Resale detection system, trained model, resale detection method and program

Abstract

To estimate whether a user books an accommodation for a resale purpose.SOLUTION: In a resale detection system 10, a sale estimation device 100 includes an output unit that outputs information showing a degree of probability in which a currently-reserved accommodation facility is reserved for the resale purpose by inputting, current reservation date determination information, which includes the determination result of whether the current reservation date of the currently-reserved accommodation facility that is an accommodation facility reserved by a prescribed subscriber, which is identified on the basis of current reservation information, meets a prescribed condition or not, and current event availability information, which indicates whether an event is held or not in a current target event facility that is an event facility having a prescribed relationship with the currently-reserved accommodation facility, which is identified on the basis of event information, on any day included in the term of the prescribed subscriber's stay in the currently-reserved accommodation facility, which is identified on the basis of the event information, into an evaluation model trained with past reservation date determination information, past event availability information and resale information as teaching data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a resale detection system, a trained model, a resale detection method, and a program. [Background technology]

[0002] An information processing device has been disclosed that estimates a risk of a reservation being canceled based on reservation information relating to a facility reservation made by a person making a reservation (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-200750 A Summary of the Invention [Problem to be solved by the invention]

[0004] The information processing device described in Patent Document 1 can estimate the risk of a reservation being canceled, and therefore can take appropriate measures for the reservation (such as invalidating the reservation) before the scheduled date, for example.

[0005] However, in recent years, the travel industry has seen an increase in damage caused by users known as "UGs (Undesirable Guests)," who not only cancel reservations but also cause various damage to accommodations. One type of damage suffered by accommodations is reservations made with the intention of reselling. Therefore, in order to prevent reservations made with the intention of reselling, it is desirable to realize a new mechanism for estimating whether or not a reservation made by a user is made with the intention of reselling.

[0006] Therefore, an object of the present invention is to make it possible to estimate whether or not a reservation for an accommodation has been made by a user with the intention of reselling. [Means for solving the problem]

[0007] A resale detection system according to one aspect of the present invention includes an acquisition unit that acquires current reservation information, which is information regarding current reservations at an accommodation facility by a specified reservationr; past reservation date determination information, which is identified based on past reservation information, which is information regarding past reservations at the accommodation facility, and which includes a past reservation date of a past reserved accommodation facility that is an accommodation facility reserved by a past reservationr who is a past reservationr, and which includes a determination result indicating whether or not the past reservation date satisfies a specified condition; past event holding status information, which is identified based on event information including information regarding the event facility where the event will be held, which indicates whether an event was held on any of the days included in the stay period of the past reserved accommodation facility at the past reserved accommodation facility, which is an event facility having a specified relationship with the past reserved accommodation facility; and past event holding status information, which is identified based on event information including information regarding the event facility where the event will be held, indicating whether or not an event was held on any of the days included in the stay period of the past reserved accommodation facility at the past reserved accommodation facility. and an output unit that outputs information indicating the degree of possibility that the currently reserved accommodation has been reserved for the purpose of resale, which is predicted by inputting, into an evaluation model trained using the above-mentioned training data, resale information indicating whether or not the currently reserved accommodation has been reserved for the purpose of resale, current reservation date determination information including a determination result indicating whether or not the current reservation date of the currently reserved accommodation, which is an accommodation reserved by the specified reservation user, identified based on the current reservation information, satisfies the specified condition, and current event holding status information indicating whether or not an event will be held at a current target event facility, which is an event facility having the specified relationship with the currently reserved accommodation, identified based on the event information, on any date included in the stay period of the specified reservation user at the currently reserved accommodation, identified based on the event information.

[0008] A trained model according to one aspect of the present invention is a trained model for causing a computer to function to output information indicating the degree of likelihood that a specific reservation party who reserves an accommodation facility has reserved the accommodation facility for the purpose of resale, and includes past reservation date determination information, which is identified based on past reservation information, which is information regarding past reservations of the accommodation facility, and which includes a determination result indicating whether or not the past reservation date satisfies a specific condition; past event holding status information, which is identified based on event information, which includes information regarding the event facility where the event will be held, indicating whether or not an event was held on any of the days included in the stay period of the past reservation party at the past reservation facility; and past reservation date determination information, which is identified based on past reservation information, which is information regarding past reservations of the accommodation facility, including a determination result indicating whether or not the past reservation date satisfies a specific condition at a past target event facility, which is an event facility having a specific relationship with the past reservation facility, The computer is caused to function to output information indicating the degree of possibility that the specified reservation person has reserved the currently reserved accommodation facility for the purpose of resale by inputting: resale information indicating whether or not the reservation has been made for the purpose of resale; current reservation date determination information, which is identified based on current reservation information, which is information regarding the current reservation of the accommodation facility by the specified reservation person, and which includes a determination result indicating whether or not the current reservation date of the currently reserved accommodation facility, which is the accommodation facility reserved by the specified reservation person, satisfies the specified condition; and current event holding status information, which indicates whether or not an event will be held at the current target event facility, which is the event facility that has the specified relationship with the currently reserved accommodation facility, identified based on the event information, on any day included in the specified reservation person's stay period at the currently reserved accommodation facility, identified based on the event information.

