfies the following formula five:

[0100] Formula 5: q2 = Z - (m + ke).

[0101] Where Q2 represents the preset lower limit value, Z represents the estimated passenger volume, m represents the reference average value for seat sales, e represents the reference standard deviation value for seat sales, and k represents the range correction coefficient.

[0102] It should be noted that the correction factor is used to adjust the upper and lower limits of seat sales, and the range correction factor is proportional to the length of time between the target flight's departure time and the actual departure time.

[0103] The range correction factor is used to indicate the time length from the current time to the departure time of the target flight and its impact on the range of seat sales.

[0104] It should be noted that the assessment of seat sales for a target flight involves several stages: the initial stage (shortly before ticket sales begin), the middle stage (some time after ticket sales have started), and the final stage (soon to close). The range of seat sales data needs to be dynamically adjusted for each stage to ensure a more accurate assessment of the sales situation.

[0105] For example, consider the initial stage of ticket sales. Shortly after ticket sales begin for the target flight, passenger demand is low, resulting in a low number of seats sold. Therefore, it is necessary to adjust the range correction parameters to broaden the scope of seat sales, lower the criteria for judging unsold seats, and improve the accuracy of the assessment.

[0106] As can be seen from the above technical solution, this application uses the average value of seat sales data from multiple reference flights as the reference mean value for seat sales, and the standard deviation of seat sales data from multiple reference flights as the reference standard deviation value for seat sales, thereby providing data support for subsequent analysis. Based on the estimated passenger volume of the target flight, the reference mean value for seat sales, the reference standard deviation value for seat sales, and the range correction coefficient, this application comprehensively considers multiple parameters to accurately determine the range of seat sales for the target flight at the current time.

[0107] In some embodiments, when the difference between the departure time of the target flight and the current time is greater than a first preset threshold, the range correction factor is the first correction factor.

[0108] For example, taking a first preset threshold of 90 days and a difference of 100 days between the departure time of the target flight and the current time as an example. If the difference between the departure time of the target flight and the current time is greater than 90 days, the range correction factor is the first correction factor. This raises the lower limit and lowers the criterion for judging sluggish sales.

[0109] In one possible implementation, if the difference between the departure time of the target flight and the current time is less than or equal to a first preset threshold and greater than or equal to a second preset threshold, the range correction coefficient is the second correction coefficient.

[0110] For example, with a first preset threshold of 90 days, a second preset threshold of 30 days, and a difference of 45 days between the departure time of the target flight and the current time, the range correction factor is the second correction factor if the departure time of the target flight is less than or equal to the first preset threshold and greater than or equal to the second preset threshold. This narrows the range of seat sales and strengthens the evaluation criteria for the number of seats available for the target flight.

[0111] In another possible implementation, if the difference between the departure time of the target flight and the current time is less than the second preset threshold, the range correction factor is the third correction factor.

[0112] For example, taking a second preset threshold of 30 days and a difference of 10 days between the departure time of the target flight and the current time. If the difference between the departure time of the target flight and the current time is less than the second preset threshold, the range correction factor is the third correction factor. At this point, the target flight is in the final stage of ticket sales, thereby further narrowing the range of seat sales and strictly monitoring the number of seats sold for the target flight.

[0113] As can be seen from the above technical solution, when the difference between the departure time of the target flight and the current time is greater than a first preset threshold, the sales of the target flight are in the ticketing stage, and the range correction coefficient is the first correction coefficient, thus relaxing the evaluation criteria. When the difference between the departure time of the target flight and the current time is less than or equal to the first preset threshold, but greater than or equal to the second preset threshold, the sales of the target flight are in the middle of the ticketing stage, and the range correction coefficient is the second correction coefficient, thus tightening the evaluation criteria. When the difference between the departure time of the target flight and the current time is less than the second preset threshold, the range correction coefficient is the third correction coefficient, indicating that the flight is in the final stage of ticketing, thus strictly controlling the evaluation criteria. The first correction coefficient is greater than the second correction coefficient, which is greater than the third correction coefficient. This achieves dynamic evaluation of the ticketing situation of the target flight at different stages.

[0114] In some embodiments, when the seat sales situation for a target flight is abnormal at the current time, the flight seat sales assessment device sends an early warning signal to staff. Staff then handle the abnormality and upload a processing report to the flight seat sales assessment device. Upon receiving the processing report, the assessment device analyzes it to identify the anomaly factors. Based on these factors, it adjusts the model parameters.

