Purchase data output system
The system enhances receipt digitization by using machine learning to estimate and output comprehensive purchase data, including official product names and additional details, addressing the limitation of abbreviated product names on receipts.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- CASH B DATA CO LTD
- Filing Date
- 2022-04-12
- Publication Date
- 2026-05-11
AI Technical Summary
Conventional systems that digitize information on receipts often limit the product name to abbreviations due to paper size constraints, resulting in less valuable data for users.
A purchase data output system utilizing machine learning to analyze receipt images, perform edge detection and plane transformation, and estimate product and non-product information, then create comprehensive purchase data by adding official product names and additional details from databases.
Provides users with highly valuable purchase data by including official product names and additional information such as store details and product attributes, enhancing the utility of digitized receipt data.
Smart Images
Figure 0007856294000001 
Figure 0007856294000002 
Figure 0007856294000003
Abstract
Description
Technical Field
[0001] The present invention relates to a purchase data output system that outputs purchase data of products.
Background Art
[0002] Conventionally, a technique for digitizing information described on a receipt has been proposed. For example, a system for analyzing information printed on a receipt has been proposed (see Patent Document 1). In the conventional system, the information included in the information (text data) printed on the receipt is decomposed into words, and based on the positional relationship of the words, a word or a group of words consisting of a plurality of words is classified into items.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] In the conventional system, although the information printed on the receipt can be digitized, the digitized information is the information printed on the receipt itself. The information printed on the receipt includes, for example, the name of the product.
[0005] However, since the number of characters that can be described on the receipt is limited by the paper size, the name of the product printed on the receipt is often a part of the official name of the product or an abbreviation. That is, the name of the product printed on the receipt is not necessarily the official name of the product. Therefore, even if the information printed on the receipt itself can be digitized, it is difficult to say that the data is highly valuable to the user who uses the data.
[0006] This invention has been made in view of the above-mentioned problems, and aims to provide a purchase data output system that can provide users with highly valuable information. [Means for solving the problem]
[0007] The present invention provides a purchase data output system for outputting product purchase data, comprising: a first machine learning unit that analyzes the relationship between a receipt image and product information and non-product information described in the receipt image using machine learning; a second machine learning unit that analyzes the relationship between product information described in the receipt image and the official name information of the product registered in a predetermined database using machine learning; an image processing unit that performs edge detection and plane transformation on a receipt image acquired from a user terminal, followed by either distortion correction or noise reduction; and, based on the relationship analyzed by the first machine learning unit, takes the processed receipt image as input and estimates and outputs product information and non-product information described in the receipt image. The purchase data output system comprises a first estimation unit and a second estimation unit that, based on the relationships analyzed by the second machine learning unit, takes the product information output from the first estimation unit as input and estimates and outputs the official name information of the product, wherein the information other than the product includes any of the following: the company name or corporate brand name of the store that sells the product, the name of the store that sells the product, the store's telephone number or store address, the purchase date or time the product was purchased, the purchase quantity or unit price or subtotal or total or tax of the product, or the payment method used when the product was purchased, and the purchase data output system comprises a purchase data creation unit that creates purchase data of the product by adding the information other than the product output from the first estimation unit to the official name information of the product output from the second estimation unit.
[0008] In this configuration, information about the products and other information listed on the receipt image is estimated and output from the receipt image obtained from the user's terminal. From the estimated product information, the official name of the product is estimated and output. Then, by adding information other than the product (for example, the store's company name or brand name, store name or phone number or store address, purchase date or time, purchase quantity or unit price or subtotal or total or tax, payment method, etc.) to the official name of the product, product purchase data is created and output. In this way, the information listed on the receipt can be converted into product purchase data. In this case, since the product purchase data includes not only the official name of the product but also information other than the product, it can provide users with highly valuable information.
[0009] Furthermore, the purchase data output system of the present invention includes a first information acquisition unit that obtains information on the correct store name, correct store telephone number, or correct store address of a store that sells the product by performing an information search based on at least one piece of information from the store name, store telephone number, or store address of a store that sells the product output from the first estimation unit, and the purchase data creation unit may create purchase data for the product by adding information on the correct store name, correct store telephone number, or correct store address of a store that sells the product to the information on the official name of the product output from the second estimation unit.
