Recommendation Systems

Abstract

The present invention recommends suitable items in a restaurant or the like such as a conveyor belt sushi restaurant.　This recommendation system 10 comprises an acquisition unit 11 for acquiring order history information indicating the order history partway through a series of orders by a user and a determination unit 12 that uses the order history information acquired by the acquisition unit as a basis to predict a value relating to the total amount of the series of orders by the user and uses the order history information and the predicted value as a basis to determine an item to recommend to the user.

Description

[Technical field]

[0001] The present invention relates to a recommendation system for recommending products. [Background technology]

[0002] 2. Description of the Related Art Conventionally, in restaurants, it has been proposed to recommend a user's next order based on the user's order history (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-91115 A Summary of the Invention [Problem to be solved by the invention]

[0004] In restaurants such as conveyor belt sushi restaurants, the products offered to users include a variety of items such as sushi and desserts, and the users order them in sequence. As described in Patent Document 1, simply using the user's order history does not necessarily allow for appropriate product recommendations in such restaurants. For example, desserts cannot be recommended at the appropriate time.

[0005] One embodiment of the present invention has been made in consideration of the above, and aims to provide a recommendation system that can recommend appropriate products in restaurants such as conveyor belt sushi restaurants. [Means for solving the problem]

[0006] In order to achieve the above object, a recommendation system according to an embodiment of the present invention includes: an acquisition unit that acquires order history information indicating an intermediate order history in a series of orders of a user; and a determination unit that predicts a value related to a total amount of a series of orders of the user based on the order history information acquired by the acquisition unit, and determines products to be recommended to the user based on the order history information and the predicted value. The determination unit performs prediction and determination using a model that inputs order history information and a value related to the total amount of a series of orders of the user at a time point prior to the time point at which the recommended product is determined, and outputs information indicating products to be recommended to the user and a value related to the total amount of a series of orders of the user at the time point at which the recommended product is determined, and the model is generated by machine learning. .

[0007] In a recommendation system according to an embodiment of the present invention, a value related to the total amount of a series of orders by a user, such as a remaining bill amount or a total bill amount, is predicted, and products to be recommended to the user are determined based on the predicted value. Therefore, the recommendation system according to an embodiment of the present invention can recommend appropriate products in a restaurant such as a conveyor belt sushi restaurant, taking into account the value related to the total amount of a series of orders by a user. Effect of the Invention

[0008] According to one embodiment of the present invention, appropriate product recommendations can be made in restaurants, such as conveyor belt sushi restaurants. [Brief description of the drawings]

[0009] [Figure 1] 1 is a diagram showing a configuration of a recommendation system according to an embodiment of the present invention. [Diagram 2] FIG. 1 illustrates information used by a recommendation system. [Diagram 3] FIG. 1 is a diagram illustrating a model used by a recommendation system. [Figure 4] 1 is a sequence diagram showing processing executed in a recommendation system according to an embodiment of the present invention. [Diagram 5] FIG. 1 is a diagram illustrating a hardware configuration of a recommendation system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of a recommendation system according to the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same elements are given the same reference numerals, and duplicated description will be omitted.

[0011] FIG. 1 shows a recommendation system 10 according to this embodiment. The recommendation system 10 is a system (device) that recommends products to a user in a restaurant or the like. For example, the recommendation system 10 recommends a product to be ordered next to a user based on the order history of the user (customer) from the time the user enters the store to the time the user leaves the store in a store where the products are provided. Note that the store that is the target of the recommendation is usually a store (e.g., a conveyor belt sushi restaurant or an izakaya) where the user does not order all products at once, but orders products sequentially (at any time). In this embodiment, one session is defined as the period from the user's visit to the store to the time the user leaves the store. The recommendation system 10 recommends a product to be ordered next in one session to the user based on the user's order history up to the middle of the session. In this embodiment, a product recommendation in a conveyor belt sushi restaurant will be described as an example.

[0012] In this embodiment, the user to be recommended may be a group of multiple users who eat and drink at the same time. In the following description, even when simply referring to a user, it is assumed to include a group of multiple users. For example, in the above-mentioned store, an order panel used by a user is provided for each table. In this case, a user orders products using one order panel for each table. Therefore, even if the group is made up of multiple users, product recommendations are required not for individuals but for the group sharing the order panel.

