WO2020230938A1 - Device for recommending receipt advertisements by using products purchased by customers in connection with pos terminal - Google Patents

Abstract

Disclosed is a device for recommending receipt advertisements by using products purchased by customers in connection with a POS terminal. A device for recommending receipt advertisements, according to the present invention, is configured to: when a request for transmitting product identifiers and advertisement images for products purchased by a user is received from a POS terminal installed in a supermarket, randomly extract the product identifiers so as to generate an eigenvector; calculate vector similarity between advertisement vectors for each of a plurality of advertisement images and the eigenvector; select an advertisement vector having the maximum value of the calculated vector similarity from among the advertisement vectors for each of the plurality of advertisement images; and then extract an advertisement image corresponding to the selected advertisement vector and transmit the advertisement image to the POS terminal, thereby enabling the advertisement image related to the product purchased by the user to be displayed on a receipt outputted through the POS terminal.

Description

Receipt advertisement recommendation device using products purchased by customers in connection with POS terminal

The present invention relates to a device capable of recommending advertisements related to products purchased by a customer in connection with a POS terminal, and an operating method thereof.

Usually, only information such as product name, product price, and product quantity for products purchased by the user is output on the receipt, and when the user receives a receipt, the user tends to immediately discard the receipt after checking the printed information.

In this way, if the user receives the receipt and immediately discards the receipt after checking the printed information, not only paper is wasted, but also it may be a little regretful in utilizing the receipt.

Accordingly, in order to prevent waste of paper and increase the utilization of receipts, a plan has been devised to output an advertisement in the additional area after allocating an additional area in addition to the existing area when printing a receipt.

In the past, when an advertisement is displayed in an additional area, there has been a tendency that only advertisements for products that are being promoted at a store or require new promotion are displayed regardless of the user's purchase intention.

However, when products that are not related to the user's purchase intention are output as advertisements, there is a problem that it is difficult for the user to purchase products again through advertisements, and thus it is difficult to lead to sales sales of affiliated stores.

Therefore, as a way to increase sales sales of affiliated stores, there is a need for a study on an advertisement recommendation technique capable of displaying advertisements for products related to products purchased by users at affiliated stores in an additional area of a receipt.

Receipt advertisement recommendation apparatus according to the present invention, when a product identifier for products purchased by a user and a request to transmit advertisement images are received from a POS terminal installed in a supermarket, the product identifiers are randomly extracted to generate an eigenvector, and a plurality of Calculates a vector similarity between the advertisement vector for each of the advertisement images and the eigenvector, selects an advertisement vector having the calculated vector similarity among the advertisement vectors for each of the plurality of advertisement images, and then selects the selected advertisement vector. By extracting an advertisement image corresponding to an advertisement vector and transmitting the advertisement image to the POS terminal, an advertisement image related to a product purchased by the user can be displayed on a receipt output through the POS terminal.

The apparatus for recommending receipt advertisements using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention includes a plurality of advertisement images and a predetermined n (n is 2) for each of the plurality of advertisement images. An advertisement information database in which advertisement vectors of the above natural number) are stored in correspondence with each other, product information in which a product identifier for each of a plurality of products and a preset unique number for each of the plurality of products correspond to each other and are stored Database, a receiving unit that receives product identifiers and advertisement image transmission requests for first products purchased by a user from a first POS (Point of Sale) terminal installed in a first supermarket, and the first POS terminal When a product identifier for the first products and a request to transmit the advertisement image are received, the n products stored in the product information database after extracting any n product identifiers from product identifiers for the first products An eigenvector generation unit that generates an n-dimensional eigenvector including an identification number for each of the identifiers as a component, and when the n-dimensional eigenvector is generated, each of the plurality of advertisement images stored in the advertisement information database A vector similarity calculation unit that calculates a vector similarity between the n-dimensional ad vector and the n-dimensional eigenvector for, based on the calculated vector similarity, among the n-dimensional ad vectors for each of the plurality of advertisement images, the An advertisement image extracting unit configured to select an n-dimensional advertisement vector having a maximum calculated vector similarity, and extract a first advertisement image stored in correspondence with the selected n-dimensional advertisement vector from the advertisement information database, and the second advertisement vector. When the 1 advertisement image is extracted, it includes an output requesting unit that transmits the first advertisement image and the advertisement image output request to the first force terminal.

