Coupon processing device, coupon processing method, and program

Abstract

Traditionally, it has not been easy to increase the effectiveness of the coupons that are distributed. [Solution] The coupon processing device 1 comprises a learning information management unit 112 which stores learning information based on two or more training data having coupon information distributed to users, product purchase information and user attribute values; a first coupon determination unit 134 which, for each of the first proportion of users among the two or more users, determines purchase information that satisfies the selection conditions using user attribute values, coupon information of the selected coupon and learning information, and determines from two or more types of coupons which are coupons corresponding to the purchase information to be distributed to the user; a second coupon determination unit 135 which, for each of the second proportion of users among the two or more users, determines the coupon to be distributed to the user without using learning information; and a coupon output unit 141 which outputs a coupon identifier that identifies the coupon to be distributed to each of the two or more users. This makes it possible to increase the effectiveness of the coupons to be distributed.

Description

Technical Field

[0001] The present invention relates to a coupon processing device that determines coupons to be distributed to users and the like.

Background Art

[0002] Conventionally, there has been an information providing device aimed at increasing the sales profit obtained from the sale of products (see Patent Document 1). Such an information providing device includes an acquisition unit that acquires information indicating the sales promotion cost of a product, and a determination unit that determines an amount or ratio to be used as a provision cost for providing a profit to a user who purchases the product among the sales promotion costs indicated by the information acquired by the acquisition unit, according to the appeal effect of the content related to the sale of the product.

[0003] Also, there has been a technology that enables the distribution of electronic coupons considering the movement state of users (see Patent Document 2).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] [[ID=4上]] However, in the prior art, it has not been easy to enhance the effectiveness of the distributed coupons.

Means for Solving the Problems

[0006] It should be noted that there seems to be a mislabeling in line ID=42 where it says "上" instead of a correct number or text. I've translated it as it is for now. If this is an error, please correct it in the original text for a more accurate translation.The coupon processing device of the first invention comprises: a user management unit that stores user information having one or more user attribute values ​​for two or more users, including two or more users to whom coupons are to be distributed; a learning information management unit that stores learning information based on two or more training data having coupon information relating to coupons previously distributed to users from two or more types of coupons, purchase information relating to the user's purchase of products, and one or more user attribute values; a first coupon determination unit that, for each of the first proportion of users among the two or more users, obtains one or more user attribute values ​​from the user management unit, determines purchase information that satisfies the selection criteria using one or more user attribute values, coupon information of a coupon selected from two or more types of coupons, and learning information, and determines a coupon that corresponds to the purchase information and is to be distributed to the user from two or more types of coupons; a second coupon determination unit that, for each of the second proportion of users among the two or more users, determines a coupon to be distributed to the user without using learning information; and a coupon output unit that outputs a coupon identifier that identifies the coupon to be distributed to each of the two or more users.

[0007] This configuration can increase the effectiveness of the coupons being distributed.

[0008] Furthermore, the coupon processing device of the second invention, compared to the first invention, is a learning model obtained by performing machine learning training using two or more training data sets in which coupon information and one or more user attribute values ​​are explanatory variables and purchase information is the objective variable; the first coupon determination unit performs machine learning prediction processing using coupon information, one or more user attribute values ​​and the learning model for each user representing a first proportion of two or more users and for each coupon information of two or more types of coupons, obtains purchase information, determines purchase information that satisfies the selection condition that the purchase information is maximized, and determines the coupon corresponding to the purchase information, which is a coupon to be distributed to the user from two or more types of coupons.

[0009] This configuration can increase the effectiveness of the coupons being distributed.

[0010] Furthermore, the coupon processing device of this third invention, compared to the second invention, includes a user management unit whose user information includes the user's coupon information and the user's purchase information, and periodically obtains coupon information, purchase information, and one or more user attribute values ​​from the user management unit for two or more user information sets, constructs two or more training data sets, and further comprises a learning unit that performs machine learning learning processing using the two or more training data sets to obtain a learning model, and the learning model used by the first coupon determination unit is the learning model obtained by the learning unit, which is the coupon processing device.

[0011] This configuration can increase the effectiveness of the coupons being distributed.

[0012] Furthermore, the coupon processing device of the fourth invention is a coupon processing device that, in addition to any one of the first to third inventions, has a user management unit whose user information includes purchase information, and further comprises a score acquisition unit that acquires a user score for each of two or more users using each user's purchase information, a first coupon determination unit that determines which coupons to distribute to a first percentage of users whose scores satisfy the coupon distribution conditions, and a second coupon determination unit that determines which coupons to distribute to a second percentage of users whose scores satisfy the coupon distribution conditions.

[0013] This configuration allows for narrowing down the users to whom coupons are distributed, thereby increasing the effectiveness of the coupons.

[0014] Furthermore, the coupon processing device of this fifth invention differs from the fourth invention in that the user information in the user management unit includes purchase information from two or more points in time, the score acquisition unit uses the purchase information from two or more points in time to determine a class that indicates the user's purchase tendency, the coupon processing device determines a class for the user from among the two or more classes, and acquires a score using that class.

[0015] This configuration allows for narrowing down the users to whom coupons are distributed, thereby increasing the effectiveness of the coupons.

[0016] In addition, the coupon processing device of the sixth invention further includes a coupon management unit that stores coupon information having a coupon identifier and a coupon amount for each of two or more types of coupons with respect to any one of the first to fifth inventions. The first coupon determination unit sequentially determines, for each user among the number of users corresponding to the first ratio among two or more users, the coupon to be distributed to the user, obtains the amount corresponding to the determined coupon from the coupon management unit, calculates the accumulation of the amounts, determines whether the accumulation has reached the budget amount, and if the accumulation has reached the budget amount, does not perform the process of further determining the user's coupon.

[0017] With such a configuration, budget management of the entire distributed coupons can be performed.

[0018] In addition, the coupon processing device of the seventh invention, with respect to any one of the first to sixth inventions, the user information in the user management unit includes the user's notification destination information, and the coupon output unit distributes coupon information identified by a coupon identifier that identifies the coupon to be distributed to the notification destination specified by the notification destination information of each of two or more users.

[0019] With such a configuration, the effectiveness of the distributed coupons can be enhanced.

Advantages of the Invention

[0020] According to the coupon processing device of the present invention, the effectiveness of the distributed coupons can be enhanced.