[0115] It should be noted that abnormal factors are influencing factors that cause abnormal seat sales. These abnormal factors can be added as external influencing factors in method S101 when analyzing the target flight, so that the flight seat sales assessment system can continuously learn and update based on the processing report, thereby improving the accuracy of seat sales assessment.

[0116] In some embodiments, such as Figure 3 As shown, the estimated passenger volume in method S101 is specifically obtained through S301-S304.

[0117] S301. Obtain the preset route of the target flight, the market share of multiple preset itineraries, and the flight segment information corresponding to each preset itinerary.

[0118] The preset route includes multiple preset itineraries. A flight segment is used to indicate the smallest travel unit in a preset itinerary, which can also be understood as each takeoff and landing of a flight segment.

[0119] For example, consider a preset route from City A to City B. The preset route might include three itineraries: "from Airport x in City A to Airport y in City B," "from Airport x in City A to Airport y in City C, then from Airport y in City C to Airport z in City B," and "from Airport x in City A to Airport n in City D, then from Airport n in City D to Airport z in City B." The itinerary "from Airport x in City A to Airport y in City C, then from Airport y in City C to Airport z in City B" comprises two segments: "from Airport x in City A to Airport y in City C" and "from Airport y in City C to Airport z in City B." Each segment is completed by one flight.

[0120] It should be noted that the preset route is used to indicate the flight path, such as from point A to point B. It should be understood that there are multiple flights from point A to point B, and the total number of passengers on all these flights is the total number of passengers on the preset route.

[0121] One possible implementation involves obtaining multiple preset itineraries within a preset route and multiple passenger preference attributes for each preset itinerary. For each preset itinerary, a weighted calculation of each passenger preference attribute is performed and then summed to determine the competitiveness of each preset itinerary. Based on the market share calculation formula and the competitiveness of each preset itinerary, the market share of each preset itinerary is determined.

[0122] Among them, passenger preference attributes indicate the various attributes that influence passengers' choice of itinerary. Competitiveness indicates the degree of passenger preference for a preset itinerary.

[0123] It should be noted that this application does not impose specific restrictions on the aforementioned passenger preference attributes, which can be selected according to the actual situation. For details on determining itinerary competitiveness, please refer to existing technologies, which will not be elaborated here.

[0124] For example, taking one of a plurality of preset trips as the target trip, the market share of the target trip satisfies the following formula six:

[0125]

[0126] Here, Share indicates the market share of the target trip, U j The value is used to represent the competitiveness of the j-th trip in the preset itinerary, and e is used to represent the natural logarithm.

[0127] S302. Based on the preset routes of the target flights, determine the time series characteristics of the route relationship diagram and the total number of passengers at the current time.

[0128] The route relationship diagram is used to indicate the influence relationship between preset routes and adjacent routes. The time series feature of the total number of passengers at the current time is used to indicate the trend of the total number of passengers at the current time over time.

[0129] For example, adjacent routes can be routes with the same origin and / or destination as the preset routes or with a geographical distance of less than a preset distance threshold (e.g., routes from Beijing to Shanghai and routes from Beijing to Nantong are adjacent routes), or routes whose origin is the destination of the preset routes are adjacent routes (e.g., routes from Beijing to Shanghai and routes from Shanghai to Guangzhou), etc.

[0130] It should be noted that the specific adjacent routes mentioned above are merely illustrative and this application does not impose any specific limitations. It should be understood that there needs to be an influence relationship between the preset route and the adjacent routes.

[0131] One possible implementation involves obtaining multiple adjacent routes of the preset route and the influence strength between the preset route and its adjacent routes. A route relationship graph is then constructed using the preset route and its adjacent routes as nodes, and the influence strength between routes as directed edges.

[0132] In the route relationship graph, the weight of the edge represents the intensity of the influence between routes.

[0133] Another possible implementation involves obtaining the total number of passengers for multiple preset time periods prior to the current time. These total passenger counts are then input into a feature extraction model to determine the time-series feature of the total number of passengers at the current time.

[0134] It should be noted that the feature extraction model is built upon a Long Short-Term Memory (LSTM) network model. The initial feature extraction model is trained using a training dataset to obtain the trained model. The training set includes the total number of passengers for multiple preset historical time periods, as well as time-series features of the total number of passengers for those same time periods. The trained feature extraction model can then determine the time-series features of the total number of passengers at the current time based on the total number of passengers for multiple preset time periods prior to the current time.