Receipt advertisement recommendation apparatus according to the present invention, when a product identifier for products purchased by a user and a request to transmit advertisement images are received from a POS terminal installed in a supermarket, the product identifiers are randomly extracted to generate an eigenvector, and a plurality of Calculates a vector similarity between the advertisement vector for each of the advertisement images and the eigenvector, selects an advertisement vector having the calculated vector similarity among the advertisement vectors for each of the plurality of advertisement images, and then selects the selected advertisement vector. By extracting an advertisement image corresponding to an advertisement vector and transmitting the advertisement image to the POS terminal, it is possible to support the display of an advertisement image associated with a product purchased by the user on a receipt output through the POS terminal.

1 is a diagram showing the structure of a device for recommending a receipt advertisement using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention.

2 is a flowchart illustrating a method of operating a device for recommending a receipt advertisement using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. This description is not intended to limit the present invention to a specific embodiment, it is to be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention. While describing each drawing, similar reference numerals have been used for similar components, and unless otherwise defined, all terms used in the present specification including technical or scientific terms refer to common knowledge in the technical field to which the present invention belongs. It has the same meaning as commonly understood by someone who has it.

In this document, when a part "includes" a certain component, it means that other components may be further included rather than excluding other components unless otherwise stated. In addition, in various embodiments of the present invention, each component, functional blocks or means may be composed of one or more sub-components, and the electrical, electronic, and mechanical functions performed by each component are electronic. A circuit, an integrated circuit, and an application specific integrated circuit (ASIC) may be implemented with various known devices or mechanical elements, and may be implemented separately or two or more may be integrated into one.

On the other hand, the blocks of the attached block diagram and the steps in the flowchart are computer program instructions that are mounted on a processor or memory of equipment capable of processing data such as a general-purpose computer, a special-purpose computer, a portable notebook computer, and a network computer to perform specified functions. It can be interpreted as meaning. Since these computer program instructions can be stored in a memory provided in a computer device or in a memory readable by a computer, the functions described in the blocks in the block diagram or in the steps in the flowchart are produced as a product containing the instruction means to perform this. It could be. In addition, each block or each step may represent a module, segment, or part of code that contains one or more executable instructions for executing the specified logical function(s). In addition, it should be noted that in some alternative embodiments, functions mentioned in blocks or steps may be executed in a different order. For example, two blocks or steps shown in succession may be performed substantially simultaneously or may be performed in reverse order, and in some cases, some blocks or steps may be omitted.

1 is a diagram showing the structure of a device for recommending a receipt advertisement using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention.

Referring to FIG. 1, a receipt advertisement recommendation device 110 using products purchased by a customer in connection with a POS terminal according to the present invention includes an advertisement information database 111, a product information database 112, a receiving unit 113, and A vector generation unit 114, a vector similarity calculation unit 115, an advertisement image extraction unit 116, and an output request unit 117 are included.

The advertisement information database 111 stores a plurality of advertisement images and advertisement vectors of predetermined n (n is a natural number of 2 or more) dimensions corresponding to each other for each of the plurality of advertisement images.

For example, information may be stored in the advertisement information database 111 as shown in Table 1 below.

Multiple advertising images n-dimensional ad vector Advertising image 1 (a ₁ , a ₂ , ..., a _n ) ex)'n=5', (3, 9, 13, 15, 19) Advertising image 2 (b ₁ , b ₂ , ..., b _n ) ex)'n=5', (7, 9, 13, 15, 19) ... ...