Brief Description of the Drawings

[0021] [Figure 1] Conceptual diagram of information system A in Embodiment 1 [Figure 2] Block diagram of the same information system A [Figure 3] Flowchart for explaining the operation example of the coupon processing device 1 [Figure 4] Flowchart for explaining an example of the score acquisition process [Figure 5] Flowchart for explaining the example of the first coupon determination process [Figure 6] Flowchart for explaining the example of the second coupon determination process [Figure 7] Flowchart for explaining the example of the coupon distribution process [Figure 8] Flowchart for explaining the example of the teacher data acquisition process [Figure 9] Flowchart for explaining the example of the user attribute value accumulation process

Mode for Carrying Out the Invention

[0022] Hereinafter, embodiments of a coupon processing apparatus and the like will be described with reference to the drawings. In the embodiments, components denoted by the same reference numerals perform the same operations, and thus repeated description may be omitted.

[0023] (Embodiment 1) In the present embodiment, based on learning information having coupon information regarding coupons distributed to a user in the past, purchase information regarding the purchase of products by the user, and user attribute values, in two or more pieces of teacher data, a coupon to be distributed to each target person at the first ratio is determined from among two or more types of coupons, and a coupon to be distributed to each target person at the second ratio is determined without using the learning information. A coupon processing apparatus will be described. Note that the learning information is, for example, a learning model obtained by a learning process of machine learning.

[0024] In the present embodiment, a coupon processing apparatus that constructs a learning model periodically (for example, every month) will be described.

[0025] In the present embodiment, based on past purchase information of products by users, a score for each user is obtained, and a coupon processing apparatus that determines, as coupon distribution target persons, users having a top score from all users will be described.

[0026] In this embodiment, a coupon processing device is described that classifies users into one of three or more classes and obtains a score for the user using the user's purchase information and class.

[0027] This embodiment describes a coupon processing device that manages the amount for each type of coupon and the budget amount for coupon distribution, and determines which coupons to distribute to eligible individuals until the budget amount is reached.

[0028] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is irrelevant. Information X and information Y may be linked, may reside in the same buffer, may information X be contained in information Y, or information Y may be contained in information X, and so on.

[0029] Furthermore, in this specification, selecting or determining information Z means obtaining information Z, obtaining a pointer to information Z, obtaining the ID of information Z, setting a flag on information Z, etc., and it is sufficient to be able to access information Z.

[0030] Figure 1 is a conceptual diagram of information system A in this embodiment. Information system A comprises a coupon processing device 1 and one or more terminal devices 2.

[0031] Coupon Processing Device 1 is a device that determines which coupons to distribute to users. Coupon Processing Device 1 may also distribute coupons to users. Coupon Processing Device 1 can be, for example, a cloud server or an ASP server, but its type is not limited.

[0032] Terminal device 2 is the user's device that receives coupon information. Terminal device 2 can be a personal computer, smartphone, or tablet device, but the type is not limited.

[0033] Figure 2 is a block diagram of information system A in this embodiment.

[0034] The coupon processing device 1 comprises a storage unit 11, a receiving unit 12, a processing unit 13, and a transmission unit 14. The storage unit 11 comprises a user management unit 111, a learning information management unit 112, and a coupon management unit 113. The processing unit 13 comprises a learning unit 131, a score acquisition unit 132, a user selection unit 133, a first coupon determination unit 134, and a second coupon determination unit 135. The transmission unit 14 comprises a coupon output unit 141.

[0035] The terminal device 2 includes a terminal storage unit 21, a terminal receiving unit 22, a terminal processing unit 23, a terminal transmission unit 24, a terminal receiving unit 25, and a terminal output unit 26.

[0036] The storage unit 11, which constitutes the coupon processing device 1, stores various types of information. These types of information include, for example, user information, learning information, and coupon information, which will be described later.

[0037] The user management unit 111 stores two or more user information entries. User information refers to information about a user. Here, a user includes users who are eligible to receive coupons. User information has one or more user attribute values. User information may also have one or more lottery entry entries, one or more code reading entry entries, and one or more coupon usage entries. User information may also have one or more purchase entries.

[0038] User attribute values ​​can be either static or dynamic. Static user attribute values ​​are those that do not change dynamically. Examples of static user attribute values ​​include gender, age, residential area, and address. Dynamic user attribute values ​​are those that can change dynamically. Examples of dynamic user attribute values ​​include lottery information, code reading information, and coupon-related information.

[0039] Lottery information refers to information about lotteries in which users can spend points to receive rewards. Rewards can be, for example, product exchange vouchers or product discount coupons, but any type of reward is acceptable as long as it provides a benefit to the user. Lottery information is, for example, statistically processed information from the lottery implementation information described later. Lottery information includes, for example, the current number of lottery points held, the cumulative number of lottery points used, and the number of times a user has participated in each type of lottery based on the number of points required. The current number of lottery points held is the current number of points held to participate in the lottery. Points increase based on, for example, the code reading implementation information described later. The cumulative number of lottery points used is the total number of points used to participate in the lottery. The period for counting the cumulative number of lottery points used can be the entire period (from user registration to the present) or a specific period (for example, the last 6 months). The number of times a user has participated in each type of lottery based on the number of points required is the number of times a user has participated in each type of lottery in a situation where there are two or more types of lotteries with different number of points required to participate.

[0040] Code reading information refers to information about the scanning of a code (e.g., a QR code (registered trademark) or barcode). The code comes with the product purchased by the user. The code may appear on the surface of a sticker attached to the product, or it may be printed on the product. The code may contain, for example, the product identifier and the place of purchase of the purchased product. Points may be awarded to the user of terminal device 2 that scanned the code upon scanning. Code reading information is, for example, statistically processed information obtained from the code scanning implementation information described later. Code reading information includes, for example, the total number of past scans, the number of scans per day during the membership registration period, the total number of scans by day of the week, the total number of scans by place of purchase, the number of scans by month, and the number of weeks with scans in the most recent X (X is a natural number of 2 or more) weeks. The total number of past scans is the total number of times the code has been scanned in the past. The total number of scans by place of purchase is the total number of times the code has been scanned in the past for each of one or more places of purchase. A "buying area" can simply be described as a location. The size of each buying area is irrelevant. A buying area can be, for example, a city, town, or village, or a grid of a predetermined size.

[0041] Coupon-related information refers to information about a user's coupons. Coupon-related information includes, for example, statistically processed information such as the coupon usage information described later. Coupon-related information includes, for example, the current number of coupons held for each coupon amount, the past usage rate for each coupon amount, the past number of uses for each coupon amount, the usage rate for each coupon amount over a specified period (e.g., the last 3 months), and the number of uses for each coupon amount over a specified period. Note that the coupon amount is the amount discounted by the coupon. Coupon amounts are, for example, "10 yen," "30 yen," and "50 yen."