[0009] A resale detection method according to one aspect of the present invention includes a computer acquiring current reservation information, which is information relating to a current reservation of an accommodation by a specified reservationr, and past reservation date determination information, which is identified based on past reservation information, which is information relating to past reservations of the accommodation, and which includes a determination result indicating whether or not the past reservation date satisfies a specified condition, past event holding status information, which is identified based on event information including information regarding an event facility where an event will be held, indicating whether an event was held on any of the dates included in the stay period of the past reservationr at the past reservation accommodation, and past reservation date determination information, which is identified based on past reservation information, which is information relating to past reservations of the accommodation, and includes a determination result indicating whether or not the past reservation date satisfies a specified condition, at a past target event facility, which is an event facility having a specified relationship with the past reservation accommodation, The method outputs information indicating the degree of possibility that the currently reserved accommodation has been reserved for the purpose of resale, which is predicted by inputting, into an evaluation model trained using resale information indicating whether or not a facility has been reserved for the purpose of resale, current reservation date determination information including a determination result indicating whether or not the current reservation date of the currently reserved accommodation facility, which is an accommodation facility reserved by the specified reservation user and identified based on the current reservation information, satisfies the specified condition, and current event holding information indicating whether or not an event will be held at a current target event facility, which is an event facility having the specified relationship with the currently reserved accommodation facility and identified based on the event information, on any date included in the stay period of the specified reservation user at the currently reserved accommodation facility, which is identified based on the event information, and

[0010] A program according to one aspect of the present invention includes a computer that acquires current reservation information, which is information regarding a current reservation of an accommodation by a specified reservationr, and past reservation date determination information, which is specified based on past reservation information, which is information regarding past reservations of the accommodation, and which includes a determination result indicating whether or not the past reservation date satisfies a specified condition, past event holding status information, which is specified based on event information including information regarding an event facility where an event will be held, indicating whether or not an event was held on any of the days included in the stay period of the past reservationr at the past reservation accommodation, and past event holding status information, which is specified based on event information including information regarding an event facility where an event will be held, indicating whether or not an event was held on any of the days included in the stay period of the past reservationr at the past reservation accommodation and outputting information indicating the degree of possibility that the currently reserved accommodation has been reserved for the purpose of resale, which is predicted by inputting, into an evaluation model trained using resale information indicating whether or not the facility has been reserved for the purpose of resale, current reservation date determination information including a determination result indicating whether or not the current reservation date of the currently reserved accommodation facility, which is an accommodation facility reserved by the specified reservation user and is identified based on the current reservation information, satisfies the specified condition, and current event holding information indicating whether or not an event will be held at a current target event facility, which is an event facility having the specified relationship with the currently reserved accommodation facility and which is identified based on the event information, on any date included in the stay period of the specified reservation user at the currently reserved accommodation facility, which is identified based on the event information, Effect of the Invention

[0011] According to the present invention, it is possible to provide a system that makes it possible to estimate whether a reservation for an accommodation facility has been made by a person intending to resell the reservation. [Brief description of the drawings]

[0012] [Figure 1] FIG. 2 is a configuration diagram showing an example of the functional configuration of various devices that make up the resale detection system. [Diagram 2]FIG. 11 is a diagram showing a reservation database D111. [Diagram 3] FIG. 13 is a diagram showing an event database D112. [Figure 4] FIG. 11 is a diagram showing a processing result database D113. [Diagram 5] FIG. 11 is a diagram showing a resale behavior database D114. [Figure 6] 4 is a flowchart showing the processing steps of the resale detection system 10. [Figure 7] FIG. 2 illustrates an example of a hardware configuration of a computer. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] ===Resale Detection System 10=== <<Overview>> The configuration of a resale detection system 10 according to this embodiment will be described with reference to FIG. 1. FIG. 1 is a diagram showing an example of the configuration of the resale detection system 10. As shown in FIG. 1, the resale detection system 10 includes, for example, a resale estimation device 100 and a facility reservation management device 200. Furthermore, the resale detection system 10 may include an evaluation model. Although not shown, the resale detection system 10 may be configured with a plurality of facility reservation management devices 200. The devices of the resale detection system 10 are connected to each other via a communication network N so as to be able to communicate with each other. The communication network N may be either a wired network or a wireless network.

[0014] The resale detection system 10 is a system that estimates whether or not a person who makes a reservation has made a reservation for the accommodation facility for the purpose of resale, for example, by using an evaluation model that has been trained using accumulated past reservation information (hereinafter referred to as "past reservation information") as training data. Specifically, the resale detection system 10 acquires information regarding the current reservation of the accommodation facility (hereinafter referred to as "current reservation information") from, for example, at least one facility reservation management device 200. The resale detection system 10 inputs the current reservation information into an evaluation model, and outputs information indicating a score (e.g., probability) of the possibility that the accommodation reservation is a reservation for the purpose of resale (hereinafter referred to as "resale evaluation information"). The resale detection system 10 may output, for example, report information including the score indicated by the resale evaluation information and information regarding the person who made the reservation (e.g., the reservation user ID and the name of the person who made the reservation) to the facility reservation management device 200.

[0015] The past reservation information is, for example, information relating to a past reservation made by a person who has made a reservation at an accommodation facility. The past reservation information is, for example, information previously transmitted from a user terminal (not shown) to the facility reservation management device 200 based on a user's operation input to the user terminal. The past reservation information is, for example, information indicating a reservation number, reservation date and time, accommodation facility, name and number of people who made the reservation, etc. Specifically, the current reservation information includes information stored in each item of a reservation database D111, which will be described later.

[0016] The current reservation information is, for example, information regarding a current reservation of a person who has made a reservation to use an accommodation facility. The current reservation information is, for example, information transmitted from a user terminal (not shown) to the facility reservation management device 200 based on a user's operation input to the user terminal. The current reservation information is, for example, information having the same items as the past reservation information.

[0017] The purpose of resale is, for example, the purpose of transferring a reservation for accommodation to another person for a fee. Such resale is carried out for the purpose of transferring the reservation for accommodation to participants of an event held near the accommodation for a price higher than the normal accommodation fee. An event is, for example, an event held at a facility such as an arena or an outdoor concert venue that can accommodate a large number of people (hereinafter referred to as an "event facility").

[0018] The evaluation model (trained model) is, for example, a regression model or a classification model. Specifically, the evaluation model may be, for example, a method using linear regression, logistic regression, random forest, boosting, support vector machine, or neural network. The evaluation model will be described later.

[0019] The resale estimation device 100 is, for example, a server device that calculates resale evaluation information for current reservation information acquired from at least one facility reservation management device 200. The resale estimation device 100 provides the facility reservation management device 200 with report information generated based on the resale evaluation information.