The product information database 112 stores a product identifier for each of a plurality of products and a preset unique number for each of the plurality of products in correspondence with each other.

For example, information may be stored in the product information database 112 as shown in Table 2 below.

Multiple product identifiers Unique number Product identifier 1 3 Product identifier 2 5 Product identifier 3 7 Product identifier 4 9 Product identifier 5 11 Product identifier 6 13 Product identifier 7 15 Product identifier 8 17 Product identifier 9 19 Product identifier 10 21 ... ...

The receiving unit 113 receives a product identifier and an advertisement image transmission request for first products purchased by a user from a first POS (Point of Sale) terminal 130 installed in a first supermarket.

For example, as a user purchases first products in a first supermarket, when the first force terminal 130 installed in the first supermarket generates purchase information for the first products, the receiving unit 113 A product identifier for the first products purchased by the user at the first supermarket and an advertisement image transmission request may be received from the first force terminal 130.

At this time, the first force terminal 130 may store product identifiers and barcodes for each of the products currently sold in the first supermarket in correspondence with each other, and the first force terminal 130 may store the A barcode reader for recognizing a barcode of may be provided.

Here, the barcode (Bar-Code) is a small black bar consisting of several on a small label printed on the product when buying and calculating a product at a local supermarket, large discount store, or department store, and the barcode includes a 3-digit country code, 4 Digit manufacturer code, 5-digit product item code, and 1-digit check digit may be included.

In this case, as the user selects the first products from among the products currently sold in the first supermarket, the clerk displays the barcode attached to the first products in the first force terminal 130. When recognized by a barcode reader, the first force terminal 130 sends a request for transmitting the advertisement image along with a product identifier for the first products stored in correspondence with the barcodes of the first products. ).

When the eigen vector generator 114 receives product identifiers for the first products and a request to transmit the advertisement image from the first force terminal 130, any n items of product identifiers for the first products are received. After the identifiers are extracted, an n-dimensional eigenvector including a unique number for each of the n product identifiers stored in the product information database 112 is generated as a component.

Here, the identification number refers to a number previously assigned by the manager in order to efficiently process product management, and among the products currently sold in the first supermarket,'hometown dumplings', a product related to dumplings. For dumplings, like '3.1', a unique number '3.2' for'Sampo Dumpling', approximate values of '3' may be assigned as unique numbers.

For example, if'n=5' and it is assumed that the product identifier for the first products and the request to transmit the advertisement image from the first force terminal 130 are received by the receipt advertisement recommendation device 110, an eigenvector is generated. The unit 114 randomly extracts product identifiers '1, 3, 5, 7, 9'among the product identifiers for the first products, and then the product identifiers '1, 3'stored in the product information database 112 5, 7, 11, 15, 19) 5 dimensional eigenvectors including components with unique numbers for each of', 5, 7, 9'can be generated.

When the n-dimensional eigenvector is generated, the vector similarity calculating unit 115 determines between the n-dimensional advertisement vector and the n-dimensional eigenvector for each of the plurality of advertisement images stored in the advertisement information database 111. Compute vector similarity.

Here, a vector similarity between the n-dimensional advertisement vector and the n-dimensional eigenvector for each of the plurality of advertisement images may be calculated according to Equation 1 below.

Here, M is the vector similarity between the two vectors, S is the cosine similarity between the two vectors, D is the Euclidean distance between the two vectors, and the cosine similarity S and the two vectors between the two vectors The Euclidean distance D between may be calculated according to Equation 2 below and Equation 3 below.

Here, S is the cosine similarity between vectors A and B and has a value between -1 and 1, and the larger the value is, the more similar the vector is, A _i is the i-th component of vector A, and B _i is the _i of vector B Means the second ingredient.