[0042] Lottery implementation information refers to information regarding a user's participation in a lottery. This information may include, for example, the user identifier of the user who participated, the number of points required for the lottery, and time information. Lottery implementation information may also include the lottery result (winner or loser). Time information refers to information that specifies a time. This time information may include, for example, the date, date and time, or time of day.

[0043] Code scanning information refers to information about the act of scanning a code. This information may include, for example, the user identifier of the user who scanned the code, a code identifier that identifies the scanned code, time information, location information or purchase location information, a product identifier that identifies the product purchased by the user, and the purchase price of the product. Note that the code identifier may be the same as the product identifier.

[0044] Coupon usage information refers to information about the use of a coupon. This information may include, for example, the user identifier of the user who used the coupon, the coupon identifier that identifies the coupon used, and the coupon amount.

[0045] Purchase information refers to information about a user's purchase of a product. This information may include, for example, a product identifier that identifies the product purchased by the user, and the purchase price. Purchase information may be the same as the code reading execution information.

[0046] The learning information management unit 112 stores learning information. This learning information includes, for example, a learning model, a correspondence table (described later), and two or more training data sets. (1) When the learning information is a learning model

[0047] A learning model is information constructed through the learning process of machine learning, and is used in the prediction process of machine learning. A learning model can also be called a learner, classifier, or classification model. The machine learning algorithm can be deep learning, random forest, decision tree, SVR, etc. Furthermore, various machine learning functions and various existing libraries can be used in machine learning, such as the TensorFlow® library, the R language's random forest module, TinySVM, etc.

[0048] The learning model here is a model obtained by providing two or more training data sets to a machine learning learning module and executing that module. The training data consists of coupon information, purchase information, and one or more user attribute values.

[0049] Coupon information refers to information about coupons. This includes, for example, a coupon identifier and the coupon's value. In this context, the coupon information refers to information about coupons previously distributed to users from among two or more types of coupons.

[0050] The one or more user attribute values ​​here preferably include the one or more dynamic attribute values ​​described above. The one or more dynamic attribute values ​​here preferably include one or more attribute values ​​from among lottery information, code reading information, and coupon-related information.

[0051] Here, coupon information and user attribute values ​​of 1 or more are explanatory variables, while purchase information is the dependent variable.

[0052] The learning model is used when the first coupon determination unit 134, described later, performs machine learning prediction processing. (2) When the learning information is a correspondence table

[0053] A correspondence table contains two or more correspondence information based on training data. Each correspondence information has a vector with two or more explanatory variables as elements and a target variable. Another correspondence information has a vector with one or more coupon information and one or more user attribute values ​​as elements and purchase information. In this case, the purchase information contains the coupon identifier of the coupon used at the time of purchase.

[0054] The correspondence table is used when the first coupon determination unit 134, described later, obtains the similarity of the vectors and uses that similarity to determine the coupon. (3) When the learning information consists of two or more training data points

[0055] The two or more training data points mentioned here are the information provided to the generating AI. In other words, if the learning information consists of two or more training data points, the first coupon determination unit 134 is asked to determine which coupon to send to the user from the generating AI.

[0056] The coupon management unit 113 stores coupon information for two or more types of coupons. Coupon information refers to information about the coupon. For example, coupon information includes a coupon identifier and the coupon amount.

[0057] The receiving unit 12 receives various instructions and information. These instructions and information include, for example, coupon distribution instructions, user registration instructions, lottery implementation information, code reading implementation information, and coupon usage information.

[0058] The receiving unit 12 receives, for example, coupon distribution instructions from the terminal device 2 of the person in charge at the company operating the coupon processing device 1. The receiving unit 12 receives user registration instructions from the terminal device 2 used by the user. The receiving unit 12 receives lottery implementation information, code reading implementation information, or coupon usage information from the terminal device 2 used by the user, or from devices not shown. Devices not shown include, for example, the server of the company operating the coupon processing device 1, or devices that constitute the store's POS system.

[0059] A user registration instruction is an instruction for user registration. A user registration instruction, for example, has one or more static user attribute values.

[0060] The processing unit 13 performs various processes. These processes include, for example, those performed by the learning unit 131, the score acquisition unit 132, the user selection unit 133, the first coupon determination unit 134, and the second coupon determination unit 135.

[0061] When the receiving unit 12 receives a user registration instruction, the processing unit 13 performs user registration processing corresponding to the user registration instruction. For example, the processing unit 13 obtains a unique user identifier, constructs user information having the user identifier and one or more user attribute values ​​provided in the user registration instruction, and stores it in the user management unit 111.

[0062] The learning unit 131 obtains coupon information, purchase information, and one or more user attribute values ​​from the user management unit 111 for each of the two or more user information entries, and constructs two or more training data sets. Next, the learning unit 131 uses these two or more training data sets to perform machine learning training and obtain a learning model.

[0063] The learning unit 131 may, for example, obtain coupon information, purchase information, and one or more user attribute values ​​from the user management unit 111 for a predetermined period from the present, construct two or more training data sets, and then use these two or more training data sets to perform machine learning training and obtain a learning model. The predetermined period is, for example, six months, but is not limited to that.

[0064] It is preferable for the learning unit 131 to periodically perform machine learning training and acquire a trained model.

[0065] The learning unit 131 may configure two or more training data sets as described above, and configure a correspondence table in which each of the two or more training data sets is used as corresponding information.

[0066] The score acquisition unit 132 acquires a user's score for each of the two or more users using the user's purchase information or one or more user attribute values. Note that the user's purchase information can also be considered a user attribute value.

[0067] The score acquisition unit 132 acquires a higher score, for example, the more the purchase amount or number of purchases indicated in the user's purchase information. The score acquisition unit 132 calculates the user's score using an increasing function that takes the purchase amount or number of purchases indicated in the user's purchase information as parameters.

[0068] The score acquisition unit 132 uses, for example, purchase information from two or more points in time or one or more user attribute values ​​to determine a class that indicates the user's behavioral tendencies. The unit determines one class for the user from among the two or more classes and acquires a score using that class. The score acquisition unit 132 acquires, for example, the score corresponding to the class from the storage unit 11. The two or more classes could be, for example, "increased depth," "prevention of outflow," and "continued maintenance."

