JP4981511B2 - How to distribute product data - Google Patents

Description

  The present invention relates to a method, a server, and a program for distributing product data.

  Conventionally, as an advertisement on the Internet, a service for displaying product information (recommended product, advertisement, etc.) customized for each user on a Web page has been provided. In such a case, for example, based on user attribute information, purchase history information, and the like, product information to be displayed is selected based on the degree of relevance.

For example, Patent Document 1 discloses a method for matching a user profile created from a usage history by calculating a similarity between concept vectors. In this way, attempts have been made to effectively present product information such as advertisements by grouping and selecting users with similar trends.
JP 2004-126975 A

  However, since the products that the user actually wants are studied from various aspects, even with the above-described technology, matching between user information and between user information and product information is not always sufficient. In particular, if only one value (feature) is defined for an attribute of a user profile, it is difficult to capture a user's preference tendency that changes depending on the relationship (condition) with other attributes.

  For example, the user's tendency to the price of a product to be purchased varies depending on the category of the product (jewel, bag, sweater, food, etc.). It cannot be characterized by one value (like expensive products, cheap products, etc.).

  SUMMARY OF THE INVENTION An object of the present invention is to provide a method of effectively recommending (advertisement) a product that is easy to draw attention by grasping the tendency of the user's price preference that varies depending on the product category.

  For the above purpose, the following are specifically provided.

(1) A method for distributing product data to a user,
Receiving past product purchase history data of a plurality of users including the user;
For each of the plurality of users based on the product purchase history data, generating distribution data indicating purchase price ranges for each product category;
Extracting second distribution data similar to the first distribution data relating to the user and relating to another user different from the user;
Selecting the product data to be distributed to the user based on a difference between the second distribution data and the first distribution data.

  According to such a configuration, the server executing the method accepts the past product purchase history data of a plurality of users, and purchase price bands for each product category for each of the plurality of users based on the product purchase history data. Distribution data is generated, second distribution data related to another user different from the user, similar to the first distribution data related to the user, is extracted, and the first distribution data in the second distribution data is extracted. Product data to be distributed to the user is selected based on the difference.

  By this, the server can represent the purchase price trend (product purchase history) that varies depending on the user as distribution data indicating the purchase price range for each product category. Product data can be distributed (advertised) based on the difference in distribution data.

  As a result, the server can present, for example, a product purchased by another user with a similar tendency (preference) to the purchase price as a recommended product. The advertising effect may increase.

  (2) The method according to (1), wherein distribution data in which a portion matching a distribution map based on the first distribution data exceeds a predetermined ratio is extracted as the second distribution data.

  According to such a configuration, when the server executing the method determines the similarity of the distribution data, the matching portion exceeds a predetermined ratio, that is, the area where two distribution maps overlap is a predetermined ratio. It is subject to exceeding.

  As a result, the server can extract users who tend to purchase in the same price range as the distribution target user regardless of the product category, and thus effectively search for users with similar price preference trends. be able to.

  (3) The method according to (1), wherein distribution data whose image similarity with a distribution map based on the first distribution data is greater than or equal to a predetermined value is extracted as the second distribution data.

  According to such a configuration, the server executing the method compares the distribution maps as images when determining the similarity of the distribution data. As a result, even if the matching area is small, a slight price range deviation or product category deviation can be determined to be similar, so that the possibility that similar distribution data can be extracted increases.

  Here, in order to determine that the deviation of the product category within a predetermined range is similar, it is desirable to generate a distribution map so that similar product categories are arranged. Accordingly, if the images are similar, it can be considered that the same product has been purchased at the same price, so that similar distribution data can be extracted effectively by the image matching technique.

  Various existing methods can be used for the image matching technique, and a suitable method is selected in accordance with the specifications of the distribution map.

(4) The distribution data further includes data indicating the number of purchases in the purchase price range,
The method according to (2) or (3), wherein the distribution map can distinguish the number of purchases.

  According to such a configuration, since the distribution data further includes data indicating the number of purchases in the purchase price range, and the distribution diagram can distinguish the number of purchases, the server executing the method includes the purchase frequency. It is possible to grasp the product purchase tendency of the user.

