KR20240017482A - Method and apparatus for recommending seat by making seat pattern automatically based on artificial intelligence - Google Patents

Abstract

The present invention relates to a group seat recommendation method and device for recommending group seats corresponding to the representative's NR (name record) number of people based on a specific seat selected by the representative.
The seat recommendation method and device according to an embodiment of the present invention can recommend the optimal group seat corresponding to the number of representatives' NR (name record) based on the specific seat selected.
According to one embodiment of the present invention, group seats can be recommended by completing the optimal pattern of group seats corresponding to the number of people in the representative's NR (name record), starting with a specific seat selected by the representative.

Description

Seat recommendation method and device for automatically completing seat patterns based on artificial intelligence {METHOD AND APPARATUS FOR RECOMMENDING SEAT BY MAKING SEAT PATTERN AUTOMATICALLY BASED ON ARTIFICIAL INTELLIGENCE}

The present invention relates to a seat recommendation method and device, and more specifically, to an artificial intelligence-based seat pattern automatic completion method that recommends group seats corresponding to the representative's NR (name record) number of people based on the specific seat selected by the representative. It relates to a seat recommendation method and device.

Currently, many people can purchase tickets to the real world, such as airplanes with reserved seats, theaters, performance halls, sports stadiums, and trains, and to virtual worlds such as the Metaverse, through the Internet. In this case, seat assignment must be made simultaneously with or after ticket purchase. And the price of seat reservation may vary depending on the location.

For example, the seat classes of current airlines are broadly divided into four classes: First, Business/Prestige, Premium Economy, and Economy. Airlines operate an aircraft seat allocation system with a mixture of two or three classes or a single economy class (especially in the case of low-cost airlines) depending on the route and operating aircraft.

Recently, with the emergence of low-cost airline business models in the market, users have started to travel by air at low costs, but they have realized that it is difficult to express the inconvenience of flying for several hours, so now even if you pay for advance seat reservation, you can feel more comfortable. There is a demand for seats.

After an airline ticket is issued, people can purchase or designate their desired airplane seat via the Internet. The current seat reservation system simply distinguishes between unselected seats and selected seats and shows them to the user, and allows the user to purchase or designate the currently unselected seat. In the seat reservation system described above, if there is a passenger, the user must purchase or designate a seat, taking into account each passenger's seat, which is inconvenient. In particular, depending on the seat assignment situation, there are many cases where passengers cannot sit close to each other and there is a possibility that they will sit apart, so they have to endure inconvenience.

Therefore, research is needed on technology that automatically recommends preferred seats by considering the user's gender, age, number of passengers, etc.

Domestic Public Patent No. 10-2003-0039950 (May 22, 2003), seat reservation system for performances, dining, lodging, transportation and other sectors

The purpose of the present invention is to provide a seat recommendation method and device using an artificial intelligence-based seat pattern automatic completion method that recommends the optimal group seat corresponding to the representative's NR (name record) number of people based on the specific seat selected by the representative. there is.

The purpose of the present invention is to provide seat recommendation using an artificial intelligence-based seat pattern automatic completion method that recommends group seats by completing the optimal pattern of group seats corresponding to the number of people in the representative's NR (name record), starting with a specific seat selected by the representative. To provide a method and device.

In order to achieve the above object, according to an embodiment of the present invention, a seat recommendation method of a seat recommendation device connected to a computer network includes the steps of displaying a seat map showing currently occupied seats; Receiving user input for selecting a specific seat from a representative; Deriving all currently recommendable group seats based on the specific seat selected by the representative and the representative's NR (Name record) number of people; calculating a recommendation value corresponding to the group type for each of the derived group sheets; and recommending a predetermined number of group seats to a representative based on the calculated results. An artificial intelligence-based seat pattern auto-completion method for seat recommendation is disclosed.

The artificial intelligence-based seat pattern auto-complete seat recommendation method and device according to an embodiment of the present invention can recommend the optimal group seat corresponding to the representative's NR (name record) number of people based on the specific seat selected. .

According to one embodiment of the present invention, group seats can be recommended by completing the optimal pattern of group seats corresponding to the number of people in the representative's NR (name record), starting with a specific seat selected by the representative.

1 is a block showing a seat recommendation system related to an embodiment of the present invention.
FIG. 2 is a block diagram of the seat recommendation device shown in FIG. 1.
Figure 3 is a flowchart showing a seat recommendation method related to an embodiment of the present invention.
Figures 4 and 5 are diagrams for explaining group passenger types related to an embodiment of the present invention.
Figure 6 is a diagram illustrating a method for determining a recommendation ranking by calculating a recommendation value related to an embodiment of the present invention.
FIG. 7 is a diagram illustrating an example in which group seats are recommended based on a representative's selection of a specific seat related to an embodiment of the present invention.