[0135] S303. Input the route relationship diagram and the time series characteristics of the passenger volume of the preset route at the current time into the trained passenger prediction model to determine the estimated passenger volume of the preset route.

[0136] The trained passenger prediction model includes graph convolutional layers and fully connected network layers.

[0137] In one possible implementation, features are extracted from the route relationship graph using graph convolutional layers to determine the relationship feature matrix. Then, a fully connected network layer processes the relationship feature matrix and the time-series features of passenger volume on the preset route at the current time to determine the total number of passengers on the preset route.

[0138] Specifically, feature extraction is performed on the network relationship characteristics between each node in the route relationship graph using graph convolutional layers to determine the relationship feature matrix. Then, the relationship feature matrix is ​​multiplied by the time-series features of the total number of passengers at the current time to determine the fusion feature matrix. The time-series feature information of the feature fusion matrix is ​​extracted using a gating mechanism in a fully connected network layer, and weights and bias coefficients are assigned to determine the estimated passenger volume for the preset route.

[0139] The estimated passenger volume for the preset route is the number of people in the origin-destination (OD) market demand based on the preset route. In other words, it is the number of people in the market who have travel demand for the preset route from origin to destination.

[0140] It should be noted that the trained passenger prediction model is built based on a Graph Convolutional Network-Long Short-Term Memory (GCN-LSTM) neural network model. This application only describes the input and output of the passenger prediction model; the specific data processing flow inside the model can be found in existing technologies and will not be elaborated here.

[0141] S304. Based on the market demand for the preset routes and the market share of multiple preset itineraries, determine the estimated passenger volume for each preset itinerary.

[0142] One possible implementation involves multiplying the market demand for passengers by the market share of each preset trip to determine the estimated number of passengers for each preset trip.

[0143] For example, the preset estimated passenger volume satisfies the following formula seven:

[0144] N = N p ×Share Formula 7.

[0145] Where N represents the estimated number of passengers for the preset itinerary. p The term "sjare" is used to represent the market demand for a preset route, while "sjare" represents the market share of a preset itinerary.

[0146] S305. Based on the estimated passenger volume of each preset itinerary and the corresponding flight section information, determine the estimated passenger volume of the target flight.

[0147] In one possible implementation, based on the estimated passenger volume of each preset itinerary and the flight segment information corresponding to each itinerary, the number of passengers on all flight segments that include the target flight in all preset itineraries is aggregated to determine the estimated passenger volume of the target flight.

[0148] For example, taking the flight segment "City A - City B" as the target flight, the estimated passenger volume of the preset itinerary "City A - City B" and the estimated passenger volume of the preset itinerary "City A - City B - City C" are both part of the estimated passenger volume of the target flight. That is, the estimated passenger volume of the target flight is determined by summing the estimated passenger volumes of all preset itineraries that include the target flight.

[0149] In some embodiments, in method S302 above, the trained passenger prediction model is trained in the following manner.

[0150] In one possible implementation, a training dataset is obtained. This dataset is then input into the passenger prediction model to obtain prediction results. Based on the prediction results and the actual results, a prediction error value is determined. If the prediction error value is greater than a preset threshold, the passenger prediction model is iteratively trained until the prediction error value is less than or equal to the preset threshold, at which point training stops.

[0151] The training dataset includes a route relationship graph within a preset historical time period and time series features of passenger volume on preset routes.

[0152] It should be noted that the prediction error value in the above method can be evaluated by selecting evaluation indicators such as mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) to iteratively train the passenger prediction model and evaluate the prediction performance of the model.

[0153] As can be seen from the above technical solution, this application extracts features from the route relationship graph through graph convolutional layers to determine the relationship feature matrix, thereby accurately determining the influence features between various routes in the route relationship graph. Furthermore, it processes the relationship feature matrix and the time-series features of the total number of passengers at the current time through fully connected network layers, thereby accurately predicting the total number of passengers on the preset routes.

[0154] As described in the above technical solution, this application obtains the preset route and flight competitiveness of the target flight, and based on the preset route, determines the time series characteristics of the route relationship diagram and the total number of passengers at the current time, thereby providing data support for subsequent passenger volume prediction of the target flight. Furthermore, the time series characteristics of the route relationship diagram and the total number of passengers at the current time are input into the trained passenger prediction model, thereby accurately determining the total number of passengers on the preset route. And based on the total number of passengers and flight competitiveness, the estimated passenger volume of the target flight is accurately determined.