  As a result, the server can extract users with similar tendencies using distribution data whose amount of information has increased according to the purchase frequency. Therefore, merchandise data based on the merchandise purchase history of users with a strong similar tendency. Can be delivered.

  Note that the data indicating the number of purchases can be expressed by, for example, a color difference in the distribution chart. According to this, for example, similarity determination including data indicating the number of purchases can be performed by image matching.

(5) The distribution data further includes data indicating purchase time,
The method according to any one of (2) to (4), wherein the distribution map is expressed in three dimensions related to the product category, the purchase price range, and the purchase time.

  According to such a configuration, the distribution data further includes data indicating the purchase time, and the distribution diagram is represented in three dimensions regarding the product category, the purchase price range, and the purchase time.

  As a result, the server that executes the method uses distribution data whose amount of information has increased depending on the purchase time, so that it can be expected that the matching accuracy will be improved, and it is possible to effectively extract users with more similar trends. There is a possibility.

  (6) The step of selecting the product data includes selecting product data corresponding to product purchase history data included in the second distribution data and not included in the first distribution data. The method according to any one of 1) to (5).

  According to such a configuration, the server that executes the method is included in the second (similar user) distribution data as the product data to be distributed to the target user, and the first (target user) distribution data is included in the distribution data. Product data corresponding to product purchase history data not included is selected.

  As a result, the server selects merchandise data that has not yet been purchased from merchandise purchase history data of users with similar purchase tendencies, so that it is possible to distribute merchandise data that is likely to result in purchase.

  As a result, product data useful for the user is presented, and an effective advertisement can be made for the product seller.

  (7) In any one of (1) to (6), the step of selecting the product data includes selecting a predetermined number of product data included in the second distribution data and having a new purchase date and time. The method described.

  According to such a configuration, the server that executes the method selects the product data distributed to the target user that is included in the second (similar user) distribution data and has a new purchase date and time.

  As a result, the server selects the recently purchased product data from the product purchase history data of users with similar purchase tendencies, so that timely product recommendation is possible. That is, it is possible to effectively present new products, products in line with recent trends, etc., and prompt users to purchase.

  As a result, there is a high possibility that product data useful for the user will be presented, and it is possible to provide an effective advertisement for the product seller.

  (8) In any one of (1) to (7), the step of selecting the product data selects product data related to a product category in which product purchase history data does not exist in the first distribution data. The method described.

  According to such a configuration, the server that executes the method preferentially selects the product data related to the product category that the user has not yet purchased, so that it is possible to present product data that is likely to be purchased in the future. it can.

  Accordingly, there is a high possibility that the product data that the user is interested in purchasing newly is presented. Moreover, it can be set as an effective advertisement as a means for acquiring new customers for the merchandise seller.

  (9) The method according to any one of (1) to (8), wherein the distribution data includes category data indicating a set of a plurality of product categories related to each other.

  According to such a configuration, the server that executes the method classifies a plurality of product categories (for example, a sweater and a jacket) that are related to each other into categories. As a result, it is possible to use the two product classification scales of the category and the product category, so that the characteristics of the product purchase history can be better understood.

(10) The method further includes a step of accepting designation of the category data,
The method according to (9), wherein the distribution data is generated based on product purchase history data relating to product categories belonging to the specified category data.

  According to such a configuration, the server that executes the method receives the specification of category data, and generates distribution data based on the product purchase history data related to the product category belonging to the specified category data.

  Thus, the server generates distribution data based on the product purchase history data related to the product category belonging to the category data based on the category data specified by the user, the Web server, or the like.

  As a result, it is possible to match trends (similarity determination) that specify a category (field of clothing, etc.) without being affected by the purchase history of other categories. Useful.

  The designation of the category data may be explicitly received from the user, or the category may be determined based on a keyword extracted from text data such as a web page.

(11) The method further includes the step of obtaining category data having a high purchase frequency in a predetermined period from the product purchase history data included in the first distribution data,
The method according to (9) or (10), wherein the step of selecting the product data selects product data in the category data having a high purchase frequency.