In Equation 3, D denotes the Euclidean distance, and A _i and B _i denote i-th components included in the two vectors. Generally, as the Euclidean distance between two vectors is smaller, the two vectors can be regarded as similar vectors, and as the Euclidean distance between two vectors is larger, the two vectors can be considered to be dissimilar vectors.

For example, a five-dimensional advertisement vector'(3, 9, 13, 15, 19)' and the five-dimensional eigenvector'(3, 19) stored in correspondence with'ad image 1'on the advertisement information database 111 The cosine similarity S between 7, 11, 15, and 19)' and the Euclidean distance D may be calculated as in Equation 4 below and Equation 5 below.

As shown in Equation 4 and Equation 5 above, between the 5-dimensional advertisement vector for each of the plurality of advertisement images and the 5-dimensional eigenvector'(3, 7, 11, 15, 19)' When the vector similarity M is calculated by deriving the cosine similarity S and the Euclidean distance D, they are shown in Table 3 below.

Five-dimensional ad vector and five-dimensional eigenvector Calculated vector similarity between 5D ad vector and 5D eigenvector ( M = S + 1/D) (3, 9, 13, 15, 19), (3, 7, 11, 15, 19) 1.34976 (0.99626+0.3535) (7, 9, 13, 15, 19), (3, 7, 11, 15, 19) 1.1929 (0.9888+0.2041) ... ...

The advertisement image extraction unit 116 selects an n-dimensional advertisement vector having a maximum value of the calculated vector similarity among n-dimensional advertisement vectors for each of the plurality of advertisement images, based on the calculated vector similarity, and , Extracts a first advertisement image stored in correspondence with the selected n-dimensional advertisement vector from the advertisement information database 111.

For example, assuming that the maximum value of the calculated vector similarity is '1.34976', the advertisement image extracting unit 116 determines 5 for each of the plurality of advertisement images based on the calculated vector similarity in Table 3 above. Among the three-dimensional advertisement vectors, the five-dimensional advertisement vector'(3, 9, 13, 15, 19)' having the maximum value of the calculated vector similarity '1.34976' is selected, and the selected 5 from the advertisement information database 111 'Ad image 1'stored in correspondence with the dimensional advertisement vector'(3, 9, 13, 15, 19)' can be extracted.

When the first advertisement image is extracted, the output request unit 117 transmits the first advertisement image and an advertisement image output request to the first force terminal 130.

At this time, according to an embodiment of the present invention, when the first force terminal 130 receives the first advertisement image and the request to output the advertisement image from the receipt advertisement recommendation device 110, the first force terminal 130 1 After arranging in a preset first area of a receipt for displaying product price information, the receipt may be printed on paper.

For example, when the first advertisement image is referred to as'advertising image 1', when'advertising image 1', which is the first advertisement image, is extracted from the receipt advertisement recommendation device 110, the output request unit 117 receives the first advertisement image. The advertisement image'ad image 1'and the advertisement image output request may be transmitted to the first force terminal 130.

At this time, when the first force terminal 130 receives the first advertisement image'advertising image 1'and the advertisement image output request from the receipt advertisement recommendation device 110, the first advertisement image'advertising image 1' After arranging in a preset first area of a receipt for displaying price information of the first products, the receipt may be output as paper.

That is, when the receipt advertisement recommendation device 110 receives a product identifier and an advertisement image transmission request for the first products purchased by the user from the first force terminal 130 installed in the first supermarket, the first product An n-dimensional eigenvector including a unique number for each of the n product identifiers is generated by extracting arbitrary n product identifiers among the product identifiers for the products, and n-dimensional for each of a plurality of advertisement images. When the vector similarity between the ad vector of and the n-dimensional eigenvector is calculated, an n-dimensional ad vector having a maximum value of the calculated vector similarity among n-dimensional ad vectors for each of the plurality of advertisement images is selected. Thereafter, by extracting the first advertisement image stored in correspondence with the selected n-dimensional advertisement vector and transmitting the first advertisement image and advertisement image output request to the first force terminal 130, the user It is possible to support to easily check advertisement images of recommended products related to purchased products on receipt paper.