[0013] The recommendation system 10 is configured by a computer such as a server device. The recommendation system 10 may be configured by a plurality of computers, i.e., a computer system.

[0014] Next, a description will be given of functions of the recommendation system 10 according to this embodiment. As shown in FIG.

[0015] The acquisition unit 11 is a functional unit that acquires order history information indicating an intermediate order history in a series of orders of a user. The acquisition unit 11 may acquire location information related to a location where a product is provided. The acquisition unit 11 may acquire time information related to a time when a product is provided. Each piece of information acquired by the acquisition unit 11 is information used to determine a product to be recommended to a user.

[0016] For example, a store where the above-mentioned products are provided has an order panel, and a user inputs the product (item) and quantity to be ordered into the order panel. The acquisition unit 11 acquires the information input into the order panel as order history information. An example of the acquired order history information is shown in FIG. 2(a). As shown in FIG. 2(a), the order history information is information in which the same session ID, ordering terminal ID, order time, product ID, and ordered quantity are associated with each other.

[0017] The same session ID is an ID (information) that identifies a session, i.e., an ID that is assigned to a transaction by the same user. The order terminal ID is an ID (information) that identifies the order panel used by the user. An ID is assigned to each order panel in a store in advance. In one session, the order panel used by the user is a specific order panel (e.g., an order panel installed at the user's seat). The same session ID is assigned, for example, to an order terminal ID when a previous transaction ends and a new order is made, i.e., when a new session begins. The order time is the time when an order was placed by the user, and is, for example, information such as the year, month, day, hour, and minute. The product ID and number ordered are an ID (information) that identifies the product ordered by the user and the number of the product ordered.

[0018] The recommendation system 10 prestores product ID management information as master data in a data server or the like. An example of the product ID management information is shown in FIG. 2(b). As shown in FIG. 2(b), the product ID management information is information in which information on the product name, price, and product ID is associated with each other. The product name and price are the product name and price of the product with the associated product ID. By referring to the product ID management information, the product name and price can be ascertained from the product ID.

[0019] The acquisition unit 11 acquires order history information every time a user places an order on the order panel. That is, the acquisition unit 11 acquires the information every time a user places an order on the order panel and one line of data in FIG. 2(a) is generated. Note that the order history information does not necessarily have to be the above information, and may be an intermediate order history in a series of orders by the user, that is, an intermediate order history in a session.

[0020] The acquisition unit 11 also acquires, as location information, information indicating the city, ward, town, or village to which the store where the product is provided belongs. The location information is acquired, for example, by reading out location information that has been input in advance and stored in the recommendation system 10 by an administrator of the recommendation system 10. Note that the location information does not necessarily have to be the above information, and may be information related to the place where the product is provided. For example, the location information may be information indicating the surrounding conditions of the store location.

[0021] Furthermore, the acquisition unit 11 acquires information on the time at which the product will be provided in one-hour units (for example, information such as 7 p.m., 8 p.m., etc.) as time information. As described above, the order history information includes information on the order time. Since the time at which the order was made (order time) and the time at which the product will be provided are generally the same, the acquisition unit 11 acquires information on the hour portion of the year, month, day, hour, and minute of the order time included in the order history information as time information. The time information acquired in this manner generally indicates the current time at which the recommendation is made. Note that the time information does not necessarily have to be the above information, and may be information related to the time at which the product will be provided.

[0022] Moreover, the acquisition of each piece of information does not necessarily have to be performed as described above, and may be performed by any method capable of acquiring the information. For example, the order history information may be acquired from a source other than the order panel. The acquisition unit 11 outputs each piece of acquired information to the determination unit 12. The order history information and time information are output from the acquisition unit 11 to the determination unit 12 each time the order history information is acquired. The location information may be output from the acquisition unit 11 to the determination unit 12 only once per session, for example, together with the output of the first order history information. Alternatively, the acquisition unit 11 may store the acquired information in a data server or the like of the recommendation system 10 so that the information can be referenced by the determination unit 12.