  According to such a configuration, the server executing the method obtains category data having a high purchase frequency in a predetermined period from among the product purchase history data included in the first (target user) distribution data, and purchases the category data. Product data in category data with high frequency is selected.

  As a result, the server preferentially selects product data of a type that the user frequently purchases, and thus there is a possibility that the product that the user is interested in can be effectively recommended.

(12) The method further includes a step of obtaining a purchase price range having a high purchase frequency in a predetermined period among the product purchase history data included in the first distribution data,
The method according to any one of (1) to (11), wherein the step of selecting the product data selects product data in a nearby purchase price range including the purchase price range having a high purchase frequency.

  According to such a configuration, the server that executes the method obtains a purchase price range having a high purchase frequency in a predetermined period from the product purchase history data included in the first (target user) distribution data, Product data in a nearby purchase price range including a purchase price range having a high purchase frequency is selected.

  As a result, the server preferentially selects the product data for the amount of money frequently purchased by the user and the amount of money nearby. As a result, there is a possibility that a product that the user can easily purchase in terms of money can be effectively recommended.

(13) A server for distributing product data to a user,
Means for receiving past product purchase history data of a plurality of users including the user;
Based on the product purchase history data, for each of the plurality of users, means for generating distribution data indicating a purchase price range for each product category;
Means for extracting second distribution data relating to another user different from the user, similar to the first distribution data relating to the user;
A server comprising: means for selecting the product data to be distributed to the user based on a difference between the second distribution data and the first distribution data.

  According to such a configuration, the same effect as in (1) can be expected by operating the server.

(14) A program for distributing product data to a user,
Receiving past product purchase history data of a plurality of users including the user;
For each of the plurality of users based on the product purchase history data, generating distribution data indicating purchase price ranges for each product category;
Extracting second distribution data similar to the first distribution data relating to the user and relating to another user different from the user;
The program which makes a server perform the step which selects the said merchandise data delivered with respect to the said user based on the difference with the said 1st distribution data in the said 2nd distribution data.

  According to such a configuration, the same effect as in (1) can be expected by executing the program on the server.

  ADVANTAGE OF THE INVENTION According to this invention, the tendency of a user's price preference which changes with goods classification can be caught, and the product advertisement which is easy to attract interest can be performed effectively.

  An example of a preferred embodiment according to the present invention will be described below based on the drawings.

[System overall configuration]
FIG. 1 is a conceptual diagram showing the overall configuration of an advertisement providing system according to an example of a preferred embodiment of the present invention.

  The matching server 10, the Web server 20, the advertisement server 30, and the user terminal device 40 are connected via a communication line (Internet). Similarity matching according to the present invention is performed by the matching server 10 and forms an advertisement providing system in cooperation with the Web server 20 and the advertisement server 30.

  The user information (product purchase history data) of the terminal device 40 is acquired by the Web server 20 and transmitted to the matching server 10. Specifically, the Web server 20 may be, for example, an EC (Electronic Commerce) site or another home page, and receives user information from the terminal device 40.

  The matching server 10 stores the product purchase history data of the user received from the web server 20, and performs similarity matching processing by comparing the stored product purchase history data of a plurality of users. The advertisement data (product data) managed by the advertisement server 30 is selected based on the product purchase history data of the users with similar product purchase tendencies extracted as a result.

  The advertisement server 30 may collectively manage a plurality of advertisement data that can be displayed in advance, or refer to and acquire advertisement data distributed to a plurality of servers based on the processing result of the matching server 10. Also good.

  The advertisement data acquired by the matching server 10 is transmitted to the Web server 20 and configured as a part of the Web page displayed on the terminal device 40. Thereby, the advertisement suitable for the user of the terminal device 40 is displayed based on the similarity matching process of the matching server 10.

[Server function configuration]
FIG. 2 is a block diagram showing each function of the matching server 10 according to an example of the preferred embodiment of the present invention.

  The matching server 10 includes a purchase status collection unit 110, a purchase distribution comparison unit 120, a similar distribution extraction unit 130, a product information selection unit 140, a product advertisement transmission unit 150, and a purchase history DB 160. Send and receive.