According to an embodiment of the present invention, in the advertisement information database 111, for each of the plurality of advertisement images, an access URL (Uniform Resource) of an advertisement website on which advertisement content corresponding to each of the plurality of advertisement images is posted. Locator) address may be additionally mapped and stored.

For example, the advertisement information database 111 may additionally store information as shown in Table 4 below.

Revenge Advertising images n-dimensional Advertising vector URL address of the ad website Advertising image 1 (a ₁ , a ₂ , ..., a _n ) ex)'n=5', (3, 9, 13, 15, 19) http://supermenz.com/nmms_12345.gif Advertising image 2 (b ₁ , b ₂ , ..., b _n ) ex)'n=5', (7, 9, 13, 15, 19) http://supermenz.com/nmms_12346.gif ... ... ...

In this case, the output request unit 117 may include an address extraction unit 118, a code generation unit 119, and a transmission unit 120.

When the first advertisement image is extracted, the address extraction unit 118 extracts a first access URL address stored in correspondence with the first advertisement image from the advertisement information database 111.

When the first access URL address is extracted, the code generator 119 generates a two-dimensional code into which the first access URL address is inserted.

Here, the two-dimensional code refers to a code capable of inserting information such as various characters on a two-dimensional image pattern, such as a quick response (QR) code.

In addition, the first force terminal 130 may be equipped with a two-dimensional code reader for recognizing a two-dimensional code into which the first access URL address is inserted.

When the 2D code is generated, the transmission unit 120 transmits the first advertisement image and the request for outputting the advertisement image to the first force terminal 130 and simultaneously outputs the 2D code and the 2D code. The request is transmitted to the first force terminal 130.

At this time, according to an embodiment of the present invention, the first force terminal 130 is configured for the first advertisement image, the advertisement image output request, the 2D code and the 2D code from the receipt advertisement recommendation device 110. When an output request is received, the first advertisement image is placed in a preset first area of a receipt for displaying price information of the first products, and the two-dimensional code is placed in a preset second area of the receipt. After that, you can print the receipt on paper.

For example, when the first advertisement image is referred to as'ad image 1'and the first access URL address is referred to as'http://supermenz.com/nmms_12345.gif', the first advertisement image is'ad image 1'. 'When extracted from the receipt advertisement recommendation device 110, the address extraction unit 118 corresponds to the first advertisement image,'advertising image 1', and the first access URL address stored in the advertisement information database 111 'Http://supermenz.com/nmms_12345.gif' can be extracted.

At this time, when the first access URL address'http://supermenz.com/nmms_12345.gif' is extracted from the receipt advertisement recommendation device 110, the code generation unit 119 is the first URL access address'http://supermenz.com/nmms_12345.gif'. It is possible to create a two-dimensional code with'://supermenz.com/nmms_12345.gif' inserted.

Then, the transmission unit 120 transmits the first advertisement image'ad image 1'and the advertisement image output request to the first force terminal 130, and at the same time, the first URL access address'http:/ /supermenz.com/nmms_12345.gif' is inserted into the 2D code and the first URL access address'http://supermenz.com/nmms_12345.gif' is inserted into the output request for the 2D code is inserted into the first force. It can be transmitted to the terminal 130.

In this way, from the receipt advertisement recommendation device 110, the first advertisement image “advertising image 1”, the advertisement image output request, and the first URL access address “http://supermenz.com/nmms_12345.gif” are inserted. When the first force terminal 130 receives a request for output of the generated 2D code and the 2D code with the first URL access address'http://supermenz.com/nmms_12345.gif' inserted, the first force The terminal 130 arranges the'advertising image 1', which is the first advertisement image, in a predetermined first area of the receipt for displaying price information of the first products, and the first URL access address'http:/' /supermenz.com/nmms_12345.gif' may be inserted into a second predetermined area of the receipt, and then the receipt may be printed on paper.