[0023] The determination unit 12 is a functional unit that predicts a value related to the total amount of a series of orders of the user based on the order history information acquired by the acquisition unit 11, and determines a product to be recommended to the user based on the order history information and the predicted value. The determination unit 12 may make the prediction and determination using a model that inputs the order history information and a value related to the total amount of a series of orders of the user at a time point prior to the time point at which the product to be recommended is determined, and outputs information indicating the product to be recommended to the user and a value related to the total amount of a series of orders of the user at the time point at which the product to be recommended is determined. The model may be generated by machine learning. The determination unit 12 may make the prediction and determination based on the location information acquired by the acquisition unit 11 as well. The determination unit 12 may make the prediction and determination based on the time information acquired by the acquisition unit 11 as well. Specifically, the determination unit 12 determines the product to be recommended to the user as follows.

[0024] The determination unit 12 recommends the next product to be ordered by the user for the session based on the order history information for each session. The determination unit 12 stores the above model in advance and uses the model to determine the product to be recommended. FIG. 3 shows a model (algorithm) used to determine the product to be recommended. As shown in FIG. 3, the model inputs information on the ordered product, the number of items ordered, location information, time information, the billing amount up to the present, and the remaining billing amount for the previous hour, and outputs information on the product to be recommended to the user (the next product) and the remaining billing amount (at the time of determining the product to be recommended). The model includes, for example, a neural network generated by machine learning. The neural network may be multi-layered, that is, generated by performing deep learning.

[0025] The remaining billing amount is a value related to the total amount of a series of orders made by the user in this embodiment. The remaining billing amount is the amount the user will spend from that point in the session. The remaining billing amount is the amount the user will spend in the entire session minus the amount the user has spent up to that point.

[0026] The determination unit 12 inputs each piece of information from the acquisition unit 11. Every time new order history information is input, that is, every time the user places a new order, the determination unit 12 predicts (outputs) information on the next product and information on the remaining billing amount using the above model.

[0027] The information on the ordered items input to the model is, for example, the item ID included in the new order history information. The information on the number of ordered items input to the model is, for example, the number of ordered items included in the new order history information. The location information input to the model is, for example, the location information input from the acquisition unit 11. The time information input to the model is, for example, the time information input from the acquisition unit 11. The location information and time information input to the model may be ID-coded.

[0028] The information on the billing amount up to the present time input to the model is, for example, calculated by the determination unit 12 from the order history information up to that time related to the session. The determination unit 12 refers to the price of each product in the product ID management information and calculates the billing amount up to the present time from the product IDs and order quantities of all the order history information related to the session. The information on the remaining billing amount for the previous time input to the model is, for example, information on the remaining billing amount output from the model using the order history information immediately before the new order history information. If the new order history information is the first order history information related to the session, there is no information on the remaining billing amount output from the model using the order history information immediately before, so information on the remaining billing amount for the previous time that is set and stored in advance is used. For example, the amount can be the average billing amount of all users in the past, or the like.

[0029] The information on the next product output from the model is, for example, a numerical value indicating the degree to which each product is recommended (a vector of the dimension of the number of products). For example, as shown in FIG. 3, the numerical value is in the range of 0 to 1, and the larger the numerical value, the higher the degree to which the product is recommended. This numerical value can also be regarded as the probability that the user will order the product. The information on the remaining bill (at the time of deciding which product to recommend) output from the model is, for example, a numerical value indicating the total amount, as shown in FIG. 3.

[0030] The determination unit 12 may characterize the information input to the model in the calculation using the model. The information to be characterized is, for example, information on the ordered product, location information, and time information as shown in FIG. 3. The characterization is performed within the model in the same manner as in the past. The characterization will be described by taking the characterization of the information on the ordered product as an example. First, the determination unit 12 inputs a product ID as information on the ordered product, and converts it into an N-dimensional vector V1 previously associated with the product ID (N is preset). Next, the determination unit 12 converts the N-dimensional vector V1 into a feature C1, which is a preset number of numerical values ​​(i.e., a vector of a preset number of dimensions). Each numerical value of the feature C1 corresponds to the numerical value of a neuron in a neural network. The conversion from the vector V1 to the feature C1 is performed based on a numerical value (weight) set for the connection between numerical values, as in normal calculations in a neural network. The vector V1 associated with the product ID and the weight for converting the vector V1 to the feature C1 are generated by machine learning. Note that the vector V1 at the start of the machine learning is composed of random numerical values. The location information and time information are also characterized in the same manner as above to be used as feature quantities.