[0155] The foregoing primarily describes the solutions provided in the embodiments of this application from a methodological perspective. It is understood that the flight seat sales assessment device, in order to achieve the aforementioned functions, includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the flight seat sales assessment method steps described in conjunction with the embodiments disclosed in this application, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0156] This application also provides an apparatus for evaluating flight seat sales. This apparatus can be a server, a CPU within the server, a module within the server for evaluating flight seat sales, or a client within the server for evaluating flight seat sales.

[0157] This application embodiment can divide the flight seat sales assessment device into functional modules or functional units according to the above method example. For example, each function can be divided into a separate functional module or functional unit, or two or more functions can be integrated into one processing unit. The integrated module can be implemented in hardware or in software functional modules or functional units. The module or unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0158] When dividing each function into modules according to its corresponding function. Figure 4 A structural diagram of an assessment device for flight seat sales provided in this application is shown below. Figure 4 As shown, the device for evaluating flight seat sales can be used to perform... Figure 1 , Figure 2 , Figure 3The method for evaluating flight seat sales is shown. The flight seat sales evaluation device 40 includes an acquisition module 401 and a processing module 402.

[0159] In one possible approach, flight seat sales data is used to indicate the difference between the number of seats sold for a reference flight at a reference time and the number of seats sold for a reference flight at its departure time. The time length between the reference time and the departure time of the reference flight is equal to the time length between the current time and the departure time of the target flight. Processing module 402 is specifically used to determine reference flight statistical characteristic data based on the flight seat sales data of multiple reference flights. The reference flight statistical characteristic data is used to reflect the distribution pattern of the difference in seat sales. Based on the reference flight statistical characteristic data and the estimated passenger volume of the target flight, the range of seat sales for the target flight at the current time is determined.

[0160] In one possible approach, the reference flight statistical characteristic data includes: the reference mean seat sales and the reference standard deviation of seat sales. The reference mean seat sales is the average of seat sales data for multiple flights, and the reference standard deviation of seat sales is the standard deviation of seat sales data for multiple flights. The processing module 402 is specifically used to determine the range of seat sales for the target flight at the current time based on the estimated passenger volume of the target flight, the reference mean seat sales, the reference standard deviation of seat sales, and the range correction coefficient. The range correction coefficient is used to indicate the impact of the time length from the current time to the departure time of the target flight on the range of seat sales.

[0161] In one possible approach, the range of seat sales includes a preset upper limit and / or a preset lower limit; the preset upper limit satisfies the following first preset formula:

[0162] Q1 = Z - (m - ke);

[0163] Where Q1 represents the preset upper limit value, Z represents the estimated passenger volume, m represents the reference average value of seat sales, e represents the reference standard deviation value of seat sales, and k represents the correction factor.

[0164] The preset lower limit value satisfies the following second preset formula:

[0165] Q2 = Z - (m + ke);

[0166] Where Q2 represents the preset lower limit value, Z represents the estimated passenger volume, m represents the reference average value for seat sales, e represents the reference standard deviation value for seat sales, and k represents the correction coefficient.

[0167] In one possible approach, the processing module 402 is specifically configured to: use a first correction coefficient when the difference between the departure time of the target flight and the current time is greater than a first preset threshold; use a second correction coefficient when the difference between the departure time of the target flight and the current time is less than or equal to the first preset threshold and greater than or equal to a second preset threshold; and use a third correction coefficient when the difference between the departure time of the target flight and the current time is less than the second preset threshold. The first correction coefficient is greater than the second correction coefficient, and the second correction coefficient is greater than the third correction coefficient.

[0168] In one possible approach, the processing module 402 is specifically used to determine that the seat sales situation of the target flight is normal when the current seat sales volume is less than or equal to a preset upper limit and greater than or equal to a preset lower limit; and to determine that the seat sales situation of the target flight is abnormal when the current seat sales volume is greater than the preset upper limit or less than the preset lower limit.