2 is a flowchart illustrating a method of operating a device for recommending a receipt advertisement using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention.

In step S210, an advertisement information database in which advertisement vectors of a plurality of advertisement images and a preset n (n is a natural number of 2 or more) dimensions for each of the plurality of advertisement images correspond to each other and is stored is maintained.

In step S220, a product information database in which a product identifier for each of a plurality of products and a preset unique number for each of the plurality of products correspond to each other and is stored is maintained.

In step S230, a product identifier for first products purchased by a user and a request to transmit an advertisement image are received from a first force terminal installed in a first supermarket.

In step S240, when product identifiers for the first products and a request to transmit the advertisement image are received from the first force terminal, n product identifiers of the product identifiers for the first products are extracted, An n-dimensional eigenvector including a unique number for each of the n product identifiers stored in the product information database is generated.

In step S250, when the n-dimensional eigenvector is generated, a vector similarity between the n-dimensional advertisement vector and the n-dimensional eigenvector for each of the plurality of advertisement images stored in the advertisement information database is calculated. do.

In step S260, based on the calculated vector similarity, an n-dimensional advertisement vector having a maximum value of the calculated vector similarity is selected from among n-dimensional advertisement vectors for each of the plurality of advertisement images, and the advertisement A first advertisement image stored in correspondence with the selected n-dimensional advertisement vector is extracted from the information database.

In step S270, when the first advertisement image is extracted, the first advertisement image and an advertisement image output request are transmitted to the first force terminal.

In this case, according to an embodiment of the present invention, when the first force terminal receives the first advertisement image and the request to output the advertisement image from the receipt advertisement recommendation device, the first advertisement image is converted to the price of the first products. After arranging in a predetermined first area of a receipt for displaying information, the receipt may be printed out on paper.

In addition, according to an embodiment of the present invention, for each of the plurality of advertisement images, an access URL address of an advertisement website on which advertisement content corresponding to each of the plurality of advertisement images is posted is added to the advertisement information database. It may be stored in correspondence with.

At this time, in step S270, when the first advertisement image is extracted, extracting a first access URL address stored in correspondence with the first advertisement image from the advertisement information database, and extracting the first access URL address Then, the step of generating a two-dimensional code in which the first access URL address is inserted, and when the two-dimensional code is generated, the first advertisement image and the request for outputting the advertisement image are transmitted to the first force terminal and the It may include transmitting a 2D code and an output request for the 2D code to the first force terminal.

At this time, according to an embodiment of the present invention, the first force terminal receives the first advertisement image, the advertisement image output request, the 2D code and the output request for the 2D code from the receipt advertisement recommendation device. Then, the first advertisement image is placed in a preset first area of a receipt for displaying price information of the first products, the two-dimensional code is placed in a preset second area of the receipt, and the receipt is You can print it out on paper.

In addition, according to an embodiment of the present invention, the calculation of the vector similarity between the n-dimensional advertisement vector and the n-dimensional eigenvector for each of the plurality of advertisement images may be performed according to Equation 1 above.

In the above, the operation method of the apparatus for recommending a receipt advertisement using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention has been described with reference to FIG. 2. Here, the operation method of the device for recommending receipt advertisement using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention is a receipt advertisement using products purchased by a customer in connection with the POS terminal described with reference to FIG. Since it may correspond to the configuration of the operation of the recommendation device 110, a more detailed description thereof will be omitted.

A method of operating a device for recommending receipt advertisements using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention may be implemented as a computer program stored in a storage medium for execution through combination with a computer.

In addition, the operation method of the apparatus for recommending receipt advertisements using products purchased by a customer in connection with a POS terminal according to an embodiment of the present invention is implemented in the form of program commands that can be executed through various computer means, Can be recorded. The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of the program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

As described above, in the present invention, specific matters such as specific components, etc., and limited embodiments and drawings have been described, but this is provided only to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. , If a person of ordinary skill in the field to which the present invention belongs, various modifications and variations are possible from these descriptions.