[0010] With this configuration, by performing an information search (for example, an internet search or a search on a store information service) based on at least one piece of information—the store name, phone number, or address—estimated from the receipt image, the correct store name, phone number, or address of the store selling the product is obtained and added to the product purchase data as information other than the product itself. This makes it possible to provide users with highly valuable information, such as the correct store name, phone number, or address of the store selling the product.
[0011] Furthermore, the purchase data output system of the present invention includes a second information acquisition unit that, based on the information of the official name of the product output from the second estimation unit, refers to the predetermined database to acquire the manufacturer name or brand name of the manufacturer that produces the product, or the category or JAN code information of the product, and the purchase data creation unit may create purchase data for the product by adding the manufacturer name or brand name of the manufacturer, or the category or JAN code information of the product, to the information of the official name of the product output from the second estimation unit.
[0012] This configuration allows the system to retrieve information such as the official product name, manufacturer's name or brand name, or product category or JAN code by referencing the database. This information is then added to the product purchase data as non-product information. This provides users with valuable information such as the manufacturer's name or brand name, or product category or JAN code.
[0013] The present invention is a method performed in a purchase data output system that outputs product purchase data, the method comprising: a first machine learning step of analyzing the relationship between a receipt image and product information and non-product information described in the receipt image using machine learning; a second machine learning step of analyzing the relationship between product information described in the receipt image and information about the official name of the product registered in a predetermined database using machine learning; an image processing step of performing edge detection and plane transformation on a receipt image acquired from a user terminal, followed by image processing of either distortion correction or noise reduction; and, based on the relationship analyzed in the first machine learning step, using the processed receipt image as input, inferring product information and non-product information described in the receipt image. The method comprises a first estimation step which outputs a fixed value, and a second estimation step which, based on the relationship analyzed in the second machine learning step, takes the product information output in the first estimation step as input and estimates and outputs the official name information of the product, wherein the information other than the product includes any of the following: the company name or corporate brand name of the store that sells the product, the name of the store that sells the product, the store's telephone number or store address, the purchase date or time the product was purchased, the purchase quantity or unit price or subtotal or total or tax of the product, or the payment method used when the product was purchased, and the method also includes a purchase data creation step which adds the information other than the product output in the first estimation step to the official name information of the product output in the second estimation step to create purchase data for the product.
[0014] This method, similar to the system described above, estimates and outputs product information and non-product information from a receipt image obtained from the user's terminal. From the estimated product information, the official name of the product is estimated and output. Then, by adding non-product information (for example, the store's company name or brand name, store name or phone number or store address, purchase date or time, purchase quantity or unit price or subtotal or total or tax, payment method, etc.) to the official name of the product, purchase data is created and output. In this way, the information on the receipt can be converted into product purchase data. In this case, since the product purchase data includes not only the official name of the product but also non-product information, it can provide users with highly valuable information. [Effects of the Invention]
[0015] According to the present invention, it is possible to provide users with highly valuable information. [Brief explanation of the drawing]
[0016] [Figure 1] This is a block diagram showing the configuration of the purchase data output system in this embodiment. [Figure 2] This block diagram shows an example of the configuration of the image processing unit in this embodiment. [Figure 3] This diagram shows an example of product information and other information listed on a receipt. [Figure 4] This figure shows an example of product data registered in the database. [Figure 5] This is a flowchart illustrating the operation of the purchase data output system in this embodiment. [Modes for carrying out the invention]
[0017] Hereinafter, the purchase data output system according to the embodiment of the present invention will be described with reference to the drawings. In this embodiment, a case of a purchase data output system used in a system for digitizing receipt information or the like will be exemplified.
[0018] The configuration of the purchase data output system according to the embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the purchase data output system according to the present embodiment. As shown in FIG. 1, the purchase data output system 1 is connected via a network N such as the Internet to a user terminal 2 used by a general user who purchases goods, and a database 3 in which product data (described later) regarding the goods is registered. The purchase data output system 1 is configured by, for example, a server device or the like. Further, the purchase data output system 1 has functions for outputting purchase data of goods, and these functions can be realized by a program installed in a memory or the like of the purchase data output system 1 (server device).