[0020] The facility reservation management device 200 is, for example, a server device that accepts reservations for accommodation facilities from users. The facility reservation management device 200 is, for example, a device operated by a business operator that operates a travel agency that accepts reservations for accommodation facilities. The facility reservation management device 200 acquires current reservation information from the person making the reservation and transmits the current reservation information to the resale estimation device 100. Then, the facility reservation management device 200 acquires resale evaluation information for each piece of current reservation information from the resale estimation device 100.

[0021] As a result, based on the report information received by the facility reservation management device 200, operators of travel agencies can take measures to prevent resale by those who may have made facility reservations with the intent to resell (for example, making reservations with the relevant person or canceling the reservation by that person).

[0022] The resale estimation device 100 and the facility reservation management device 200 may be, for example and without limitation, a server computer, a personal computer (for example and without limitation, a desktop, laptop, tablet, etc.), a media computing platform (e.g., cable, satellite set-top box, digital video recorder), a handheld computing device (e.g., a Personal Digital Assistant (PDA), email client, etc.), or other types of computers or communications platforms.

[0023] ===Resale Estimation Device 100=== The functional configuration of the resale estimation device 100 will be described with reference to Fig. 1. Fig. 1 is a configuration diagram showing an example of the functional configuration of various devices that make up the resale detection system 10. The resale estimation device 100 acquires current reservation information of the reservation holder's accommodation facility from the facility reservation management device 200, and outputs report information based on the current reservation information.

[0024] As shown in FIG. 1, the resale estimation device 100 includes functional units, for example, a memory unit 110, an acquisition unit 120, a specification determination unit 130, a learning unit 140, an estimation unit 150, and an output unit 160.

[0025] The storage unit 110 may include, for example, a reservation database D111, an event database D112, a processing result database D113, and a resale behavior database D114.

[0026] The reservation database D111, the event database D112, the processing result database D113, and the resale behavior database D114 are, for example, databases for efficiently learning the evaluation model of the learning unit 140 described below. These databases are, for example, databases in which each piece of data is extracted, edited, and compiled by a user. Furthermore, these databases do not have to be stored in the storage unit 110, for example, and may be databases of other businesses. In this case, the resale estimation device 100 may obtain data necessary for learning the evaluation model of the learning unit 140 from the database of the other business.

[0027] The reservation database D111 will be described with reference to Fig. 2. Fig. 2 is a diagram showing the reservation database D111. The reservation database D111 is a database that compiles current reservation information and past reservation information related to a plurality of accommodation facilities and a plurality of reservation persons. As shown in Fig. 2, the reservation person database D111 includes items such as [accommodation facility ID], [reservation ID], [reservation person ID], [reservation person name], [stay period], [reservation acceptance date and time], [child use], and [planned check-in time], for example.

[0028] [Accommodation ID] stores, for example, an identification code that can identify the accommodation facility. [Reservation ID] stores, for example, an identification code that can identify reservation information. [Reserving person ID] stores, for example, an identification code that can identify the reserving person. [Reserving person name] stores, for example, the name of the reserving person. [Staying period] stores, for example, the staying period at the time of booking. The staying period refers to, for example, the period from the check-in date to the check-out date. In the following, when "any day included in the staying period" is mentioned, for example, if the check-in date is "January 1, 2023" and the check-out date is "January 3, 2023", it refers to any day among "January 1, 2023, January 2, 2023, and January 3, 2023". [Reservation date and time] stores, for example, the date and time when the facility reservation management device 200 accepts the reservation from the reserving person (hereinafter referred to as "reservation date"). [Child use] stores, for example, information indicating whether or not a child will stay with the reserving person. [Planned check-in time] stores, for example, information indicating the planned check-in time of the person planning to stay at the accommodation facility.

[0029] The event database D112 will be described with reference to Fig. 3. Fig. 3 is a diagram showing the event database D112. The event database D112 is a database that includes information on event facilities having a predetermined relationship with accommodation facilities and event hosting, for example.

[0030] The predetermined relationship is, for example, a relationship in which the accommodation facility and the event facility are within a predetermined distance, a relationship in which they are in the same city, ward, town, or village, a relationship in which they are previously associated by the operator of the resale estimation device 100, etc. Hereinafter, the same applies to the "predetermined relationship."

[0031] As shown in FIG. 3, the event database D112 may include, for example, items such as [accommodation facility ID], [event facility ID], [event announcement date], and [event holding date].

[0032] [Accommodation ID] stores, for example, an identification code that can identify an accommodation facility. [Event Facility ID] stores, for example, an identification code that can identify an event facility. [Event ID] stores, for example, an identification code that can identify an event held at each event facility. [Event Announcement Date] stores, for example, the date the event was announced to the public (for example, the date it was announced on the website of the event organizer). [Event Holding Date] stores, for example, the date the event is held at the event facility.

[0033] The processing result database D113 will be described with reference to Fig. 4. Fig. 4 is a diagram showing the processing result database D113. The processing result database D113 is a database including the determination results of whether or not the reservation date satisfies a predetermined condition. As shown in Fig. 4, the determination result database D113 may include items such as [Reservation ID], [Whether or not an event is held], [First determination result], [Second determination result], and [Third determination result].

[0034] [Reservation ID] stores, for example, an identification code that can identify reservation information.

[0035] [Event held or not] stores information about whether an event is held or not. The event held or not information is information identified by the identification determination unit 130 described later, and indicates whether an event was held or not at an event facility (hereinafter referred to as the "target event facility") that has a specified relationship with the reserved accommodation facility on any of the days included in the accommodation period at the time of reservation by the reservationr at the accommodation facility reserved by the reservationr (hereinafter referred to as the "reserved accommodation facility"). For example, if an event was held at the target event facility, "1" is stored, and if not, "0" is stored.

[0036] The first judgment result is stored in [First judgment result]. The first judgment result is information indicating the result of judgment made by the specific judgment unit 130 described later, and is, for example, a judgment result indicating whether or not the reservation date made by the booker is within a predetermined number of days of the announcement date of the event to be held at the target event facility (hereinafter referred to as the "target event"). If the reservation date is within the predetermined number of days of the announcement date, "1" is stored, and if not, "0" is stored.