  The purchase status collection unit 110 receives user information (product purchase history data) from the Web server 20 and stores it in the purchase history DB 160. Thereby, information related to the purchase price of the products received from a plurality of users is stored, and this becomes processing target data for similarity matching and product data selection described later.

  The purchase distribution comparison unit 120 refers to the product purchase history data stored in the purchase history DB 160, generates distribution data indicating the relationship between the product category in which similar products are grouped and the purchase price, and each of the plurality of users. Compare distribution data.

  The similar distribution extraction unit 130 extracts distribution data related to other users similar to the distribution data related to the user of the terminal device 40 based on the comparison result between the distribution data by the purchase distribution comparison unit 120. As a result, it is possible to acquire users with similar purchase price trends (preferences) and distribution data (product purchase history data).

  The product information selection unit 140 selects product data to be advertisement information based on the difference between the distribution data of similar users extracted by the similar distribution extraction unit 130 and the distribution data of users of the terminal device 40. This product data is managed by the advertisement server 30.

  The product advertisement transmission unit 150 generates product advertisement data to be displayed on the web page based on the product data selected by the product information selection unit 140 and transmits the product advertisement data to the web server 20. As a result, the matching server 10 transmits product advertisement data to the Web server 20 based on the received product purchase history data.

[Product purchase history table]
FIG. 3 is a diagram showing a product purchase history table according to an example of the preferred embodiment of the present invention.

  The product purchase history table in the purchase history DB 160 stores the product purchased by the user, the product category, the purchase price, and the purchase date in association with the ID for identifying the user. Thereby, the matching server 10 can manage the purchase amount and the purchase time for each product category for each user who uses the advertisement providing system of the present embodiment as a history.

[Distribution map]
FIG. 4 is an example of a distribution diagram based on a product purchase history according to an example of the preferred embodiment of the present invention.

  The horizontal axis is the product category (jewels, bags, cars, etc.), and the vertical axis is the monetary range (5,000 yen or more but less than 10,000 yen, 10,000 yen or more and less than 20,000 yen), and is expressed in a two-dimensional matrix. . Each cell corresponds to one product category and a price range, and a distribution map is obtained by filling the corresponding cell according to the product category and the price of the product purchased by the user.

  Here, the width of the amount of money indicated by one cell can be a predetermined width set in advance, but is not limited to this. For example, a logarithmic value may be assigned with a predetermined width. According to this, it is possible to widen the range of the amount of money in the high-priced product and narrow the range of the amount of money in the low-priced product. Alternatively, it is possible to eliminate the inconvenience that data that is too wide is separated.

[Processing flow]
FIG. 5 is a diagram showing a flow of processing in the matching server 10 according to an example of the preferred embodiment of the present invention.

  In step S <b> 11, the purchase status collection unit 110 receives user information from the Web server 20. Specifically, it is data indicating the product category and the amount related to the product purchased by the user of the terminal device 40, and is stored in the purchase history DB 160 as a product purchase history table.

  In step S12, the purchase distribution comparison unit 120 refers to the product purchase history data related to the user of the terminal device 40 stored in the purchase history DB 160, and reads the distribution data. Specifically, it is data corresponding to the distribution chart shown in FIG. 4 and may be generated at the time of reference, but distribution data generated in advance may be stored in the purchase history DB 160.

  In step S <b> 13, the purchase distribution comparison unit 120 reads out distribution data (distribution map) regarding a user different from the user of the terminal device 40.

  In step S14, the purchase distribution comparison unit 120 compares the distribution data (distribution map) read in steps S12 and S13, and calculates the similarity. Specifically, for example, the coincidence area between the distribution diagrams is obtained, and the coincidence ratio is used as the similarity. Also, distribution maps as images can be compared by image matching technology.

  FIG. 6 shows an example of distribution data determined to be similar in the similarity evaluation. The distribution data regarding the user of the terminal device 40 is represented by the distribution chart of (A). Compared with this, the distribution chart of (B) agrees at a high rate. In the distribution chart of (C), although the matching ratio is not so high as in (B), it can be determined that the images (graphics) formed by the painted cells are similar.

  For image matching, an existing technique can be used, and can be realized by a method capable of determining similarity of figures.