[0019] As shown in FIG. 1, the purchase data output system 1 includes an input / output unit 4, a storage unit 5, a first system 6, a second system 7, and a third system 8. The input / output unit 4 is an input / output interface for inputting and outputting various data for outputting purchase data of goods. Here, an example in which the input interface and the output interface are integrally configured will be described, but the input interface and the output interface may be configured separately. The storage unit 5 is configured by a memory or the like, and stores various data for outputting purchase data of goods.
[0020] Also, as shown in FIG. 1, the first system 6 includes a first machine learning unit 9, an image processing unit 10, and a first estimation unit 11, the second system 7 includes a second machine learning unit 12 and a second estimation unit 13, and the third system 8 includes a first information acquisition unit 14, a second information acquisition unit 15, and a purchase data creation unit 16. Further, as shown in FIG. 2, the image processing unit 10 includes an edge detection unit 17, a planar conversion unit 18, a binarization unit 19, a smoothing unit 20, and a noise removal unit 21.
[0021] The first machine learning unit 9 analyzes, by machine learning, the relationship between the receipt image and the information of products and non-product information described in the receipt image. Any method such as deep learning using a neural network can be used for this machine learning.
[0022] For example, in the case of a neural network, the receipt image is input to the input layer, and the information of products and non-product information described in the receipt image is output from the output layer. Then, the weighting coefficients between the neurons of the neural network are optimized by supervised learning using the analysis data (teacher data) in which the data (receipt image) input to the input layer and the data (product information and non-product information) output from the output layer are associated.
[0023] The receipt image to be determined (the receipt image acquired from the user terminal 2) is input to the input / output unit 4. In the image processing unit 10, image processing (preprocessing) is performed on the input receipt image (the receipt image to be determined) to make it easier to estimate product information and non-product information from the receipt image. First, in the edge detection unit 17, edge detection is performed on the input receipt image. For example, by edge detection, an image of a rectangular image area including the image area where the receipt is shown is extracted from the input receipt image. Known techniques can be used for edge detection.
[0024] Next, in the image processing unit 10, in the planar conversion unit 18, a process (planar conversion process) is performed to convert the image of the rectangular image area (the image of the receipt seen from the front, the image with perspective) extracted by edge detection into a planar image (the image without perspective) of the receipt seen from the front. Known techniques can also be used for the planar conversion process.
[0025] Next, in the image processing unit 10, the binarization unit 19 performs a process (binarization process) to convert the planar image (color image) of the receipt into a black and white binarized image. Then, the smoothing unit 20 performs a process (smoothing process) to correct the distortion of the binarized image of the receipt, and the noise reduction unit 21 performs a process (noise reduction process) to remove noise (such as dots in the image) contained in the binarized image of the receipt. Known techniques can also be used for the binarization process, smoothing process, and noise reduction process.
[0026] The first estimation unit 11 takes a processed receipt image as input and estimates and outputs information about the products and other information described in the receipt image, based on the relationships analyzed by the first machine learning unit 9. For example, in the neural network described above, the processed receipt image is input to the input layer, and the information about the products and other information described in the receipt image is estimated and output from the output layer, thereby estimating the information about the products and other information described in the receipt.
[0027] Figure 3 shows an example of product information and non-product information listed on a receipt. As shown in Figure 3, the receipt includes product information such as the product name, for example, "Brand M Product Name A". In addition, the receipt includes non-product information such as the name of the store selling the product, for example, "K Corporation", the store's brand name, for example, "Brand K T Town Store", the store's phone number, for example, "03-XXXX-XXXX", the store's address, for example, "1 T Town, K Ward, Tokyo", the purchase date, for example, "April 4, 2022", the purchase time, for example, "15:00", the quantity of items purchased, for example, "2 items", the unit price, for example, "100 yen", the subtotal, for example, "200 yen", the total, for example, "220 yen", the tax, for example, "20 yen", and the payment method, for example, "cash".
[0028] The second machine learning unit 12 analyzes the relationship between the product information listed on the receipt image and the official name information of that product registered in database 3 using machine learning. This machine learning can utilize any method, such as deep learning using neural networks.
[0029] For example, in the case of a neural network, the information of the products listed on the receipt image is input to the input layer, and the information of the official name of the product registered in database 3 is output from the output layer. Then, the weighting coefficients between neurons in the neural network are optimized through supervised learning using analysis data (training data) that links the data input to the input layer (information of the products listed on the receipt image) and the data output from the output layer (information of the official name of the product registered in database 3).