[0169] In one possible approach, the acquisition module 401 is specifically used to acquire the preset route of the target flight, the market share of multiple preset itineraries, and the flight segment information corresponding to each preset itinerary. The preset route includes multiple preset itineraries, and the flight segment is used to indicate the smallest itinerary unit in the preset itinerary, with one flight segment corresponding to one preset flight. Based on the preset route, the time series characteristics of the route relationship graph and the total number of passengers at the current time are determined. The route relationship graph is used to indicate the influence relationship between the preset route and adjacent routes, and the time series characteristics of the total number of passengers at the current time are used to indicate the trend of the total number of passengers at the current time changing over time. The route relationship graph and the time series characteristics of the total number of passengers at the current time are input into the trained passenger prediction model to determine the estimated passenger volume of the preset route. Based on the estimated passenger volume of the preset route and the market share of multiple preset itineraries, the estimated passenger volume of each preset itinerary is determined. Based on the estimated passenger volume of each preset itinerary and the flight segment information corresponding to each preset itinerary, the estimated passenger volume of the target flight is determined.

[0170] In one possible approach, the trained passenger prediction model includes a graph convolutional layer and a fully connected network layer. The processing module 402 is specifically used to extract features from the route relationship graph through the graph convolutional layer to determine the relationship feature matrix; and to process the relationship feature matrix and the time series features of the current time preset route passenger volume through the fully connected network layer to determine the total number of passengers on the preset route.

[0171] This disclosure also provides a computer-readable storage medium storing instructions that, when executed by a processor of an electronic device, enable the electronic device to perform the flight seat sales assessment method provided in the embodiments of this disclosure described above.

[0172] This disclosure also provides a computer program product containing instructions that, when run on an electronic device, cause the electronic device to execute the flight seat sales assessment method provided in the above-described embodiments of this disclosure.

[0173] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections having one or more wires; portable computer disks; hard disks; random access memory (RAM); read-only memory (ROM); erasable programmable read-only memory (EPROM); registers; hard disks; optical fibers; portable compact disc read-only memory (CD-ROM); optical storage devices; magnetic storage devices; or any suitable combination thereof; or any other form of computer-readable storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may reside in an application-specific integrated circuit (ASIC). In the embodiments of this application, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0174] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0175] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0176] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the classified units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0177] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0178] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, essentially, or the part that contributes to the prior art, or a complete or partial classification of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0179] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for evaluating flight seat sales, characterized in that, The method includes: The system obtains the current seat sales volume of the target flight, the estimated passenger volume at the current time, and the seat sales data of multiple reference flights. The reference flights are those in the historical flight data where the similarity between the external influencing factors and the external influencing factors of the target flight is greater than a similarity threshold. The external influencing factors are used to indicate the influencing factors that affect the seat sales of the target flight. Based on the estimated passenger volume of the target flight and the seat sales data of the multiple reference flights, the range of seat sales for the target flight at the current time is determined, wherein the range of seat sales is the range of the number of seats sold for the target flight at the current time. Based on the current seat sales volume of the target flight and the range of seat sales, determine whether the seat sales situation of the target flight is abnormal at the current time.

2. The method according to claim 1, characterized in that, The flight seat sales data is used to indicate the difference between the number of seats sold for the reference flight at the reference time and the number of seats sold for the reference flight at the departure time, wherein the time length between the reference time and the departure time of the reference flight is equal to the time length between the current time and the departure time of the target flight. The determination of the range of seat sales for the target flight at the current time, based on the estimated passenger volume of the target flight and the seat sales data of the multiple reference flights, includes: Based on the flight seat sales data of the multiple reference flights, statistical characteristic data of the reference flights are determined. The statistical characteristic data of the reference flights are used to reflect the distribution pattern of the difference in the number of seats sold. Based on the statistical characteristics of the reference flight and the estimated passenger volume of the target flight, the range of seat sales for the target flight at the current time is determined.

3. The method according to claim 2, characterized in that, The reference flight statistical characteristic data includes: the reference mean of seat sales and the reference standard deviation of seat sales, wherein the reference mean of seat sales is the average value of the seat sales data of the multiple reference flights, and the reference standard deviation of seat sales is the standard deviation of the seat sales data of the multiple reference flights; Determining the range of seat sales for the target flight at the current time based on multiple flight seat sales data and the estimated passenger volume of the target flight includes: Based on the estimated passenger volume of the target flight, the reference average seat sales, the reference standard deviation of seat sales, and the range correction factor, the range of seat sales for the target flight at the current time is determined. The range correction factor is used to indicate the impact of the time length from the current time to the departure time of the target flight on the range of seat sales.