[0037] The second judgment result is stored in [Second judgment result]. The second judgment result is information indicating the result of judgment made by the specific judgment unit 130 described later, and is, for example, a judgment result indicating whether or not the reservation date made by the person who made the reservation was included between the announcement date, which announced that the target event would be held at the target event facility, and the opening date of the target event. If the reservation date is included between the announcement date and the opening date, "1" is stored, and if not, "0" is stored.

[0038] The third judgment result is stored in [Third judgment result]. The third judgment result is information indicating the result of judgment made by the specific judgment unit 130 described later, and is, for example, a judgment result indicating whether or not multiple reservations have been registered by the same reservation person on the day of the target event at an accommodation facility that has a predetermined relationship with the target event facility. For example, if multiple reservations have been registered by the same reservation person, "1" is stored, and if not, "0" is stored.

[0039] The resale behavior database D114 will be described with reference to Fig. 5. Fig. 5 is a diagram showing the resale behavior database D114. The resale behavior database D114 is a database that includes information indicating whether or not a past reservation was a reservation for an accommodation facility for the purpose of resale. As shown in Fig. 5, the resale behavior database D114 may include items such as [Reservation ID], [Reserver ID], and [Resale behavior flag].

[0040] [Reservation ID] stores, for example, an identification code capable of identifying the reservation information. [Reserved User ID] stores, for example, an identification code capable of identifying the person who made the reservation. [Resale Flag] stores a resale flag indicating whether the reservation indicated by the past reservation information was an accommodation reservation for the purpose of resale. For example, the resale flag is set by obtaining information regarding whether a reservation in the past reservation information was actually made for resale from a business operator that operates a travel agency that accepts reservations for accommodation facilities, and storing "1" for reservation information that was a reservation for the purpose of resale, and "0" for reservation information that was not a reservation for the purpose of resale.

[0041] The acquiring unit 120 acquires, for example, the current reservation information and the past reservation information from the facility reservation management device 200. The current reservation information and the past reservation information are stored in the storage unit 110.

[0042] The identification determination unit 130 references the reservation database D111 and the event database D112 to generate current and past event holding status information and reservation date determination information.

[0043] The event holding information is information indicating whether or not an event will be held at the target event facility on any of the days included in the current or past stay period of the person who made the reservation at the reserved accommodation facility. The process of generating the event holding information in the identification determination unit 130 will be described below.

[0044] In the process of generating event holding status information, first, the identification determination unit 130 refers to the reservation database D111 to identify the stay period of each reservation, and then refers to the event database D112 to identify the target event facility corresponding to each reserved accommodation facility of the reservation. The identification determination unit 130 refers to the event database D112 to determine whether an event will be held on any of the dates included in the stay period of the reservation at each of the identified target event facilities. This allows the identification determination unit 130 to generate current or past event holding status information.

[0045] In the above, the identification determination unit 130 identifies the target event facility by referring to the event database D112 in which the accommodation facility is associated with an event facility having a predetermined relationship, but this is not limited to the above. For example, the target event facility may be identified by referring to a database in which the accommodation facility is not associated with an event facility having a predetermined relationship. Specifically, the identification determination unit 130 may identify the target event facility by referring to a database in which the coordinates of the location of the accommodation facility are stored and a database in which the coordinates of the location of the event facility are stored, and identifying the coordinates of the event facility within a predetermined distance from the coordinates of the accommodation facility.

[0046] The reservation date determination information includes, for example, a first determination result, a second determination result, and a third determination result indicating whether the reservation date satisfies a predetermined condition. Hereinafter, a process of generating the first determination result, the second determination result, and the third determination result in the specific determination unit 130 will be described.

[0047] In the process of generating the first determination result, first, the identification determination unit 130 identifies the reservation date of the reservation by referring to the reservation database D111, and identifies the target event facility corresponding to the reserved accommodation facility of the reservation by referring to the event database D112. The identification determination unit 130 identifies the event announcement date of the target event at the identified target event facility. The identification determination unit 130 judges whether the reservation date of the reservation is included within a predetermined number of days of the event announcement date corresponding to the reservation date. The predetermined number of days may be, for example, a period of five days before the announcement date and five days after the announcement date, or five days after the announcement date, based on the announcement date. In this way, the identification determination unit 130 can generate the current or past first determination result.

[0048] In the process of generating the second determination result, first, the identification determination unit 130 identifies the reservation date of the reservation by referring to the reservation database D111, and identifies the target event facility corresponding to the reserved accommodation facility of the reservation by referring to the event database D112. The identification determination unit 130 identifies the event announcement date and the event holding date of the target event at the identified target event facility. The identification determination unit 130 judges whether the reservation date of the reservation is included between the event announcement date corresponding to the reservation date and the event holding date of the target event. Specifically, when the announcement date is "January 1" and the holding date is "April 1", if the reservation date is "January 10", the determination result is "included". This allows the identification determination unit 130 to generate the current or past second determination result.

[0049] In the process of generating the third determination result, first, the identification determination unit 130 identifies the reservation date of the reservation by referring to the reservation database D111, and identifies the target event facility corresponding to the reserved accommodation facility of the reservation by referring to the event database D112. Furthermore, the identification determination unit 130 identifies, for example, accommodation facilities having a predetermined relationship with the target event facility. The identification determination unit 130 identifies the date on which the target event will be held at the target event facility. The identification determination unit 130 determines whether or not one reservation person has made multiple reservations on the date on which the target event will be held at at least one of the identified accommodation facilities. Here, when there are multiple identified accommodation facilities, the identification determination unit 130 may determine whether or not there are multiple reservations by one reservation person for all of the multiple accommodation facilities, or may determine whether or not there are multiple reservations by one reservation person for one accommodation facility. In this way, the identification determination unit 130 can generate the current or past third determination result.

[0050] The learning unit 140 trains an evaluation model using, for example, the past reservation information stored in the storage unit 110 as teacher data.