[0069] The "Increased Depth" class is a class of users who are increasingly purchasing products and using the codes attached to those products. The score acquisition unit 132, for example, obtains the depth at the most recent point in time (the first time point) using the calculation formula "Depth = Total number of code reads / Number of code reads", and obtains the depth at the point before the first time point (the second time point) using the same calculation formula. The score acquisition unit 132, for example, determines whether the "Increased Depth" condition is met, such as "Depth at the first time point - Depth at the second time point" being equal to or greater than a threshold. If the increased depth condition is met, the class of the user is set to "Increased Depth".

[0070] The "leak prevention" class is a class of users who have purchased a product and used the code attached to that product, and whose actions should be prevented from ceasing. For example, the score acquisition unit 132 might classify a user as "leak prevention" if a continuous user who had been reading codes up until M1 (e.g., "M1=3") months ago has not read a code for M1-1 months (e.g., 2 months) consecutively.

[0071] The "Continued Use" class is the class of users who have purchased a product and have been using the code attached to that product for at least M2 (for example, "M2=2") months. The score acquisition unit 132 classifies users who, for example, had "Code Read" in the previous month and also have the user attribute value "Code Read" in the current month as "Continued Use".

[0072] The user selection unit 133 determines, for each of two or more users, whether the user's score meets the coupon distribution conditions. The user selection unit 133 selects users who meet the coupon distribution conditions. Selecting a user involves, for example, obtaining a user identifier. The coupon distribution conditions include, for example, a score being above or above a threshold, or a score being within the top N (where N is 1 or a natural number greater than or equal to 2).

[0073] The user selection unit 133 may, for example, select users who belong to a specific class. Examples of such classes include "increased depth," "prevention of outflow," and "continued maintenance."

[0074] The first coupon determination unit 134 obtains one or more user attribute values ​​from the user management unit 111 for each of the first proportion of users out of two or more users. Next, the first coupon determination unit 134 uses the one or more user attribute values, the coupon information of the selected coupon from two or more types of coupons, and the learning information to determine purchase information that satisfies the selection criteria, and then determines from two or more types of coupons the coupon corresponding to the purchase information to be distributed to the user. The first proportion is, for example, "90 percent," but is not limited to that. The first proportion may also be "100 percent." Furthermore, the two or more users that form the base when the first coupon determination unit 134 selects the first proportion of users may be users selected by the user selection unit 133, users of the user management unit 111, or randomly selected users, etc.

[0075] The first coupon determination unit 134 may decide which coupons to distribute to only one or more users selected by the user selection unit 133. In other words, the first coupon determination unit 134 may, for example, decide which coupons to distribute to a first-tenth of the number of users whose scores meet the coupon distribution conditions (two or more users).

[0076] The first coupon determination unit 134 may determine which users to distribute coupons to or which coupons to distribute, taking into consideration the total budget amount for all coupons to be distributed. In other words, the first coupon determination unit 134 sequentially determines which coupons to distribute to each user, for each of the first proportion of users out of two or more users, obtains the amount corresponding to the determined coupon from the coupon management unit 113, calculates the cumulative amount, determines whether the cumulative amount has reached the budget amount, and if the cumulative amount has reached the budget amount, it is preferable not to perform the process of determining coupons for users further.

[0077] The first coupon determination unit 134 determines which coupons to distribute to the user, for example, using a learning model, a correspondence table, or a generation AI. (1) When using a learning model

[0078] The first coupon determination unit 134 performs machine learning prediction processing using coupon information, one or more user attribute values, and a learning model for each user representing the first proportion of two or more users, and for each coupon information of two or more types of coupons, to obtain purchase information. Next, the first coupon determination unit 134 determines the purchase information that satisfies the selection condition that the purchase information is the largest, and determines the coupon to be distributed to the user from two or more types of coupons that corresponds to the purchase information. The purchase information here is, for example, the number of products sold relative to the amount of the coupon used at the time of purchase (number of products sold / coupon amount), and the amount of the products sold relative to the amount of the coupon used at the time of purchase (amount of products sold / coupon amount).

[0079] The first coupon determination unit 134 associates, for example, code reading execution information and coupon usage information that are paired with the same user identifier and whose differences in the included time information are within a threshold. Then, the first coupon determination unit 134 uses the number of sales items "1" corresponding to the associated code reading execution information, or the amount of sales items included in the code reading execution information, and the amount of the coupon included in the coupon usage information to obtain the number of sales items in relation to the amount of the coupon used at the time of purchase, or the amount of sales items in relation to the amount of the coupon used at the time of purchase. (2) When using a correspondence table

[0080] The first coupon determination unit 134 performs the following coupon selection process for each user, representing the first percentage of the two or more users. In other words, the first coupon determination unit 134 constructs a vector for each of the two or more types of coupon information, with one or more coupon information items and one or more user attribute values ​​as elements. It then calculates the similarity between this vector and the vectors of the two or more corresponding information items in the correspondence table, and obtains the coupon identifier that is paired with the vector with the highest similarity. (3) When using a generated AI

[0081] The first coupon determination unit 134 acquires two or more training data points. This process is the same as the process performed by the learning unit 131 described above.

[0082] Next, the first coupon determination unit 134 provides the two or more training data to a generating AI (not shown). Then, for each of the two or more users, the first coupon determination unit 134 configures a prompt having one or more user attribute values, prompting the generating AI to answer with a coupon type suitable for the user, provides this prompt to the generating AI, and obtains a coupon identifier to be distributed to each user from the generating AI. Note that the generating AI is, for example, ChatGPT (registered trademark), but is not limited to that.

[0083] The second coupon determination unit 135 determines which coupons to distribute to each user, without using learning information, for each user that is the second-highest percentage of the total number of users (2 or more users). Note that the second-highest percentage is, for example, "10%", but is not limited to any specific number.

[0084] The second coupon determination unit 135 preferably determines which coupons to distribute to the second-highest number of users among the two or more users whose scores meet the coupon distribution conditions. The two or more users who meet the coupon distribution conditions are, for example, users selected by the user selection unit 133, or users corresponding to user information managed by the user management unit 111.

[0085] The second coupon determination unit 135 determines, for example, a coupon to be distributed to each of two or more users, randomly, for each of the second-highest number of users.

[0086] The second coupon determination unit 135 determines, for example, for each of two or more users representing a second-percentage number of users, which coupons to distribute in the order in which the user identifiers are listed and the coupon information is listed.

[0087] The second coupon determination unit 135 may determine coupons for a second-tenths number of users from the coupons determined by the first coupon determination unit 134, without using the learned information. In this case, for the coupons of the second-tenths number of users, the coupons determined by the first coupon determination unit 134 will not be adopted, and the coupons determined by the second coupon determination unit 135 will be adopted.