  In step S15, the purchase distribution comparison unit 120 determines whether or not a predetermined number of distribution data has been compared. That is, steps S13 to S15 are repeated until the comparison with the distribution data regarding users belonging to a preset group (for example, a group of users residing in the same region or a group of users of the same age) is completed.

  Alternatively, the loop of steps S13 to S15 may be repeated a predetermined number of times, or the comparison process may be terminated at the stage where distribution data having a similarity equal to or higher than a predetermined level is found, and the process may be shifted to step S16.

  In step S16, the similar distribution extraction unit 130 extracts distribution data having a similarity equal to or higher than a predetermined level based on the distribution data comparison result in step S14. This makes it possible to acquire purchase history data and a user whose purchase price is similar to the user of the terminal device 40.

  It should be noted that the distribution data to be extracted may be a plurality of data having a degree of similarity equal to or higher than a predetermined value, or may be one that is most similar. Alternatively, a predetermined number may be extracted in descending order of similarity.

  In step S17, the product information selection unit 140 selects a product to advertise based on the distribution data extracted in step S16. Specifically, for example, the product category and the price range are determined based on the difference between the distribution data.

  For example, in FIG. 6B, the cell denoted by reference numeral 51 is a portion that is not painted in the distribution diagram of FIG. That is, since the user of (B) who has a similar tendency has purchased, but the user of (A) can be determined to be a product group that has not yet been purchased, it can be expected that the advertising effect on the user of (A) is high. . Therefore, the product belonging to the product category and the price range corresponding to this cell is selected as the advertising product. Then, the product information selection unit 140 receives product data (advertisement data) related to the selected product from the advertisement server 30.

  In step S <b> 18, the product advertisement transmission unit 150 transmits the advertisement data related to the product received in step S <b> 17 to the Web server 20. Accordingly, the page data is appropriately updated in the Web server 20, and the product advertisement can be performed on the Web page displayed on the terminal device 40.

[Similar comparison and product selection example]
7-10 is a figure which shows the variation of the similarity comparison in the distribution map which concerns on an example of suitable embodiment of this invention, and the selection method of advertisement goods.

  In the distribution chart shown in FIG. 7, the purchase frequency (number of purchases) in the same product category and price range is further expressed. That is, the cell indicated by reference numeral 52 indicates that the purchase frequency is high by making it darker than the other cells. Thus, for example, according to similarity evaluation by image matching, a more detailed similarity can be obtained by expressing the purchase frequency as an image.

  Note that the method of expressing the purchase frequency can be indicated by, for example, a color difference or a predetermined number of gradation expressions. The purchase frequency may be the cumulative number of product purchases or the number of purchases within a predetermined period.

  In the distribution diagram shown in FIG. 8, the purchase history is concentrated in the area indicated by reference numeral 53. In such a case, it can be determined that the user is highly interested in the product category of the area. Therefore, an effective advertisement can be performed by selecting product data belonging to this product category.

  In the distribution diagram shown in FIG. 9, the purchase history is concentrated in the area indicated by reference numeral 54. In such a case, it can be determined that the user is highly interested in the amount of money in the area. Therefore, an effective advertisement can be performed by selecting product data belonging to this monetary range and its surroundings.

  The distribution diagram shown in FIG. 10 schematically shows a product purchase history obtained by adding a purchase category to a product category and a purchase price. Since the distribution data is expressed as a three-dimensional solid figure, the evaluation of the similarity is based on the volume of the matching part and the matching processing of the three-dimensional image.

  According to this, since the amount of information relating to the purchase history is further increased as compared with the two-dimensional plane figure, it is possible to expect more detailed comparison and improvement in accuracy of similarity evaluation.

[Hardware configuration of server]
FIG. 11 is a diagram illustrating an example of a hardware configuration of the matching server 10 described in FIG. The matching server 10 includes a CPU (Central Processing Unit) 1010 constituting the control unit 101 (in the multiprocessor configuration, a plurality of CPUs such as a CPU 1012 may be added), a bus line 1005, a communication I / F 1040, a main memory 1050, a BIOS. (Basic Input Output System) 1060, a USB port 1090, an I / O controller 1070, input means such as a keyboard and mouse 1100, and a display device 1022.