[0051] The evaluation model is a trained model that predicts the possibility that a reservation made by a user for an accommodation facility is for the purpose of resale. The evaluation model is trained using, for example, past reservation date determination information (hereinafter referred to as "past reservation date determination information"), event holding status information (hereinafter referred to as "past event holding status information"), and resale information as training data.

[0052] In other words, when the trained evaluation model receives current reservation date determination information (hereinafter referred to as "current reservation date determination information") and event holding status information (hereinafter referred to as "current event holding status information") identified based on current reservation information, it outputs resale evaluation information that indicates the degree of possibility that the reserved accommodation facility has been booked for the purpose of resale.

[0053] Specifically, the evaluation model can be derived as a regression model with the “resale flag” in the resale database D114, which indicates whether or not a reservation for accommodation at an accommodation facility was made for the purpose of resale, as the objective variable (correct answer data), and with past reservation date determination information and past event holding status information extracted from the processing result database D113 as explanatory variables (features).

[0054] For example, the evaluation model is obtained as a regression model as shown in the following formula (1). y=W 1 X 1 +W 2 X 2 +···+W n X n (1)

[0055] In formula (1), "X" is an explanatory variable corresponding to information extracted from the processing result database D113. "y" is an objective variable corresponding to the "resale flag", which is "1" if the reservation is for the purpose of resale, and is "-1" or "0" if the reservation is not for the purpose of resale.

[0056] In addition, in formula (1), "W" is a coefficient of "X" and indicates a weighting value. 1" is "X 1 " is the weight value of "W 2 " is "X 2 " is the weight value of "W n " is "X n In this way, the formula (1) includes an explanatory variable "X" corresponding to the past reservation date determination information and the past event holding information extracted from the processing result database D113, and a variable including a weight value "W" (for example, "W 1 X 1 ") is used to create the

[0057] Here, the basis for the fact that using information contained in the processing result database D113 as training data for training an evaluation model makes it possible to accurately estimate whether a reservation is for the purpose of resale will be described.

[0058] Reservations for resale purposes tend to be made at accommodation facilities around the target event facility around the announcement date of the event. Therefore, the resale detection system 10 can accurately estimate whether the reservation is for resale purposes by using information (e.g., the first judgment result or the second judgment result) as training data on whether the reservation date for the accommodation facilities around the target event facility satisfies a predetermined condition in relation to the announcement date of the event at the target event facility.

[0059] In addition, reservations for resale purposes tend to be made by a single person making multiple reservations for accommodation facilities around the target event facility. Therefore, the resale detection system 10 can accurately estimate whether or not a reservation is for resale purposes by using information on whether or not a single person has made multiple reservations for accommodation facilities around the target event facility (e.g., the third judgment result) as training data.

[0060] In other words, the resale detection system 10 can capture notable trends in reservations for the purpose of resale, and solve the problem of not being able to properly estimate whether a reservation is for the purpose of resale by simply having an evaluation model learn general reservation information (e.g., name, age, accommodation, number of nights stayed, etc.).

[0061] The estimation unit 150 inputs the current reservation date judgment information and the current event holding information into the evaluation model, thereby acquiring the resale evaluation information from the evaluation model. The estimation unit 150 generates report information using the resale evaluation information (e.g., a score). The report information may be, for example, information indicating only a score indicating that the reservation is for the purpose of resale, or may be information including the score as well as information capable of identifying the reservation holder included in the current reservation information (e.g., a reservation holder ID and contact information, etc.). This allows a business operator operating a travel agency to take measures to prevent resale by a reservation holder who may have made a facility reservation for the purpose of resale, based on the report information received by the facility reservation management device 200.

[0062] The output unit 160 transmits the report information generated by the estimation unit 150 to the facility reservation management device 200.

[0063] <<Modifications>> In the above, it has been explained that the resale estimation device 100 uses past reservation date determination information and past event holding / non-holding information as training data for training the evaluation model, but the following information may also be included in the training data for training the evaluation model.

[0064] The resale estimation device 100 may include information indicating the "number of days of stay during the event period" in the teacher data. The number of days of stay during the event period is the number of days an event is held at an event facility that has a predetermined relationship with the accommodation facility on the day the reservation is scheduled to check in and the day the reservation is scheduled to check out. The number of days of stay during the event period is generated by identifying the check-in date and the check-out date based on the past reservation information (reservation database D111) and identifying the number of days a target event is held at the target event facility based on the event database D112. Reservations for resale purposes tend to be made multiple times during the period when the event is held. Therefore, the resale estimation device 100 can accurately estimate whether or not a reservation is for resale purposes by having an evaluation model learn using information indicating the number of days of stay during the event period included in the teacher data.

[0065] The resale estimation device 100 may include information indicating "the difference between the name of the person who made the reservation and the name of the guest staying at the hotel" in the teacher data. The difference between the name of the person who made the reservation and the name of the guest staying at the hotel is information indicating whether there is a difference between the name of the person who made the reservation and the name of the person who plans to stay at the hotel. The difference between the name of the person who made the reservation and the name of the guest staying at the hotel is information that can be identified based on past reservation information. In reservations for resale purposes, the name of the person who made the reservation and the name of the guest staying at the hotel tend to match in most reservations in normal use, but tend to show a mismatch in almost all reservations. Therefore, the resale estimation device 100 can accurately estimate whether the reservation is for resale purposes by having the evaluation model learn using information indicating the difference between the name of the person who made the reservation and the name of the guest staying at the hotel.

[0066] The resale estimation device 100 may include information indicating "the difference between the reservation person's phone number and the guest's phone number" in the teacher data. The difference between the reservation person's phone number and the guest's phone number is information indicating whether or not there is a difference between the reservation person's phone number and the phone number of the guest who plans to stay, as registered by the reservation person. The difference between the reservation person's phone number and the guest's phone number is information that can be identified based on past reservation information. In reservations for resale purposes, the reservation person's phone number and the guest's phone number tend to match in most reservations in normal use, but tend to show a mismatch in almost all reservations. Therefore, the resale estimation device 100 can accurately estimate whether or not the reservation is for resale by having the evaluation model learn using information indicating the difference between the reservation person's phone number and the guest's phone number in the teacher data.