[0030] Figure 4 shows an example of product data registered in database 3. As shown in Figure 4, database 3 registers product data such as the official name of the product, such as "Brand M Product Name A 6-can pack Box of 180g x 6 cans", the name of the manufacturer that produces the product, such as "Manufacturer M", the brand name of the manufacturer, such as "Brand M", the product category, such as "Soft Drink", and the JAN code, such as "XX-XXXXX-XXXXX-X".
[0031] The second estimation unit 13 takes the product information output from the first estimation unit 11 as input, based on the relationships analyzed by the second machine learning unit 12, and estimates and outputs the official name information of that product. For example, in the neural network described above, the product information output from the first estimation unit 11 is input to the input layer, and the official name information of that product registered in the database 3 is estimated and output from the output layer, thereby estimating the official name information of the product written on the receipt. For example, the second estimation unit 13 takes the product information "Brand M Product Name A" output from the first estimation unit 11 as input and estimates and outputs the official name information of that product, "Brand M Product Name A 6-can pack Box of 180g x 6 cans".
[0032] The first information acquisition unit 14 obtains the correct store name, correct store phone number, or correct store address information by performing an information search (for example, an internet search) based on at least one piece of information output from the first estimation unit 11, which is the store name, store phone number, or store address of the store that sells the product. For example, based on the store name "Brand KT Town Store" output from the first estimation unit 11, the correct store address "1-1 T Town, K Ward, Tokyo" is obtained.
[0033] The second information acquisition unit 15 obtains information such as the manufacturer's name or brand name of the product, or the product's category or JAN code, by referring to the database 3 based on the official product name information output from the second estimation unit 13. For example, based on the official product name "Brand M Product Name A 6-can pack Box of 180g x 6 cans" output from the second estimation unit 13, the manufacturer's name "Manufacturer M", the manufacturer's brand name "Brand M", the product category "Soft drink", and the JAN code "XX-XXXXX-XXXXX-X" are obtained.
[0034] The purchase data creation unit 16 creates product purchase data by adding information other than the product output from the first estimation unit 11 to the official product name information output from the second estimation unit 13. The purchase data creation unit 16 can also create product purchase data by adding the correct store name, correct store telephone number, or correct store address information to the official product name information output from the second estimation unit 13. Furthermore, the purchase data creation unit 16 can create product purchase data by adding the manufacturer's name or brand name, or the product category or JAN code information to the official product name information output from the second estimation unit 13. The product purchase data thus created is output from the input / output unit 4.
[0035] The operation of the purchase data output system 1, configured as described above, will be explained with reference to the flowchart in Figure 5.
[0036] As shown in Figure 5, in the purchase data output system 1 of this embodiment, first, the first machine learning unit 9 analyzes the relationship between the receipt image and the product information and non-product information described in the receipt image using machine learning (S1). Then, the second machine learning unit 12 analyzes the relationship between the product information described in the receipt image and the information of the official name of that product registered in the database 3 using machine learning (S2).
[0037] Next, the image processing unit 10 performs edge detection and plane transformation on the receipt image acquired from the user terminal 2, and then applies either distortion correction or noise reduction image processing (S3). Then, the first estimation unit 11 takes the processed receipt image as input and estimates and outputs product information and non-product information described in the receipt image based on the relationships analyzed by the first machine learning unit 9 (S4). Furthermore, the second estimation unit 13 takes the product information output by the first estimation unit 11 as input and estimates and outputs the official name information of the product based on the relationships analyzed by the second machine learning unit 12 (S5).
[0038] Then, in the purchase data creation unit 16, information other than the product output in the first estimation step is added to the official name information of the product output by the second estimation unit 13 to create product purchase data (S6). At this time, the purchase data creation unit 16 may also create product purchase data by adding the correct store name, correct store telephone number, or correct store address information of the store that sells the product to the official name information of the product output from the second estimation unit 13. Alternatively, the purchase data creation unit 16 may also create product purchase data by adding the manufacturer's name or brand name, or the product category or JAN code information to the official name information of the product output from the second estimation unit 13. The product purchase data created in this way is then output from the input / output unit 4.