[0031] The determination unit 12 inputs the feature C1 of the ordered item to an RNN (Recurrent Neural Network) included in the model. The determination unit 12 also inputs the feature C2 (feature layer of the previous time) after processing by the RNN, which is obtained by calculation of the previous order history information of the same session, to the RNN. The determination unit 12 adds these features C1 and C2 to generate feature C3. The feature C3 generated here is input to the RNN as the above feature C2 when calculating the next order history information of the same session. If the new order history information is the first order history information related to the session, since no calculation has been performed on the previous order history information, a feature that has been set and stored in advance (for example, a feature with all element values ​​set to 0) is used as feature C2. By using the RNN in this way, it is possible to output information on the next item and the remaining billing amount, taking into account all items ordered up to that point, including the order.

[0032] The determination unit 12 generates combined layer information from the information input to the model. The combined layer information is a simple horizontal combination of each piece of information (arranged side by side). For information on ordered items, the feature C3 generated by the RNN is used as the combined layer information. For location information and time information, the characterized feature C1 is used as the combined layer information. For information on the number of items ordered, the billing amount up to the present, and the remaining billing amount for the previous hour, the input information is used as is as the combined layer information.

[0033] The determination unit 12 calculates the information to be output from the information of the combined layers, i.e., the information of the next product and the remaining billing amount. This calculation is performed based on the numerical values ​​(weights) set for the connections between numerical values, similar to normal calculations in neural networks. The weights are generated by machine learning.

[0034] The above model can be generated by machine learning, for example, by using order history information of a session where a transaction has already been completed as learning data (teaching data). In this case, for the information on the next product corresponding to the output, the value of the product actually ordered after the new order history information corresponding to the input is set to 1, and the values ​​of other products are set to 0. Furthermore, the information on the remaining billing amount can be calculated by subtracting the billing amount up to the new order history information corresponding to the input from the billing amount for the entire session. Note that the model may be generated by the recommendation system 10, or by a system (device) other than the recommendation system 10.

[0035] The above model used in the recommendation system 10 is expected to be used as a program module that is a part of artificial intelligence software. The model is used in a computer having a CPU and a memory. Specifically, the CPU of the computer operates in accordance with instructions from the model stored in the memory to input information to the input layer of the neural network, perform calculations based on the learned weighting coefficients in the neural network, and output results from the output layer of the neural network.

[0036] The determination unit 12 determines a product to be recommended to the user based on the calculated information of the next product. For example, the determination unit 12 determines the products with the top N product values ​​indicated in the information of the next product as the products to be recommended to the user. Here, N is a preset and stored value. The determination unit 12 outputs information of the determined product. For example, the determination unit 12 transmits information of the determined product to an order panel used by the user to display the product. At this time, the determination unit 12 may display the product name and price by referring to product ID management information. Note that the determination of the product to be recommended to the user does not necessarily have to be performed as described above, but may be performed based on the calculation result. In addition, the output of information may be performed by a method other than the above.

[0037] The determination unit 12 uses the calculated remaining bill amount information for the recommendation for the next order history information as described above, without outputting the information in a format that the user can refer to. The above is the function of the recommendation system 10 according to this embodiment.

[0038] Next, the process executed by the recommendation system 10 according to this embodiment (the operation method performed by the recommendation system 10) will be described with reference to the flowchart of FIG. 4. This process is performed on the above-mentioned session basis when a user uses a store. In this process, the acquisition unit 11 acquires information used to determine products to be recommended to the user (S01). The acquired information includes order history information, location information, and time information. The order history information is information for each order placed by a user, and the following process is performed each time order history information corresponding to one order is acquired.

[0039] Next, the determination unit 12 uses a pre-stored model to calculate (predict) information on the next product and information on the remaining billing amount from the information acquired by the acquisition unit 11 (S02). The determination unit 12 also determines a product to be recommended to the user based on the information on the next product. Next, the determination unit 12 recommends the determined product (S03).