4. The method according to claim 3, characterized in that, The range of seat sales includes a preset upper limit and / or a preset lower limit; The preset upper limit value satisfies the following first preset formula: Q1 = Z - (m - ke); Wherein, Q1 represents the preset upper limit value, Z represents the estimated passenger volume, m represents the reference average value of seat sales, e represents the reference standard deviation value of seat sales, and k represents the correction coefficient. The preset lower limit value satisfies the following second preset formula: Q2 = Z - (m + ke); Where Q2 represents the preset lower limit value, Z represents the estimated passenger volume, m represents the reference average value for seat sales, e represents the reference standard deviation value for seat sales, and k represents the correction coefficient.

5. The method according to claim 3 or 4, characterized in that, If the difference between the departure time of the target flight and the current time is greater than a first preset threshold, the range correction coefficient is the first correction coefficient. If the difference between the departure time of the target flight and the current time is less than or equal to the first preset threshold and greater than or equal to the second preset threshold, the range correction coefficient is the second correction coefficient. If the difference between the departure time of the target flight and the current time is less than the second preset threshold, the range correction coefficient is the third correction coefficient. The first correction factor is greater than the second correction factor, and the second correction factor is greater than the third correction factor.

6. The method according to claim 4, characterized in that, Based on the current seat sales volume of the target flight and the range of seat sales, determine whether the seat sales situation of the target flight is abnormal at the current time; If the current seat sales volume is less than or equal to the preset upper limit and greater than or equal to the preset lower limit, the seat sales situation of the target flight is considered normal. If the current seat sales volume is greater than the preset upper limit or less than the preset lower limit, the seat sales situation of the target flight is considered abnormal.

7. The method according to claim 1, characterized in that, Obtaining the estimated passenger volume corresponding to the current time includes: Obtain the preset route of the target flight, the market share of multiple preset itineraries, and the airsegment information corresponding to each preset itinerary. The preset route includes multiple preset itineraries, and the airsegment is used to indicate the smallest itinerary unit in the preset itinerary. One airsegment corresponds to one preset flight. Based on the preset routes, a route relationship diagram and the time series characteristics of the total number of passengers at the current time are determined. The route relationship diagram is used to indicate the influence relationship between the preset routes and adjacent routes, and the time series characteristics of the total number of passengers at the current time are used to indicate the trend of the total number of passengers at the current time changing over time. The route relationship diagram and the time series features of the total number of passengers at the current time are input into the trained passenger prediction model to determine the estimated passenger volume of the preset route. Based on the estimated passenger volume of the preset routes and the market share of the multiple preset itineraries, the estimated passenger volume of each preset itinerary is determined. The estimated passenger volume of the target flight is determined based on the estimated passenger volume of each preset itinerary and the corresponding flight section information.

8. The method according to claim 7, characterized in that, The trained passenger prediction model includes graph convolutional layers and fully connected network layers. The step of inputting the route relationship graph and the time-series features of the total number of passengers at the current time into the trained passenger prediction model to determine the total number of passengers on the preset route includes: The graph convolutional layer is used to extract features from the route relationship graph to determine the relationship feature matrix; The estimated passenger volume for the preset route is determined by processing the relation feature matrix and the time series features of the total number of passengers at the current time through the fully connected network layer.

9. A device for evaluating the sales of flight seats, characterized in that, The device includes: The acquisition module is used to acquire the current seat sales volume of the target flight, the estimated passenger volume at the current time, external influencing factors, and the seat sales data of multiple reference flights. The external influencing factors are used to indicate the influencing factors that affect the seat sales of the target flight. The reference flights are flights in the history of flights whose external influencing factors have a similarity greater than a similarity threshold with the external influencing factors of the target flight. The processing module is used to determine the range of seat sales for the target flight at the current time based on the estimated passenger volume of the target flight, the external influencing factors, and the seat sales data of the multiple reference flights. The range of seat sales is the range of the number of seats sold for the target flight at the current time. The processing module is also used to determine whether the seat sales situation of the target flight is abnormal at the current time, based on the current seat sales volume of the target flight and the range of seat sales situation.

10. An electronic device, characterized in that, It includes a memory and a processor; the memory and the processor are coupled; the memory is used to store instructions executable by the processor; when the processor executes the instructions, it performs the method as described in any one of claims 1-8.

Images