Therefore, the spirit of the present invention is limited to the described embodiments and should not be defined, and all things that are equivalent or equivalent to the claims as well as the claims to be described later fall within the scope of the spirit of the present invention. .

Claims (5)

An advertisement information database in which advertisement vectors of a plurality of advertisement images and a predetermined n (n is a natural number of 2 or more) dimensions corresponding to each other are stored for each of the plurality of advertisement images; A product information database in which a product identifier for each of a plurality of products and a preset unique number for each of the plurality of products are stored in correspondence with each other; A receiving unit configured to receive a product identifier for first products purchased by a user and a request to transmit an advertisement image from a first POS (Point of Sale) terminal installed in a first supermarket; When a product identifier for the first products and a request to transmit the advertisement image are received from the first POS terminal, n product identifiers of the product identifiers for the first products are extracted, and then the product information database An eigenvector generator for generating an n-dimensional eigenvector including a unique number for each of the n product identifiers stored as a component; A vector similarity calculator configured to calculate a vector similarity between the n-dimensional advertisement vector and the n-dimensional eigenvector for each of the plurality of advertisement images stored in the advertisement information database when the n-dimensional eigenvector is generated; Based on the calculated vector similarity, an n-dimensional advertisement vector having a maximum value of the calculated vector similarity is selected from among n-dimensional advertisement vectors for each of the plurality of advertisement images, and the selected advertisement vector is selected from the advertisement information database. an advertisement image extracting unit for extracting a first advertisement image stored in correspondence with an n-dimensional advertisement vector; And When the first advertisement image is extracted, an output request unit that transmits the first advertisement image and an advertisement image output request to the first force terminal Receipt advertisement recommendation device using products purchased by the customer in connection with the POS terminal including a. The method of claim 1, The first force terminal When the request for outputting the first advertisement image and the advertisement image is received from the receipt advertisement recommendation device, the first advertisement image is placed in a predetermined first area of a receipt for displaying price information of the first products, and the Receipt advertisement recommendation device using products purchased by customers in conjunction with a POS terminal that prints out receipts on paper. The method of claim 1, In the advertisement information database For each of the plurality of advertisement images, an access URL (Uniform Resource Locator) address of an advertisement website on which advertisement content corresponding to each of the plurality of advertisement images is posted is additionally corresponding and stored, The output request unit An address extractor configured to extract a first access URL address stored in correspondence with the first advertisement image from the advertisement information database when the first advertisement image is extracted; A code generator for generating a two-dimensional code into which the first access URL address is inserted when the first access URL address is extracted; And When the 2D code is generated, a request for outputting the first advertisement image and the advertisement image is transmitted to the first force terminal, and a request for outputting the 2D code and the 2D code is transmitted to the first force terminal. Transmission part to transmit Receipt advertisement recommendation device using products purchased by the customer in connection with the POS terminal including a. The method of claim 3, The first force terminal When an output request for the first advertisement image, the advertisement image output request, the 2D code and the 2D code is received from the receipt advertisement recommendation device, the first advertisement image is displayed as price information of the first products. Receipt using the products purchased by the customer by interlocking with a POS terminal that arranges the receipt in a preset first area of the receipt for use and places the two-dimensional code in a preset second area of the receipt and outputs the receipt on paper Ad recommendation device. The method of claim 1, The calculation of the vector similarity between the n-dimensional advertisement vector and the n-dimensional eigenvector for each of the plurality of advertisement images is a receipt using the products purchased by the customer by interlocking with a force terminal performed according to Equation 1 below. Ad recommendation device. [Equation 1]

Here, M is the vector similarity between two vectors, S is the cosine similarity between the two vectors, and D is the Euclidean distance between the two vectors.

Images