[0088] The transmitting unit 14 transmits various types of information. These types of information include, for example, a coupon identifier. The coupon identifier is information that identifies the type of coupon. The coupon identifier can also be called the coupon itself. The coupon identifier may be the same as the coupon information. For example, the coupon identifier specifies the amount of discount applied when purchasing a product.

[0089] The coupon output unit 141 outputs a coupon identifier that identifies the coupon to be distributed to each of two or more users. The coupon output unit 141 may also output coupon information that includes the coupon identifier.

[0090] Here, "output" usually refers to transmission to terminal device 2, but it may also be a concept that includes transmission to other external devices, storage on a recording medium, and handover of processing results to other processing devices or other programs.

[0091] The coupon output unit 141 transmits coupon information, identified by a coupon identifier that identifies the coupon to be distributed, to, for example, two or more notification destinations specified by the notification destination information of each user.

[0092] The terminal storage unit 21, which constitutes the terminal device 2, stores various types of information. These types of information include, for example, a user identifier.

[0093] The terminal reception unit 22 receives various information and instructions. These various information and instructions include, for example, user registration instructions, lottery instructions, code scanning instructions, and coupon usage instructions. A lottery instruction is an instruction to conduct a lottery. A code scanning instruction is an instruction to scan a code. A coupon usage instruction is an instruction to use a coupon.

[0094] The means of inputting various information and instructions can be anything, such as a touch panel, keyboard, mouse, or menu screen.

[0095] The terminal processing unit 23 performs various processes. These processes include, for example, converting received information and instructions into information and instructions for transmission. Other processes include, for example, converting received information into information for output.

[0096] The terminal transmission unit 24 transmits various information and instructions to the coupon processing device 1. These various information and instructions include, for example, user registration instructions, lottery implementation information, code reading implementation information, coupon usage information, and user identifiers.

[0097] The terminal receiving unit 25 receives various types of information from the coupon processing device 1. These types of information include, for example, coupon information.

[0098] The terminal output unit 26 outputs various types of information. These types of information include, for example, coupon information.

[0099] Here, "output" is a concept that includes display on a screen, projection using a projector, printing with a printer, sound output, transmission to an external device, storage on a recording medium, and transfer of processing results to other processing devices or other programs.

[0100] The storage unit 11, user management unit 111, learning information management unit 112, coupon management unit 113, and terminal storage unit 21 are preferably made of non-volatile recording media, but can also be made of volatile recording media.

[0101] The process by which information is stored in the storage unit 11, etc. is not relevant. For example, information may be stored in the storage unit 11, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 11, etc., or information input via an input device may be stored in the storage unit 11, etc.

[0102] The receiving unit 12 and the terminal receiving unit 25 are usually implemented by wireless or wired communication means, but they may also be implemented by means of receiving broadcasts.

[0103] The processing unit 13, learning unit 131, score acquisition unit 132, user selection unit 133, first coupon determination unit 134, second coupon determination unit 135, and terminal processing unit 23 can typically be implemented using a processor, memory, etc. The processing procedures of the processing unit 13, etc., are usually implemented in software, and this software is recorded on a recording medium such as ROM. However, it may also be implemented in hardware (dedicated circuitry). The processor can be a CPU, MPU, GPU, etc., and the type is not limited.

[0104] The transmitting unit 14, the coupon output unit 141, and the terminal transmitting unit 24 are usually implemented by wireless or wired communication means, but may also be implemented by broadcasting means.

[0105] The terminal reception unit 22 can be implemented using device drivers for input means such as touch panels and keyboards, or control software for menu screens, etc.

[0106] The terminal output unit 26 may or may not be considered to include output devices such as a display or speakers. The terminal output unit 26 can be implemented using driver software for an output device, or a driver software for an output device and an output device.

[0107] Next, an example of the operation of the coupon processing device 1 will be explained using the flowchart in Figure 3.

[0108] (Step S301) The processing unit 13 determines whether or not it is time to distribute coupons. If it is time to distribute coupons, the process proceeds to step S302; otherwise, it proceeds to step S307. The timing for distributing coupons is, for example, when a predetermined time arrives and a coupon distribution instruction is received from the terminal device 2 of the person in charge at the operating company of the coupon processing device 1.

[0109] (Step S302) The score acquisition unit 132 acquires the scores of two or more users. An example of this score acquisition process will be explained using the flowchart in Figure 4.

[0110] (Step S303) The user selection unit 133 selects two or more users to whom coupons will be distributed, using the scores of two or more users.

[0111] (Step S304) The first coupon determination unit 134 performs the first coupon determination process. An example of the first coupon determination process will be explained using the flowchart in Figure 5.

[0112] (Step S305) The second coupon determination unit 135 performs the second coupon determination process. An example of the second coupon determination process will be explained using the flowchart in Figure 6.

[0113] (Step S306) The coupon output unit 141 distributes coupons to each of the two or more users. The process returns to step S301. An example of such coupon distribution process will be explained using the flowchart in Figure 7.

[0114] (Step S307) The processing unit 13 determines whether or not it is time to compose learning information. If it is time to compose learning information, the process proceeds to step S308; otherwise, it proceeds to step S311. Note that the timing for compiling learning information is usually at regular intervals. For example, the timing for compiling learning information is 12:00 on the 1st of each month, or 0:00 on the last day of March, June, September, and December.

[0115] (Step S308) The learning unit 131 acquires two or more training data points. An example of such training data acquisition process will be explained using the flowchart in Figure 8.

[0116] (Step S309) The learning unit 131 acquires learning information using two or more training data sets. For example, the learning unit 131 performs machine learning training using two or more training data sets and acquires a learning model.

[0117] (Step S310) The learning unit 131 stores the learning information acquired in step S309 in the learning information management unit 112. Return to step S301.

[0118] (Step S311) The receiving unit 12 determines whether or not it has received the lottery information. If it has received the lottery information, it proceeds to step S312; otherwise, it proceeds to step S313. The lottery information is associated with a user identifier.

[0119] (Step S312) The processing unit 13 associates the lottery implementation information received in step S311 with a user identifier and stores it in the user management unit 111. Return to step S301.

[0120] (Step S313) The receiving unit 12 determines whether or not it has received code reading execution information. If it has received code reading execution information, it proceeds to step S314; otherwise, it proceeds to step S315. The code reading execution information is associated with a user identifier.