[0067] The resale estimation device 100 may include information indicating the "number of unique guest names within X days of reservation" in the teacher data. The number of unique guest names within X days of reservation is, for example, information indicating the number of unique guest names, which is a history of changes in the name of the guest who plans to stay within a predetermined number of days from the day the reservation by the reservationr is accepted. The number of unique guest names within X days of reservation is information that can be specified by specifying the transition of the guest name in one reservation in the reservation database D111. In a reservation for resale purposes, there is a tendency that the name of the guest who plans to stay is changed when the resale is completed. Therefore, the resale estimation device 100 can accurately estimate whether the reservation is for resale or not by including information indicating the number of unique guest names within X days of reservation in the teacher data and training the evaluation model. In this case, for example, when the resale estimation device 100 acquires current reservation information in which the guest name has been changed, the identification determination unit 130 refers to the reservation database D111 to identify the past reservation information of the current reservationr, and specifies the above-mentioned number of unique guests. Then, the resale estimation device 100 inputs the identified number of unique items into the evaluation model.

[0068] The resale estimation device 100 may include information indicating the "number of unique guest phone numbers within X days of reservation" in the teacher data. The number of unique guest phone numbers within X days of reservation is, for example, information indicating the number of unique phone numbers of guests who plan to stay within a predetermined number of days from the day when the reservation by the reservationr is accepted. The number of unique guest phone numbers within X days of reservation is information that can be specified by specifying the change in the phone number of the guests who plan to stay in one reservation in the reservation database D111. In a reservation for the purpose of resale, there is a tendency that the phone number of the guests who plan to stay changes when the resale is completed. Therefore, the resale estimation device 100 can accurately estimate whether the reservation is for resale or not by including information indicating the number of unique guest phone numbers within X days of reservation in the teacher data and having the evaluation model learn. In this case, for example, when the resale estimation device 100 acquires current reservation information in which the guest phone number has been changed, the identification determination unit 130 refers to the reservation database D111 to identify past reservation information of the current reservationr, and specifies the above-mentioned number of unique phone numbers. Then, the resale estimation device 100 inputs the identified number of unique items into the evaluation model.

[0069] Facility Reservation Management Device 200 Returning to FIG. 1, the functional configuration of the facility reservation management device 200 will be described. As shown in FIG. 1, the facility reservation management device 200 includes, as functional units, for example, a storage unit 210, an acquisition unit 220, and a transmission unit 230. The storage unit 210 has, for example, a reservation database D211 that stores reservation information of current and past reservation holders. The information stored in the reservation database D211 may be, for example, information included in the reservation database D111 shown in FIG. 2. The acquisition unit 220 acquires current reservation information from the reservation holder. In addition, the acquisition unit 220 acquires report information from the resale estimation device 100. The transmission unit 230 transmits the current reservation information acquired from the reservation holder to the resale estimation device 100. In addition, the transmission unit 230 may transmit past reservation information to the resale estimation device 100 in response to a request from the resale estimation device 100.

[0070] ===Processing Procedure=== The processing procedure of the resale detection system 10 will be described with reference to Fig. 6. Fig. 6 is a flowchart showing the processing procedure of the resale detection system 10. Note that Fig. 6 describes, as an example, the processing procedure in which the resale estimation device 100 transmits report information indicating the possibility that the reservation is for the purpose of resale to the facility reservation management device 200 based on current reservation information related to a reservation of an accommodation facility by a booker.

[0071] In step S101 , the facility reservation management device 200 receives, for example, an operational input from a user, and transmits current reservation information regarding accommodation facilities to the resale estimation device 100 .

[0072] In step S102, the resale estimation device 100 stores the current reservation information in the storage unit 110. At this time, the resale estimation device 100, for example, refers to the reservation database D111 and the event database D112, and generates current reservation date determination information and current event holding / non-holding information from the current reservation information.

[0073] In step S103, the resale estimation device 100 inputs, for example, the current reservation date determination information and the current event holding / non-holding information to the evaluation model.

[0074] In step S104, the resale estimation device 100 generates resale evaluation information from, for example, a score indicating the degree to which the reservation is for the purpose of resale from the evaluation model.

[0075] In step S105 , the resale estimation device 100 generates report information based on the resale evaluation information, and transmits the report information to the facility reservation management device 200 .

[0076] In step S106, the facility reservation management device 200 displays the report information on the display unit.

[0077] This allows the resale detection system 10 to accurately estimate whether or not a reservation is for the purpose of resale, based on current reservation information for accommodation facilities.

[0078] ===Hardware Configuration=== 7, an example of a hardware configuration for implementing the resale estimation device 100 and the facility reservation management device 200 using a computer 1000 will be described. Note that the various functions of the resale estimation device 100 and the facility reservation management device 200 can be implemented by multiple devices.

[0079] FIG. 7 is a diagram showing an example of a hardware configuration of a computer. As shown in FIG. The computer 1000 includes, for example, a processor 1001 , a memory 1002 , a storage device 1003 , an input I / F unit 1004 , a data I / F unit 1005 , a communication I / F unit 1006 , and a display unit 1007 .

[0080] The processor 1001 is a control unit that controls various processes in the computer 1000 by executing programs stored in the memory 1002 .

[0081] The memory 1002 is a storage medium such as a RAM (Random Access Memory), etc. The memory 1002 temporarily stores the program code of the program executed by the processor 1001 and data required when the program is executed.

[0082] The storage device 1003 is a non-volatile storage medium such as a hard disk drive (HDD), a flash memory, etc. The storage device 1003 stores an operating system and various programs for implementing the above-mentioned components.

[0083] The input I / F unit 1004 is a device for receiving input from a user. Specific examples of the input I / F unit 1004 include a keyboard, a mouse, a touch panel, various sensors, a wearable device, etc. The input I / F unit 1004 may be connected to the computer 1000 via an interface such as a Universal Serial Bus (USB).