Hereinafter, a seat recommendation method and device using an artificial intelligence-based seat pattern automatic completion method related to an embodiment of the present invention will be described with reference to the drawings.

As used herein, singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “consists of” or “comprises” should not be construed as necessarily including all of the various components or steps described in the specification, and some of the components or steps may be included in the specification. It may not be included, or it should be interpreted as including additional components or steps.

The seat recommendation method and device related to an embodiment of the present invention can be applied to seat recommendation in performance halls such as theaters, airplanes, sports halls, trains, etc. where seat designation is required.

In addition, the seat recommendation method and device related to an embodiment of the present invention can be applied not only to seat recommendation in the real world, but also to seat recommendation in a virtual world such as a metaverse where each seat has a different location and seat designation must be performed.

In this specification, a representative is a representative of a group and may include a representative reservation holder of tickets to a performance hall such as a theater, an airplane, a sports stadium, a train, etc., and a representative participant in a virtual world.

In this specification, the representative's NR (Name record) information may include the representative and companion information in the case of ticket reservation or participation in a specific game. For example, NR (Name record) information may include an airline ticket PNR (Passenger name record).

In this specification, a companion refers to a member who forms a group with the representative when recommending a group seat, and may include fellow passengers, accompanying visitors, accompanying participants, etc.

In this specification, group type may mean a group type determined based on NR (Name record) information. The group type may include group passenger type, group participant type, group visitor type, etc.

Hereinafter, in the embodiment, airplane seat recommendation is explained as an example of seat recommendation of the present invention, but the present invention is not limited thereto.

1 is a block showing a seat recommendation system related to an embodiment of the present invention.

As shown, the airplane seat recommendation system may include a user terminal 100, a seat recommendation device 200, a travel agency server 300, and an airline server 350. Here, the user can include a representative reservation holder in the Passenger name record (PNR).

The user terminal 100, seat recommendation center 200, travel agency server 300, and airline server 350 may be connected to a computer network.

The user may access the airplane seat pre-purchase or designation page directly through the user terminal 100, or may access the airplane seat purchase or designation page through the travel agency server 300 or the airline server 350.

Additionally, when an airline ticket is purchased through a travel agency or airline, the seat recommendation device 200 may receive airline ticket reservation information from the travel agency server 300 or the airline server 350. An example of airline ticket reservation information is the Passenger name record (PNR). Passenger name record (PNR) information is information about airline ticket reservations and may include passenger name, passenger name, passenger passport information, departure information, return information, etc.

Below, PNR (Passenger name record) will be explained as an example of airline ticket reservation information.

The seat recommendation device 200 may determine a group passenger type based on PNR information and derive all currently available group seats corresponding to the determined passenger type. In addition, the seat recommendation device 200 may determine a ranking by calculating a recommendation value for each group seat that can be derived. And according to the ranking results, a certain number of group seats can be recommended to the representative reservation holder.

FIG. 2 is a block diagram of the seat recommendation device shown in FIG. 1.

As shown, the seat recommendation device 200 includes a receiver 210, a passenger type determination unit 215, a group pattern database (DB) 220, a recommendable seat derivation unit 225, and an air seat database (DB). It may include (230), a recommendation value calculation unit 235, a group seat recommendation unit 240, a seat designation unit 245, and a control unit 250.

The receiving unit 210 may receive the PNR of the user (eg, representative reservation holder) from the travel agency server 300 or the airline server 350. Through the PNR, the passenger's name, co-passenger's name, passenger's passport information, departure information, return information, etc. can be confirmed. Additionally, the receiving unit 210 can receive purchase statistics for each seat.

Meanwhile, the PNR information may be received, but the seat recommendation device 200 may also generate it on its own.

The receiving unit 210 may receive group sheet information selected by the representative reservationist.

The receiving unit 210 may receive a user input for selecting a specific seat for the representative.

The passenger type determination unit 215 may determine the group passenger type based on the PNR information. The passenger type determination unit 215 is an example of a group type determination unit that determines the group type.

Group sheet pattern information is stored in the group pattern database (DB) 220. The group seat pattern information may include shape information of group seats. The shape of the group seats may be square, L-shaped, I-shaped, etc.

And seat information may be stored in the air seat database (DB) 230. The seat information may include seat map information for all seats on the airplane and information on currently occupied seats.