[0121] (Step S314) The processing unit 13 associates the code reading execution information received in step S313 with a user identifier and stores it in the user management unit 111. The process returns to step S301.

[0122] (Step S315) The receiving unit 12 determines whether or not it has received coupon usage information. If coupon usage information has been received, it proceeds to step S316; otherwise, it proceeds to step S317. Note that the coupon usage information is associated with a user identifier.

[0123] (Step S316) The processing unit 13 associates the coupon usage information received in step S315 with a user identifier and stores it in the user management unit 111. Return to step S301.

[0124] (Step S317) The processing unit 13 determines whether or not it is time to accumulate user attribute values. If it is time to accumulate user attribute values, it proceeds to step S318; otherwise, it returns to step S301.

[0125] (Step S318) The processing unit 13 assigns 1 to counter i.

[0126] (Step S319) The processing unit 13 determines whether or not the i-th user identifier exists in the user management unit 111. If the i-th user identifier exists, the process proceeds to step S320; otherwise, the process returns to step S301.

[0127] (Step S320) The processing unit 13 obtains and stores the user attribute value of the user identified by the i-th user identifier. An example of the user attribute value storage process will be explained using the flowchart in Figure 9. The user attribute value storage process is typically the process of obtaining and storing dynamic user attribute values.

[0128] (Step S321) The processing unit 13 increments counter i by 1. The process returns to step S319.

[0129] In the flowchart shown in Figure 3, processing is terminated by power-off or processing termination interrupts.

[0130] Next, an example of the score acquisition process in step S302 will be explained using the flowchart in Figure 4.

[0131] (Step S401) The score acquisition unit 132 assigns 1 to counter i.

[0132] (Step S402) The score acquisition unit 132 determines whether or not the i-th user for whom the score is to be acquired exists. If the i-th user exists, the process proceeds to step S403; otherwise, it returns to the higher-level process. The score acquisition unit 132 determines, for example, whether or not the information for the i-th user exists in the user management unit 111.

[0133] (Step S403) The score acquisition unit 132 acquires one or more user attribute values ​​from the user management unit 111 that are present in the i-th user information.

[0134] (Step S404) The score acquisition unit 132 assigns 1 to counter j.

[0135] (Step S405) The score acquisition unit 132 determines whether the j-th class exists. If the j-th class exists, the unit proceeds to step S406; otherwise, it proceeds to step S410. The types of classes are usually predetermined. Examples of class types include "depth increase," "outflow prevention," and "continued maintenance."

[0136] (Step S406) The score acquisition unit 132 acquires one or more user attribute values ​​to be used to determine the j-th class. Note that the one or more user attribute values ​​here may include lottery implementation information, code reading implementation information, or coupon usage information.

[0137] (Step S407) The score acquisition unit 132 uses one or more user attribute values ​​obtained in step S406 to determine whether the i-th user belongs to the j-th class. If the user belongs to the j-th class, the unit proceeds to step S408; otherwise, the unit proceeds to step S409.

[0138] (Step S408) The score acquisition unit 132 associates the class identifier of the j-th class with the user identifier of the i-th user. Proceed to step S410.

[0139] (Step S409) The score acquisition unit 132 increments counter j by 1. Return to step S405.

[0140] (Step S410) The score acquisition unit 132 acquires one or more user attribute values ​​other than a class identifier in order to acquire a score.

[0141] (Step S411) The score acquisition unit 132 acquires the user's score using the user's class identifier and one or more user attribute values ​​acquired in step S410. Note that the score acquisition unit 132 may also acquire the user's score using only the user's class identifier.

[0142] (Step S412) The score acquisition unit 132 stores the score acquired in step S411, associating it with the user identifier of the i-th user.

[0143] (Step S413) The score acquisition unit 132 increments counter i by 1. Return to step S402.

[0144] Next, an example of the first coupon determination process in step S304 will be explained using the flowchart in Figure 5.

[0145] (Step S501) The first coupon determination unit 134 obtains the first percentage (for example, 90%) of user identifiers.

[0146] (Step S502) The first coupon determination unit 134 assigns 1 to counter i.

[0147] (Step S503) The first coupon determination unit 134 determines whether the i-th user identifier exists among the user identifiers obtained in step S501. If the i-th user identifier exists, the process proceeds to step S504; otherwise, it returns to the higher-level process.

[0148] (Step S504) The first coupon determination unit 134 obtains one or more user attribute values ​​from the user management unit 111 that are paired with the i-th user identifier.

[0149] (Step S505) The first coupon determination unit 134 assigns 1 to counter j.

[0150] (Step S506) The first coupon determination unit 134 determines whether or not the j-th coupon type exists. If the j-th coupon type exists, the unit proceeds to step S507; otherwise, the unit proceeds to step S512.

[0151] (Step S507) The first coupon determination unit 134 obtains coupon information for the j-th coupon type from the storage unit 11. Note that the coupon information here may consist only of, for example, a coupon identifier.

[0152] (Step S508) The first coupon determination unit 134 obtains the learning model from the learning information management unit 112.

[0153] (Step S509) The first coupon determination unit 134 uses one or more user attribute values, coupon information, and a learning model to perform machine learning prediction processing and obtain purchase information. The purchase information here is, for example, the number of items purchased in the past month.

[0154] (Step S510) The first coupon determination unit 134 temporarily stores the purchase information obtained in step S509, associating it with the coupon identifier of the j-th coupon.

[0155] (Step S511) The first coupon determination unit 134 increments counter j by 1. Return to step S506.

[0156] (Step S512) The first coupon determination unit 134 obtains a coupon identifier corresponding to the largest purchase information and associates the coupon identifier with the i-th user identifier.

[0157] (Step S513) The first coupon determination unit 134 increments counter i by 1. The process returns to step S503.

[0158] Next, an example of the second coupon determination process in step S305 will be explained using the flowchart in Figure 6.

[0159] (Step S601) The second coupon determination unit 135 obtains the user identifiers of the second percentage (for example, 10%). The obtained user identifiers are, for example, the user identifiers of users for whom the first coupon determination unit 134 has not determined whether to distribute coupons.

[0160] (Step S602) The second coupon determination unit 135 assigns 1 to counter i.

[0161] (Step S603) The second coupon determination unit 135 determines whether the i-th user identifier exists among the user identifiers obtained in step S601. If the i-th user identifier exists, the process proceeds to step S604; otherwise, it returns to the higher-level process.