[0040] Next, when the user places a new order, the acquisition unit 11 acquires new order history information and time information related to the order (S01). Note that location information only needs to be acquired once in a session. Next, the determination unit 12 calculates (predicts) information on the next product and the remaining billing amount from the new order history information, and makes a recommendation based on this (S02, S03). The above processes of S01 to S03 are repeated every time the user places a new order. The above is the process executed by the recommendation system 10 according to this embodiment.

[0041] In this embodiment, the remaining bill is predicted, and the product recommended to the user is determined based on the remaining bill. The remaining bill reflects the user's characteristics related to the product the user will order next. Therefore, according to this embodiment, appropriate products can be recommended in consideration of the user's remaining bill in restaurants such as conveyor belt sushi restaurants. For example, if the remaining bill is large, it is assumed that the user will still order more, and sushi products can be recommended, whereas if the remaining bill is small, it is assumed that the meal is nearing the end, and dessert can be recommended. In this way, products can be recommended at the appropriate time.

[0042] The value relating to the total amount of a series of orders of a user is not limited to the remaining bill amount used in this embodiment, but may be anything that can grasp the orders of the entire session. For example, the total bill amount of the entire session may be used as the value relating to the total amount of a series of orders of a user.

[0043] Also, as in this embodiment, a recommendation may be made using a model generated by machine learning that predicts both the information on the next product and the information on the remaining billing amount. That is, a recommendation may be made using multitask learning, which is a deep learning method that solves multiple problems by sharing knowledge. By sharing the intermediate layers of multiple predictions as described above, useful features common to the problems can be learned. Furthermore, the prediction accuracy can be improved by complementing each other between tasks. As a result, according to this embodiment, a more appropriate product can be recommended. However, the information on the next product and the information on the remaining billing amount may be calculated by different models. Alternatively, the calculation may be performed without using a model generated by machine learning.

[0044] In addition, recommendations may be made using location information and time information as in the present embodiment. With this configuration, it is possible to recommend an appropriate product according to the location and time at which the product is provided. However, the above information does not necessarily need to be used for the recommendation. Also, recommendations may be made using information other than the above.

[0045] In the above embodiment, the recommendation of products at a conveyor belt sushi restaurant has been described as an example, but other recommendations may also be targeted. The target of recommendations by the recommendation system 10 is not limited to products at restaurants such as conveyor belt sushi restaurants and izakayas, but may be products that are sequentially ordered by users. For example, it may be a recommendation of the next song to be sung by the user at a karaoke store. In this case, the recommended product is a song (offered by the user). In this case, the value relating to the total amount of a series of orders by the user may be the number of remaining songs or the total number of songs in the session.

[0046] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. The method of realizing each functional block is not particularly limited. That is, each functional block may be realized by using one device that is physically or logically combined, or may be realized by using two or more devices that are physically or logically separated and directly or indirectly connected (for example, by wire, wirelessly, etc.). The functional blocks may be realized by combining the one device or the multiple devices with software.

[0047] Functions include, but are not limited to, judgment, determination, judgment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs the function of transmission is called a transmitting unit or a transmitter. In either case, as described above, there is no particular limitation on the method of realization.

[0048] For example, the recommendation system 10 according to an embodiment of the present disclosure may function as a computer that performs the information processing of the present disclosure. Fig. 5 is a diagram showing an example of a hardware configuration of the recommendation system 10 according to an embodiment of the present disclosure. The above-mentioned recommendation system 10 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

[0049] In the following description, the term "apparatus" may be replaced with a circuit, a device, a unit, etc. The hardware configuration of the recommendation system 10 may be configured to include one or more of the apparatuses shown in the figure, or may be configured to exclude some of the apparatuses.

[0050] Each function in the recommendation system 10 is realized by loading a specific software (program) onto hardware such as a processor 1001 and a memory 1002, causing the processor 1001 to perform calculations, control communications via a communication device 1004, and control at least one of the reading and writing of data in the memory 1002 and the storage 1003.

[0051] The processor 1001, for example, operates an operating system to control the entire computer. The processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, each function in the above-mentioned recommendation system 10 may be realized by the processor 1001.

[0052] Furthermore, the processor 1001 reads out a program (program code), a software module, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002, and executes various processes according to the read out programs. As the program, a program that causes a computer to execute at least a part of the operations described in the above-mentioned embodiment is used. For example, each function in the recommendation system 10 may be realized by a control program stored in the memory 1002 and operated in the processor 1001. Although the above-mentioned various processes have been described as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The program may be transmitted from a network via a telecommunication line.