The recommendable sheet derivation unit 225 may derive all currently recommendable group sheets using the group sheet pattern information and the sheet information. When the representative selects a specific seat that is not occupied, the recommendable seat derivation unit 225 may derive all currently recommendable group seats to include the specific seat.

The recommendation value calculation unit 235 may calculate a recommendation value for each of all the derived group sheets. The recommendation value can be calculated using the position value and pattern value of each seat included in the group seat. We will explain later how to calculate the recommendation value.

The group seat recommendation unit 240 may recommend a predetermined number of group seats to the representative to include a specific seat selected by the representative based on the calculated recommendation value.

When the representative selects a recommended group seat, the seat designation unit 245 may designate a group seat for the corresponding PNR based on the selected group seat information.

The control unit 250 includes a reception unit 210, a passenger type determination unit 215, a group pattern database (DB) 220, a recommendable seat derivation unit 225, an air seat database (DB) 230, and a recommendation value. The calculation unit 235, group seat recommendation unit 240, and seat assignment unit 245 can be generally controlled.

Figure 3 is a flowchart showing a seat recommendation method related to an embodiment of the present invention.

The receiving unit 210 may receive PNR information of the representative reservationist from the travel agency server 300 or the airline server 350 (S310).

The passenger type determination unit 215 may determine the group passenger type based on the PNR information (S320).

4 to 5 are diagrams for explaining passenger types related to an embodiment of the present invention.

As shown in FIG. 4, passenger types can be classified according to a plurality of criteria. For example, the number of people, gender, age, passenger composition, etc. may be multiple standards.

For example, when classifying passenger types according to the number of people, they are divided into types from 1 person to 9 people, and when classifying passenger types by gender, there are only men, only women, a mixture of men and women, and the same number of men and women passengers. In some cases, it can be divided into cases where men and women are mixed and the number of male and female passengers is different.

The above multiple criteria for classifying passenger types may be used alone, or several may be used in combination.

Additionally, as shown in FIG. 5, passenger types may be divided into care type, couple type, community type, etc. In the example shown, the passenger types are divided into three, but they may be divided into more passenger types or fewer passenger types.

The passenger type may correspond to one of a plurality of passenger types if predefined conditions are satisfied. For example, if the PNR of the representative carer includes a physically vulnerable person, it may be classified as a care type. For example, passengers such as infants (under 2 years of age), children (under 5 years of age), and senior citizens (over 65 years of age) may be classified as physically vulnerable.

For example, if the PNR has an even number of adults (2, 4, 6, 8), an equal number of men and women, and an age difference of 10 years or less between men and women, it can be classified as a couple type. .

Meanwhile, the receiver 210 may receive a user input for selecting a specific seat of the representative from a seat map displaying currently occupied seats (S330).

The recommendable sheet derivation unit 225 can derive all currently recommendable group sheets using group sheet pattern information and sheet information. In this case, the recommendable seat derivation unit 225 may derive all currently recommendable group seats so that the specific seat selected by the representative is included (S340). The group seat pattern information may include shape information of group seats. The shape of the group seats may be square, L-shaped, I-shaped, etc. Additionally, the seat information may include seat map information for all seats on the airplane and information on currently occupied seats.

The recommendation value calculation unit 235 may calculate a recommendation value for each of all the derived group sheets (S350). The recommendation value calculation unit 235 may calculate the recommendation value using the position value and pattern value of each seat included in each derived group seat.

The variables that determine the location value include a variable (D) representing the value according to the distance to the exit in the vertical direction of the seat, a variable (WMA) representing the value depending on whether the horizontal position of the seat is a window, an aisle, or a center seat, There is a variable (LMR) that represents the value depending on whether the horizontal position of the seat belongs to the right block, the middle block, or the left block.

Variables that determine the pattern value include a variable (S) representing the value according to the area per seat and a variable (US) representing the value according to the degree of seat separation from other passengers.

Additionally, when calculating the recommendation value, a variable (P) representing the value according to the price per seat can be used.

The recommendation value can be calculated using Equation 1 below. Equation 1 below is an example of using variables when calculating a recommendation value, but the present invention is not limited to this.

Here, V is the recommended value, D is a variable representing the value according to the distance to the exit in the vertical direction of the seat, WMA is a variable representing the value according to whether the horizontal position of the seat is a window, aisle, or center seat, and LMR is a variable that represents the value according to whether the horizontal position of the seat belongs to the right block, the middle block, or the left block, S is a variable that represents the value according to the area per seat, and B is the proximity to fellow passengers. US is a variable that represents the value according to the degree of seat separation from other passengers, and P is a variable that represents the value according to the price per seat.