[0084] The data I / F unit 1005 is a device for inputting data from outside the computer 1000. A specific example of the data I / F unit 1005 is a drive device for reading data stored in various storage media. The data I / F unit 1005 may be provided outside the computer 1000. In that case, the data I / F unit 1005 is connected to the computer 1000 via an interface such as a USB.

[0085] The communication I / F unit 1006 is a device for performing data communication via the Internet N, either wired or wirelessly, with an external device of the computer 1000. The communication I / F unit 1006 may be provided outside the computer 1000. In that case, the communication I / F unit 1006 is connected to the computer 1000 via an interface such as a USB.

[0086] The display unit 1007 is a device for displaying various types of information. Specific examples of the display unit 1007 include a liquid crystal display, an organic EL (Electro-Luminescence) display, a display of a wearable device, and the like. The display unit 1007 may be provided outside the computer 1000. In that case, the display unit 1007 is connected to the computer 1000 via, for example, a display cable. In addition, when a touch panel is adopted as the input I / F unit 1004, the display unit 1007 can be configured integrally with the input I / F unit 1004.

[0087] ===Summary=== <1> The resale detection system 10 includes an acquisition unit 120 that acquires current reservation information, which is information regarding a current reservation of an accommodation by a specified reservationr, past reservation date determination information, which is information regarding past reservations of the accommodation, and which includes a determination result indicating whether or not the past reservation date satisfies a specified condition, and past event holding status information, which is specified based on an event database D112 (event information) including information regarding an event facility where an event will be held, indicating whether an event was held on any of the days included in the stay period of the past reservationr at the past reservation accommodation, at a past target event facility, which is an event facility having a specified relationship with the past reservation accommodation, and an output unit 160 that outputs information indicating the degree of possibility that the currently reserved accommodation has been reserved for the purpose of resale, which is predicted by inputting, into an evaluation model trained using resale information indicating whether or not a previously reserved accommodation has been reserved for the purpose of resale, current reservation date determination information including a determination result indicating whether or not a current reservation date of a currently reserved accommodation that is an accommodation booked by a specific reservationr, identified based on the current reservation information, satisfies a specific condition, and current event holding information indicating whether or not an event will be held at a current target event facility that is an event facility that has a specific relationship with the currently reserved accommodation, identified based on the event information, on any date included in the stay period of the specific reservationr at the currently reserved accommodation, identified based on the event information, to an evaluation model trained using the above as training data. This enables the resale detection system 10 to estimate whether or not an accommodation has been reserved by a reservationr for the purpose of resale.

[0088] <2> The resale detection system 10 further includes a learning unit 140 that learns an evaluation model using the past reservation date determination information, the past event holding / non-holding information, and the resale information as training data. This enables the resale detection system 10 to estimate whether or not a reservation for an accommodation facility has been made by a person with the intention of reselling the reservation.

[0089] <3> In the resale detection system 10, the evaluation model is an evaluation model trained using as training data past reservation date determination information including a first determination result indicating whether a past reservation date is included within a predetermined number of days of the announcement date of an event held at a target event facility in the past specified based on the event information, past event holding status information, and resale information. As a result, the resale detection system 10 uses an evaluation model trained with information capturing the tendency for the purpose of resale in relation to the relationship between the reservation date when accommodation facilities surrounding the target event facility were booked and the announcement date of the event held at the target event facility, and thus it is possible to accurately estimate whether the purpose of resale is present.

[0090] <4> In the resale detection system 10, the evaluation model is an evaluation model trained using as training data past reservation date determination information, past event holding status information, and resale information, including a second determination result indicating whether or not the past target event, which is an event held at a past target event facility, was included in the period from the announcement date of the announcement of the holding of the past target event to the holding date of the past target event, and a third determination result indicating whether or not multiple reservations were made by one of the past reservation holders at an accommodation facility having a predetermined relationship with the past target event facility on the holding date of the past target event. As a result, the resale detection system 10 uses an evaluation model trained with information capturing the tendency for the purpose of resale in relation to the date of announcement of the holding of the event at the target event facility, and thus it is possible to accurately estimate whether or not the purpose of resale is the purpose of resale.

[0091] <5> In the resale detection system 10, the evaluation model is an evaluation model trained using as training data the past reservation date determination information including the second determination result and the third determination result indicating whether or not multiple reservations were made by one of the past reservation users on the date of the past target event at an accommodation facility having a predetermined relationship with the past target event facility, the past event holding status information, and the resale information. As a result, the resale detection system 10 uses an evaluation model trained with information capturing the tendency for the purpose of resale, that is, whether or not one reservation user has made multiple reservations at accommodation facilities around the target event facility, and therefore can accurately estimate whether or not the purpose of resale is present.

[0092] The above-described embodiments are intended to facilitate understanding of the present invention, and are not intended to limit the present invention. The elements of the embodiments, as well as their arrangements, materials, conditions, shapes, sizes, etc., are not limited to those shown as examples and can be changed as appropriate. In addition, configurations shown in different embodiments can be partially substituted or combined with each other. [Explanation of symbols]

[0093] 10... resale detection system, 100... resale estimation device, 110... memory unit, 120... acquisition unit, 130... identification judgment unit, 140... learning unit, 150... estimation unit, 160... output unit, 200... facility reservation management device.

Claims

1. An acquisition unit that acquires current reservation information, which is information regarding a current reservation of an accommodation facility by a specified reservation person; past reservation date determination information including a past reservation date of a past reservation accommodation facility that is a reservation made by a past reservation user who is a past reservation user, the past reservation date being specified based on past reservation information, which is information related to past reservations of the accommodation facility, and a determination result indicating whether the past reservation date satisfies a predetermined condition; past event holding information indicating whether an event was held on any of the days included in the stay period of the past reservation user at the past reservation accommodation facility at a past target event facility that has a predetermined relationship with the past reservation accommodation facility, the past target event facility being identified based on event information including information regarding the event facility where the event is to be held; resale information indicating whether the previously reserved accommodation was reserved for resale purposes; and For the evaluation model trained using the training data, current reservation date determination information including a determination result indicating whether or not the current reservation date of the currently reserved accommodation facility, which is the accommodation facility reserved by the specified reservation user, specified based on the current reservation information satisfies the specified condition; and Current event holding status information indicating whether or not an event will be held at a current target event facility, which is the event facility having the specified relationship with the currently reserved accommodation facility and which is identified based on the event information, on any date included in the stay period of the specified reservation user at the currently reserved accommodation facility, which is identified based on the event information; and an output unit that outputs information indicating a degree of possibility that the currently reserved accommodation facility has been reserved for the purpose of resale, the degree being predicted by inputting the above information; Includes a resale detection system.