[0053] The memory 1002 is a computer-readable recording medium, and may be composed of at least one of, for example, a read only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store a program (program code), software modules, etc. that are executable to perform information processing according to an embodiment of the present disclosure.

[0054] The storage 1003 is a computer-readable recording medium, and may be composed of at least one of an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, and the like. The storage 1003 may be called an auxiliary storage device. The storage medium provided in the recommendation system 10 may be, for example, a database including at least one of the memory 1002 and the storage 1003, a server, or other suitable medium.

[0055] The communication device 1004 is hardware (transmission / reception device) for performing communication between computers via at least one of a wired network and a wireless network, and is also called, for example, a network device, a network controller, a network card, or a communication module.

[0056] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that performs output to the outside. The input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0057] In addition, each device such as the processor 1001 and the memory 1002 is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0058] Furthermore, the recommendation system 10 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0059] The order of the steps, sequences, flow charts, etc. of each aspect / embodiment described in this disclosure may be changed unless inconsistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0060] The input and output information may be stored in a specific location (e.g., memory) or may be managed using a management table. The input and output information may be overwritten, updated, or added to. The output information may be deleted. The input information may be transmitted to another device.

[0061] The determination may be based on a value represented by a single bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0062] Each aspect / embodiment described in this disclosure may be used alone, in combination, or switched according to execution. In addition, notification of specific information (e.g., notification that "X is the case") is not limited to being done explicitly, but may be done implicitly (e.g., not notifying the specific information).

[0063] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described in the present disclosure. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0064] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0065] Additionally, software, instructions, information, etc. may be transmitted or received over a transmission medium. For example, if the software is transmitted from a website, server, or other remote source using wired and / or wireless technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave, etc.), then these wired and / or wireless technologies are included within the definition of transmission media.

[0066] As used in this disclosure, the terms "system" and "network" are used interchangeably.

[0067] Furthermore, the information, parameters, etc. described in this disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information.

[0068] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, search, inquiry (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in a memory), and the like. In addition, "judgment" and "decision" can include considering resolving, selecting, choosing, establishing, comparing, etc., to be a "judgment" or "decision." In other words, "judgment" and "decision" can include considering some action to be a "judgment" or "decision." In addition, "judgment" can be interpreted as "assuming," "expecting," "considering," etc.

[0069] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using at least one of one or more wires, cables, and printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0070] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0071] Any reference to an element using a designation such as "first," "second," etc., used in this disclosure does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must precede the second element in some way.

[0072] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Further, when used in this disclosure, the term "or" is not intended to be an exclusive or.

[0073] In this disclosure, where articles have been added due to translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0074] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different." [Explanation of symbols]

[0075] 10...recommendation system, 11...acquisition unit, 12...decision unit, 1001...processor, 1002...memory, 1003...storage, 1004...communication device, 1005...input device, 1006...output device, 1007...bus.

Claims

1. an acquisition unit that acquires order history information indicating an intermediate order history of a series of orders made by a user; a determination unit that predicts a value related to a total amount of a series of orders of a user based on the order history information acquired by the acquisition unit, and determines a product to be recommended to the user based on the order history information and the predicted value; Equipped with the determination unit inputs order history information and a value related to a total amount of a series of orders of the user at a time point prior to the time point at which the recommended product is determined, and performs prediction and determination using a model that outputs information indicating products to be recommended to the user and a value related to a total amount of a series of orders of the user at the time point at which the recommended product is determined; The model is a recommendation system generated by machine learning.

2. The recommendation system according to claim 1 , wherein the determination unit predicts a remaining bill amount or a total bill amount as a value related to the total amount of a series of orders of the user.

3. The acquisition unit acquires location information relating to a location where the product is provided, The recommendation system according to claim 1 or 2, wherein the determination unit makes predictions and decisions based also on the location information acquired by the acquisition unit.

4. The acquisition unit acquires time information relating to a time when the product will be provided, The recommendation system according to any one of claims 1 to 3, wherein the determination unit makes predictions and decisions based also on time information acquired by the acquisition unit.

Images