[0162] (Step S604) The second coupon determination unit 135 randomly determines a coupon identifier for the coupon to be distributed to the user identified by the i-th user identifier. The second coupon determination unit 135 associates the coupon identifier with the i-th user identifier.

[0163] (Step S605) The second coupon determination unit 135 increments counter i by 1. Return to step S603.

[0164] Next, an example of the coupon distribution process in step S306 will be explained using the flowchart in Figure 7.

[0165] (Step S701) The coupon output unit 141 assigns 1 to counter i.

[0166] (Step S702) The coupon output unit 141 determines whether or not the i-th user identifier exists among the user identifiers of the users to whom the coupon has been decided. If the i-th user identifier exists, the process proceeds to step S703; otherwise, it returns to the higher-level process.

[0167] (Step S703) The coupon output unit 141 obtains a coupon identifier that is paired with the i-th user identifier.

[0168] (Step S704) The coupon output unit 141 obtains notification destination information paired with the i-th user identifier from the user management unit 111.

[0169] (Step S705) The coupon output unit 141 sends the coupon information identified by the coupon identifier obtained in step S703 to the notification destination indicated by the notification destination information obtained in step S704.

[0170] (Step S706) The coupon output unit 141 increments counter i by 1. Return to step S702.

[0171] Next, an example of the training data acquisition process in step S308 will be explained using the flowchart in Figure 8.

[0172] (Step S801) The learning unit 131 assigns 1 to counter i.

[0173] (Step S802) The learning unit 131 determines whether the i-th user identifier exists in the user management unit 111. If the i-th user identifier exists, the process proceeds to step S803; otherwise, it returns to the higher-level process.

[0174] (Step S803) The learning unit 131 obtains lottery information for a predetermined period from the user management unit 111, which is paired with the i-th user identifier. The predetermined period is, for example, the six months from six months ago to today.

[0175] (Step S804) The learning unit 131 obtains code reading information for a predetermined period from the user management unit 111, which is the code reading information paired with the i-th user identifier.

[0176] (Step S805) The learning unit 131 obtains coupon-related information for a predetermined period from the user management unit 111, which is the coupon-related information paired with the i-th user identifier.

[0177] (Step S806) The learning unit 131 obtains the purchase information that is paired with the i-th user identifier.

[0178] (Step S807) The learning unit 131 constructs training data for the i-th user identifier, which includes lottery information, code reading information, coupon-related information, and purchase information, and stores it in a buffer (not shown). Typically, the lottery information, code reading information, and coupon-related information are explanatory variables, and the purchase information is the dependent variable.

[0179] (Step S808) The learning unit 131 increments counter i by 1. Return to step S802.

[0180] Next, an example of the user attribute value accumulation process in step S320 will be explained using the flowchart in Figure 9.

[0181] (Step S901) The processing unit 13 assigns 1 to counter i.

[0182] (Step S902) The processing unit 13 determines whether there is an i-th lottery piece of information that has not yet been acquired and should be acquired. If there is an i-th lottery piece of information, the unit proceeds to step S903; otherwise, the unit proceeds to step S905.

[0183] (Step S903) The processing unit 13 obtains lottery implementation information paired with the target user identifier from the user management unit 111. The processing unit 13 uses the lottery implementation information to obtain the i-th lottery information.

[0184] (Step S904) The processing unit 13 increments counter i by 1. The process returns to step S902.

[0185] (Step S905) The processing unit 13 assigns 1 to counter i.

[0186] (Step S906) The processing unit 13 determines whether there is an i-th code reading information that has not yet been acquired and should be acquired. If there is an i-th code reading information, the unit proceeds to step S909; otherwise, the unit proceeds to step S911.

[0187] (Step S907) The processing unit 13 obtains code reading execution information paired with the target user identifier from the user management unit 111. The processing unit 13 uses the code reading execution information to obtain the i-th code reading information.

[0188] (Step S908) The processing unit 13 increments counter i by 1. The process returns to step S906.

[0189] (Step S909) The processing unit 13 assigns 1 to counter i.

[0190] (Step S910) The processing unit 13 determines whether there is an i-th coupon-related piece of information that has not yet been acquired and should be acquired. If there is an i-th coupon-related piece of information, the process goes to step S911; otherwise, the process goes to step S913.

[0191] (Step S911) The processing unit 13 obtains coupon usage information paired with the target user identifier from the user management unit 111. The processing unit 13 uses the coupon usage information to obtain the i-th coupon-related information.

[0192] (Step S912) The processing unit 13 increments counter i by 1. The process returns to step S912.

[0193] (Step S913) The processing unit 13 stores the acquired lottery information, acquired code reading information, and acquired coupon-related information in the user management unit 111, paired with the target user identifier. It then returns to the higher-level processing unit.

[0194] As described above, this embodiment makes it possible to increase the effectiveness of the coupons that are distributed.

[0195] Furthermore, according to this embodiment, the number of users to whom coupons are distributed can be narrowed down in order to further enhance the effectiveness of the coupons being distributed.

[0196] Furthermore, according to this embodiment, it is possible to manage the overall budget for the coupons to be distributed.

[0197] The processing in this embodiment may be implemented in software. This software may be distributed via software download or the like. Alternatively, this software may be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments in this specification. The software that implements information system A in this embodiment is the following program. In other words, this program is a program that enables a computer to access a user management unit that stores user information for two or more users, including two or more users to whom coupons are to be distributed, each having one or more user attribute values ​​for two or more users, and a learning information management unit that stores learning information based on two or more training data having coupon information about coupons previously distributed to users from two or more types of coupons, purchase information about the user's purchase of products, and one or more user attribute values. The computer then functions as a first coupon determination unit that, for each of the first proportion of the two or more users, obtains one or more user attribute values ​​from the user management unit, determines purchase information that satisfies the selection criteria using the one or more user attribute values, the coupon information of the coupon selected from the two or more types of coupons, and the learning information, and determines the coupon that corresponds to the purchase information and is to be distributed to the user from the two or more types of coupons; a second coupon determination unit that, for each of the second proportion of the two or more users, determines the coupon to be distributed to the user without using the learning information; and a coupon output unit that outputs a coupon identifier that identifies the coupon to be distributed to each of the two or more users.

[0198] Furthermore, the computer running the above program may be a single computer or multiple computers. In other words, it may perform centralized processing or distributed processing.

[0199] Furthermore, in each of the above embodiments, each process may be implemented by centralized processing by a single device, or by distributed processing by multiple devices.