2. a learning unit that learns the evaluation model using the past reservation date determination information, the past event holding / non-holding information, and the resale information as training data; The resale detection system of claim 1 .

3. The evaluation model is the past reservation date determination information including a first determination result indicating whether or not the past reservation date is included within a predetermined number of days from the publication date of the event held at the past target event facility specified based on the event information; The information on whether past events have been held; The resale information; This is an evaluation model trained using the training data. The resale detection system according to claim 1 or 2.

4. The evaluation model is A second determination result indicating whether or not the event was included in the period from the announcement date of the announcement of the holding of the past target event, which is an event held at the past target event facility, to the holding date of the past target event; a third determination result indicating whether or not one of the past reservation users has made multiple reservations on the date of the past target event at an accommodation facility having the predetermined relationship with the past target event facility; The past reservation date determination information including The information on whether the past event was held, The resale information; This is an evaluation model trained using the training data. The resale detection system according to claim 1 or 2.

5. The evaluation model is the second determination result; and the third determination result indicating whether or not a plurality of reservations have been made by one of the past reservation users on the date of the past target event at an accommodation facility having the predetermined relationship with the past target event facility; The past reservation date determination information including The information on whether the past event was held, The resale information; This is an evaluation model trained using the training data. The resale detection system of claim 4 .

6. A trained model for causing a computer to function to output information indicating the degree of likelihood that a given reservationr who reserves an accommodation has reserved the accommodation for the purpose of resale, past reservation date determination information including a past reservation date of a past reservation accommodation facility that is a reservation made by a past reservation user who is a past reservation user, the past reservation date being specified based on past reservation information, which is information related to past reservations of the accommodation facility, and a determination result indicating whether the past reservation date satisfies a predetermined condition; past event holding information indicating whether an event was held on any of the days included in the stay period of the past reservation user at the past reservation accommodation facility at a past target event facility that has a predetermined relationship with the past reservation accommodation facility, the past target event facility being identified based on event information including information regarding the event facility where the event is held; and resale information indicating whether the previously reserved accommodation was reserved for resale purposes; and This was trained using the training data, current reservation date determination information including a determination result indicating whether or not the current reservation date of the currently reserved accommodation facility, which is the accommodation facility reserved by the specified reservation user, satisfies the specified condition, which is specified based on current reservation information, which is information regarding the current reservation of the accommodation facility by the specified reservation user; and Current event holding status information indicating whether or not an event will be held at a current target event facility, which is the event facility having the specified relationship with the currently reserved accommodation facility and which is identified based on the event information, on any date included in the stay period of the specified reservation user at the currently reserved accommodation facility, which is identified based on the event information; and By inputting, A trained model for causing a computer to function to output information indicating the degree of likelihood that the specified reservation user has reserved the currently reserved accommodation for the purpose of resale.

7. The computer Obtaining current reservation information, which is information regarding a current reservation made by a given reservation party at the accommodation; past reservation date determination information including a past reservation date of a past reservation accommodation facility that is a reservation made by a past reservation user who is a past reservation user, the past reservation date being specified based on past reservation information, which is information related to past reservations of the accommodation facility, and a determination result indicating whether the past reservation date satisfies a predetermined condition; past event holding information indicating whether an event was held on any of the days included in the stay period of the past reservation user at the past reservation accommodation facility at a past target event facility that has a predetermined relationship with the past reservation accommodation facility, the past target event facility being identified based on event information including information regarding the event facility where the event is held; and resale information indicating whether the previously reserved accommodation was reserved for resale purposes; and For the evaluation model trained using the training data, current reservation date determination information including a determination result indicating whether or not the current reservation date of the currently reserved accommodation facility, which is the accommodation facility reserved by the specified reservation user, specified based on the current reservation information satisfies the specified condition; and Current event holding status information indicating whether or not an event will be held at a current target event facility, which is the event facility having the specified relationship with the currently reserved accommodation facility and which is identified based on the event information, on any date included in the stay period of the specified reservation user at the currently reserved accommodation facility, which is identified based on the event information; and outputting information indicating the degree of possibility that the currently reserved accommodation has been reserved for the purpose of resale, which is predicted by inputting the Implement resale detection methods.

8. On the computer, Obtaining current reservation information, which is information regarding a current reservation made by a given reservation party at the accommodation; past reservation date determination information including a past reservation date of a past reservation accommodation facility that is a reservation made by a past reservation user who is a past reservation user, the past reservation date being specified based on past reservation information, which is information related to past reservations of the accommodation facility, and a determination result indicating whether the past reservation date satisfies a predetermined condition; past event holding information indicating whether an event was held on any of the days included in the stay period of the past reservation user at the past reservation accommodation facility at a past target event facility that has a predetermined relationship with the past reservation accommodation facility, the past target event facility being identified based on event information including information regarding the event facility where the event is held; and resale information indicating whether the previously reserved accommodation was reserved for resale purposes; and For the evaluation model trained using the training data, current reservation date determination information including a determination result indicating whether or not the current reservation date of the currently reserved accommodation facility, which is the accommodation facility reserved by the specified reservation user, specified based on the current reservation information satisfies the specified condition; and Current event holding status information indicating whether or not an event will be held at a current target event facility, which is the event facility having the specified relationship with the currently reserved accommodation facility and which is identified based on the event information, on any date included in the stay period of the specified reservation user at the currently reserved accommodation facility, which is identified based on the event information; and outputting information indicating the degree of possibility that the currently reserved accommodation has been reserved for the purpose of resale, which is predicted by inputting the A program that executes the following.

Images