The initial values of the variables may be determined based on actual seat reservation purchase data.

Figure 6 is a diagram illustrating a method for determining a recommendation ranking by calculating a recommendation value related to an embodiment of the present invention.

Pattern value can be calculated from the product of B and US. In this case, the initial B value can be assigned as horizontally next to the same row = 1, same row between corridors = 0.5, diagonal digits between the front and back rows = 0.1, and front and back digits = 0.

Additionally, US can be calculated by Equation 2 below.

Here, a block refers to a group of consecutive seats adjacent to each other. For example, 2A, 2B, and 2C may be composed of one block, and 2D, 2E, and 2F may be composed of another block.

And the position value can be calculated as the product of D, WMA, LMR, and S. The initial D value can be assigned as the 1st row with the closest distance to the exit = 1, the next 2nd row = 0.75, etc.

In addition, the initial S value can be assigned as 1 for general seats, and 2.32 for wide seats such as 2D seats.

And the initial value of WMA may be set differently depending on whether it is a window seat, a middle seat, or an aisle seat.

Additionally, the initial value of LMR may be assigned differently depending on whether the horizontal position of the seat belongs to the right block, the middle block, or the left block.

And the P value may be assigned differently depending on the price per seat.

And by substituting the assigned variable values into Equation 1 above, the total recommendation value can be calculated. And the recommendation ranking of the group sheet may be determined according to the calculated recommendation value.

The group sheet recommendation unit 240 may recommend a predetermined number of group sheets to the representative reservation holder based on the calculated recommendation value (S360).

For example, the group seat recommendation unit 240 completes the group seat pattern corresponding to the number of NR members of the representative so that the selected specific seat is included, and displays the completed group seat pattern on the seat map. You can make a recommendation.

The group seat recommendation unit 240 completes the group seat pattern corresponding to the number of NR members of the representative by sequentially adding unoccupied seats starting from the selected specific seat, and applies the completed group seat pattern to the It can also be displayed on a sheet map.

FIG. 7 is a diagram illustrating an example in which group seats are recommended based on a representative's selection of a specific seat related to an embodiment of the present invention.

As shown, the representative can select a specific seat of his/her choice among the currently unoccupied seats (S1). Representatives can select a specific seat by touching it (S2). When the specific seat is selected, the group seat recommendation unit 240 sequentially adds unoccupied seats starting from the selected specific seat (S3, S4, S5), thereby selecting a group corresponding to the number of NR members of the representative. The seat pattern may be completed (S6), and the completed group seat pattern 710 may be displayed on the seat map.

The artificial intelligence-based seat pattern auto-complete seat recommendation method and device according to an embodiment of the present invention can recommend the optimal group seat corresponding to the representative's NR (name record) number of people based on the specific seat selected. .

According to one embodiment of the present invention, group seats can be recommended by completing the optimal pattern of group seats corresponding to the number of people in the representative's NR (name record), starting with a specific seat selected by the representative.

The artificial intelligence-based seat pattern automatic completion type seat recommendation method and device described above is not limited to the configuration and method of the above-described embodiments, and the embodiments are each so that various modifications can be made. All or part of the embodiments may be selectively combined.

100: user terminal
200: Seat recommendation device
210: receiving unit
215: Passenger type determination unit
220: Group pattern database
225: Recommended sheet derivation unit
230: Air seat database
235: Recommendation value calculation unit
240: Group sheet recommendation section
245: Seat assignment section
250: control unit
300: Travel agency server
350: Airline server

Claims (3)

In the seat recommendation method of a seat recommendation device connected to a computer network,
Displaying a seat map showing currently occupied seats;
Receiving user input for selecting a specific seat from a representative;
Deriving all currently recommendable group seats based on the specific seat selected by the representative and the representative's NR (Name record) number of people;
calculating a recommendation value corresponding to the group type for each of the derived group sheets; and
An artificial intelligence-based seat pattern auto-completion method for seat recommendation, comprising the step of recommending a predetermined number of group seats to a representative based on the calculated results. The method of claim 1, wherein the group sheet recommendation step is
Completing a group seat pattern corresponding to the number of NR members of the representative to include the selected specific seat;
An artificial intelligence-based seat pattern auto-completion method for seat recommendation, comprising the step of displaying the completed group seat pattern on the seat map. The method of claim 1, wherein the group sheet recommendation step is
Completing a group seat pattern corresponding to the number of NRs of the representative by sequentially adding unoccupied seats starting from the selected specific seat;
An artificial intelligence-based seat pattern auto-completion method for seat recommendation, comprising the step of displaying the completed group seat pattern on the seat map.