[0200] It goes without saying that the present invention is not limited to the embodiments described above, and various modifications are possible, all of which are also included within the scope of the present invention. [Industrial applicability]

[0201] As described above, the coupon processing device according to the present invention has the effect of increasing the effectiveness of the coupons being distributed, and is useful as a server for distributing coupons. [Explanation of Symbols]

[0202] A Information Systems 1. Coupon Processing Device 2 Terminal devices 11 Storage Unit 12 Receiver 13 Processing Unit 14. Transmitter 21 Terminal storage section 22 Terminal Reception Section 23 Terminal Processing Unit 24 Terminal transmission unit 25 Receiving part of the terminal 26 Terminal output section 111 User Management Department 112 Learning Information Management Department 113 Coupon Management Department 131 Learning Department 132 Score Acquisition Section 133 User Selection Section 134 First Coupon Decision Department 135 Second Coupon Decision Section 141 Coupon output section

Claims

1. A user management unit that stores user information for two or more users, including two or more users who are eligible to receive coupons, and each user has one or more user attribute values ​​for two or more users. A learning information management unit stores learning information based on two or more training data sets, which include coupon information related to coupons previously distributed to the user from among two or more types of coupons, purchase information related to the user's product purchases, and one or more user attribute values. For each user representing a first proportion of the two or more users, one or more user attribute values ​​are obtained from the user management unit, and using these one or more user attribute values, coupon information of a coupon selected from two or more types of coupons, and learning information, purchase information that satisfies the selection criteria is determined, and a first coupon determination unit determines from the two or more types of coupons the coupon that corresponds to the purchase information and is to be distributed to the user. For each of the two or more users, representing a second proportion, a second coupon determination unit determines which coupon to distribute to the user without using the learning information. A coupon processing device comprising: a coupon output unit that outputs a coupon identifier that identifies a coupon to be distributed to each of the two or more users mentioned above.

2. The aforementioned learning information is This is a machine learning model obtained by performing a machine learning process using two or more training data sets, with coupon information and one or more user attribute values ​​as explanatory variables and the purchase information as the dependent variable. The aforementioned first coupon determination unit is The coupon processing device according to claim 1, wherein for each user representing a first proportion of the two or more users, and for each coupon information of two or more types of coupons, machine learning prediction processing is performed using the coupon information, the one or more user attribute values, and the learning model to obtain purchase information, purchase information that satisfies a selection condition which is that the purchase information is the largest, and a coupon corresponding to the purchase information, which is to be distributed to the user, is determined from the two or more types of coupons.

3. The user information in the user management unit includes the user's coupon information and the user's purchase information. The system further comprises a learning unit that periodically obtains coupon information, purchase information, and one or more user attribute values ​​from the user management unit, constructs two or more training data sets, performs machine learning training using the two or more training data sets, and obtains the learning model. The coupon processing apparatus according to claim 2, wherein the learning model used by the first coupon determination unit is a learning model acquired by the learning unit.

4. The user information in the user management unit includes purchase information, For each of the two or more users, the system further comprises a score acquisition unit that acquires a user score using the purchase information of each user. The aforementioned first coupon determination unit is The coupons to be distributed to the first percentage of users among the two or more users whose scores meet the coupon distribution conditions are determined as follows: The aforementioned second coupon determination unit is, The coupon processing device according to any one of claims 1 to 3, which determines which coupons to distribute to the second percentage of users among two or more users who meet the coupon distribution conditions based on the score.

5. The user information in the user management unit includes purchase information from two or more points in time, The aforementioned score acquisition unit, The coupon processing device according to claim 4, which uses purchase information from two or more of the aforementioned time points to determine a class that indicates the user's purchasing tendency, selects a class of the user from among the two or more classes, and uses that class to obtain the score.

6. The system further comprises a coupon management unit that stores coupon information for each of two or more types of coupons, which includes a coupon identifier and the coupon amount. The aforementioned first coupon determination unit is A coupon processing device according to any one of claims 1 to 5, wherein for each user representing a first proportion of the two or more users, a coupon to be distributed to the user is determined sequentially, the amount corresponding to the determined coupon is obtained from the coupon management unit, the cumulative amount is calculated, and it is determined whether the cumulative amount has reached the budget amount, and if the cumulative amount has reached the budget amount, the process of determining the user's coupon is not performed further.

7. The user information in the user management unit includes the user's notification destination information. The aforementioned coupon output unit is A coupon processing device according to any one of claims 1 to 6, which distributes coupon information identified by a coupon identifier that identifies the coupon to be distributed to the notification destinations specified by the notification destination information of each of the two or more users.

8. A coupon processing method implemented by a user management unit which stores two or more user information having one or more user attribute values ​​for two or more users, including two or more users to whom coupons are to be distributed; a learning information management unit which stores learning information based on two or more training data having coupon information related to coupons previously distributed to users from two or more types of coupons, purchase information related to the user's purchase of products, and one or more user attribute values; a first coupon determination unit; a second coupon determination unit; and a coupon output unit, wherein The first coupon determination step involves the coupon determination unit obtaining one or more user attribute values ​​from the user management unit for each of the two or more users representing a first proportion of the users, determining purchase information that satisfies the selection criteria using the one or more user attribute values, the coupon information of the coupon selected from two or more types of coupons, and the learning information, and determining from the two or more types of coupons the coupon that corresponds to the purchase information and is to be distributed to the user, The second coupon determination unit performs a second coupon determination step in which, for each user representing a second proportion of the two or more users, it determines the coupon to be distributed to the user without using the learning information, A coupon processing method comprising: a coupon output step in which the coupon output unit outputs a coupon identifier that identifies a coupon to be distributed to each of the two or more users.

9. A computer that can access a user management unit that stores user information for two or more users, including two or more users to whom coupons are distributed, each having one or more user attribute values ​​for two or more users, and a learning information management unit that stores learning information based on two or more training data sets, each having coupon information related to coupons previously distributed to users from among two or more types of coupons, purchase information related to the user's product purchases, and one or more user attribute values, For each user representing a first proportion of the two or more users, one or more user attribute values ​​are obtained from the user management unit, and using these one or more user attribute values, coupon information of a coupon selected from two or more types of coupons, and learning information, purchase information that satisfies the selection criteria is determined, and a first coupon determination unit determines from the two or more types of coupons the coupon that corresponds to the purchase information and is to be distributed to the user. For each of the two or more users, representing a second proportion, a second coupon determination unit determines which coupon to distribute to the user without using the learning information. A program to function as a coupon output unit that outputs a coupon identifier to identify the coupons to be distributed to each of the two or more users mentioned above.

Images