[0039] According to the purchase data output system 1 of this embodiment, product information and non-product information described in the receipt image are estimated and output from the receipt image acquired from the user terminal 2. The official name of the product is then estimated and output from the estimated product information. By adding non-product information (for example, the store's company name or brand name, store name or store phone number or store address, purchase date or purchase time, purchase quantity or unit price or subtotal or total or tax, payment method, etc.) to the official name of the product, product purchase data is created and output. In this way, the information described in the receipt can be converted into product purchase data. In this case, since the product purchase data includes not only the official name of the product but also non-product information, it can provide users with highly valuable information.
[0040] Furthermore, in this embodiment, by performing an information search (for example, an internet search or a search on a store information service) based on at least one piece of information estimated from the receipt image—either the store name, the store phone number, or the store address—the correct store name, phone number, or store address of the store selling the product is obtained and added to the product purchase data as information other than the product itself. This makes it possible to provide users with highly valuable information, such as the correct store name, phone number, or address of the store selling the product.
[0041] Furthermore, in this embodiment, by referring to database 3, information such as the manufacturer's name or brand name, or the product's category or JAN code can be obtained from the product's official name, and this information is added to the product's purchase data as information other than the product itself. This makes it possible to provide users with highly valuable information such as the manufacturer's name or brand name, or the product's category or JAN code.
[0042] Although embodiments of the present invention have been described above by example, the scope of the present invention is not limited to these, and modifications and alterations can be made within the scope described in the claims depending on the purpose. [Industrial applicability]
[0043] As described above, the purchase data output system according to the present invention has the effect of providing users with highly valuable information and is useful as a system for digitizing receipt information. [Explanation of Symbols]
[0044] 1. Purchase Data Output System 2 User terminals 3 Databases 4 Input / output section 5 Storage section 6. System 1 7. System 2 8. Third System 9. First Machine Learning Department 10 Image Processing Unit 11 First estimation part 12. Second Machine Learning Department 13 Second estimation part 14 1st Information Acquisition Department 15 2nd Information Acquisition Department 16. Purchasing Data Creation Department 17 Edge detection unit 18 Planar transformation section 19. Binarization section 20 Smoothing section 21 Noise Reduction Section N Network
Claims
1. A purchase data output system that outputs product purchase data, The first machine learning unit analyzes the relationship between a receipt image and the product information and other information listed in the receipt image using machine learning. A second machine learning unit analyzes the relationship between the product information listed in the aforementioned receipt image and the official name information of the product registered in a predetermined database using machine learning. An image processing unit that performs edge detection and plane transformation on a receipt image acquired from a user terminal, and then applies either distortion correction or noise reduction to the image processing, Based on the relationships analyzed by the first machine learning unit, the first estimation unit takes the processed receipt image as input and estimates and outputs product information and non-product information described in the receipt image. Based on the relationships analyzed by the second machine learning unit, the second estimation unit takes the product information output from the first estimation unit as input and estimates and outputs the official name information of the product. Equipped with, Information other than the aforementioned product includes any of the following: the company name or brand name of the store selling the product, the name of the store selling the product, the store's telephone number or address, the date or time of purchase of the product, the quantity or unit price or subtotal or total or tax of the product, or the payment method used when purchasing the product. The aforementioned purchase data output system is A purchase data creation unit creates purchase data for the product by adding information other than the product output from the first estimation unit to the information of the official name of the product output from the second estimation unit, Equipped with, The system includes a first information acquisition unit that obtains information on the correct store name, correct store telephone number, or correct store address of a store that sells the product by performing an information search based on at least one piece of information output from the first estimation unit, which is the store name, store telephone number, or store address of the store that sells the product. The purchase data creation unit adds information about the correct store name, correct store telephone number, or correct store address that sells the product to the information about the official name of the product output from the second estimation unit, thereby creating purchase data for the product.
2. A purchase data output system that outputs product purchase data, The first machine learning unit analyzes the relationship between a receipt image and the product information and other information listed in the receipt image using machine learning. A second machine learning unit analyzes the relationship between the product information listed in the aforementioned receipt image and the official name information of the product registered in a predetermined database using machine learning. An image processing unit that performs edge detection and plane transformation on a receipt image acquired from a user terminal, and then applies either distortion correction or noise reduction to the image processing, Based on the relationships analyzed by the first machine learning unit, the first estimation unit takes the processed receipt image as input and estimates and outputs product information and non-product information described in the receipt image. Based on the relationships analyzed by the second machine learning unit, the second estimation unit takes the product information output from the first estimation unit as input and estimates and outputs the official name information of the product. Equipped with, Information other than the aforementioned product includes any of the following: the company name or brand name of the store selling the product, the name of the store selling the product, the store's telephone number or address, the date or time of purchase of the product, the quantity or unit price or subtotal or total or tax of the product, or the payment method used when purchasing the product. The aforementioned purchase data output system is A purchase data creation unit creates purchase data for the product by adding information other than the product output from the first estimation unit to the information of the official name of the product output from the second estimation unit, Equipped with, The system includes a second information acquisition unit that, based on the information of the official name of the product output from the second estimation unit, obtains the manufacturer name or brand name of the manufacturer of the product, or the category or JAN code information of the product, by referring to the predetermined database. The purchase data creation unit is a purchase data output system that creates purchase data for a product by adding the manufacturer's name or brand name, or the product's category or JAN code information, to the information of the official name of the product output from the second estimation unit.
3. A method performed by a purchase data output system that outputs product purchase data, The aforementioned method, The first machine learning step involves analyzing the relationship between a receipt image and the product information and other information listed in the receipt image using machine learning. A second machine learning step involves using machine learning to analyze the relationship between the product information listed on the receipt image and the official name information of the product registered in a predetermined database. Image processing steps include: performing edge detection and plane transformation on a receipt image acquired from a user terminal, followed by image processing of either distortion correction or noise reduction; Based on the relationships analyzed in the first machine learning step, the first estimation step takes the processed receipt image as input and estimates and outputs product information and non-product information described in the receipt image. Based on the relationships analyzed in the second machine learning step, the second estimation step takes the product information output in the first estimation step as input and estimates and outputs the information of the official name of the product. Equipped with, Information other than the aforementioned product includes any of the following: the company name or brand name of the store selling the product, the name of the store selling the product, the store's telephone number or address, the date or time of purchase of the product, the quantity or unit price or subtotal or total or tax of the product, or the payment method used when purchasing the product. The aforementioned method, A purchase data creation step involves adding information other than the product output in the first estimation step to the information of the official name of the product output in the second estimation step to create purchase data for the product, Includes, The above method further, The process includes a first information acquisition step in which information is obtained by performing an information search based on at least one of the following pieces of information output in the first estimation step: the name of the store that sells the product, the store's telephone number, or the store's address, thereby obtaining the correct name of the store that sells the product, the correct telephone number, or the correct address of the store that sells the product. The purchase data creation step involves adding information about the correct store name, correct store telephone number, or correct store address that sells the product to the information about the official name of the product output in the second estimation step, thereby creating purchase data for the product.
4. A method performed by a purchase data output system that outputs product purchase data, The aforementioned method, The first machine learning step involves analyzing the relationship between a receipt image and the product information and other information listed in the receipt image using machine learning. A second machine learning step involves using machine learning to analyze the relationship between the product information listed on the receipt image and the official name information of the product registered in a predetermined database. Image processing steps include: performing edge detection and plane transformation on a receipt image acquired from a user terminal, followed by image processing of either distortion correction or noise reduction; Based on the relationships analyzed in the first machine learning step, the first estimation step takes the processed receipt image as input and estimates and outputs product information and non-product information described in the receipt image. Based on the relationships analyzed in the second machine learning step, the second estimation step takes the product information output in the first estimation step as input and estimates and outputs the information of the official name of the product. Equipped with, Information other than the aforementioned product includes any of the following: the company name or brand name of the store selling the product, the name of the store selling the product, the store's telephone number or address, the date or time of purchase of the product, the quantity or unit price or subtotal or total or tax of the product, or the payment method used when purchasing the product. The aforementioned method, A purchase data creation step involves adding information other than the product output in the first estimation step to the information of the official name of the product output in the second estimation step to create purchase data for the product, Includes, The above method further, Based on the information of the official name of the product output in the second estimation step, the second information acquisition step includes obtaining the manufacturer name or brand name of the manufacturer of the product, or the category or JAN code of the product, by referring to the predetermined database. The purchase data creation step involves adding the manufacturer's name or brand name, or the product's category or JAN code information, to the official name information of the product output in the second estimation step to create purchase data for the product.