  Storage means such as a tape drive 1072, a hard disk 1074, an optical disk drive 1076, and a semiconductor memory 1078 can be connected to the I / O controller 1070.

  The BIOS 1060 stores a boot program executed by the CPU 1010 when the matching server 10 is activated, a program depending on the hardware of the matching server 10, and the like.

  The hard disk 1074 constituting the storage unit 107 stores various programs for the matching server 10 to function as a server and programs for executing the functions of the present invention. Further, various databases can be configured as necessary.

  As the optical disc drive 1076, for example, a DVD-ROM drive, a CD-ROM drive, a DVD-RAM drive, or a CD-RAM drive can be used. In this case, the optical disk 1077 corresponding to each drive is used. A program or data can be read from the optical disk 1077 by the optical disk drive 1076 and provided to the main memory 1050 or the hard disk 1074 via the I / O controller 1070.

  The program provided to the matching server 10 is provided by being stored in a recording medium such as the hard disk 1074, the optical disk 1077, or a memory card. This program may be installed in the matching server 10 and executed by being read from a recording medium via the I / O controller 1070 or downloaded via the communication I / F 1040.

  The aforementioned program may be stored in an internal or external storage medium. Here, as a storage medium constituting the storage unit 107, in addition to the hard disk 1074, the optical disk 1077, or the memory card, a magneto-optical recording medium such as an MD or a tape medium can be used. Further, a storage device such as a hard disk 1074 or an optical disk library provided in a server system connected to a dedicated communication line or the Internet may be used as a recording medium, and the program may be provided to the matching server 10 via the communication line.

  Here, the display device 1022 displays a screen for accepting data input to the user or displays a screen of a result of arithmetic processing performed by the matching server 10, and includes a cathode ray tube display device (CRT), a liquid crystal display device ( LCD).

  Here, the input means accepts input by the user, and may be configured by a keyboard, a mouse 1100, and the like.

  The communication I / F 1040 is a network adapter for enabling the matching server 10 to be connected to a terminal via a dedicated network or a public network. The communication I / F 1040 may include a modem, a cable modem, and an Ethernet (registered trademark) adapter.

  Although the above example mainly demonstrated the matching server 10, the function demonstrated above can also be implement | achieved by installing a program in a computer and operating the computer as a server apparatus. Therefore, the functions realized by the server described as an embodiment in the present invention can be realized by executing the above-described method by the computer, or by introducing the above-mentioned program into the computer and executing it. It is.

  Further, the web server 20 and the advertisement server 30 have the same configuration as the matching server 10.

[Device hardware configuration]
The terminal device 40 also has the same configuration as the matching server 10 described above. In the above-described example, an example realized by a so-called computer has been described. Furthermore, as long as the principle of the present invention is applicable, it can be realized by various terminals such as a mobile phone, a PDA (Personal Data Assistant), and a game machine. It's okay.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

It is a conceptual diagram showing the whole structure of the advertisement provision system which concerns on an example of suitable embodiment of this invention. It is a block diagram showing each function of the matching server 10 which concerns on an example of suitable embodiment of this invention. It is a figure which shows the goods purchase history table which concerns on an example of suitable embodiment of this invention. It is an example of the distribution map based on the goods purchase history which concerns on an example of suitable embodiment of this invention. It is a figure which shows the flow of a process in the matching server 10 which concerns on an example of suitable embodiment of this invention. The example of the distribution data judged to be similar in the similarity evaluation which concerns on an example of suitable embodiment of this invention is shown. It is a figure which shows the variation of the similarity comparison in the distribution map which concerns on an example of suitable embodiment of this invention, and the selection method of advertisement goods. It is a figure which shows the variation of the similarity comparison in the distribution map which concerns on an example of suitable embodiment of this invention, and the selection method of advertisement goods. It is a figure which shows the variation of the similarity comparison in the distribution map which concerns on an example of suitable embodiment of this invention, and the selection method of advertisement goods. It is a figure which shows the variation of the similarity comparison in the distribution map which concerns on an example of suitable embodiment of this invention, and the selection method of advertisement goods. It is a figure which shows an example of the hardware constitutions of the matching server 10 which concerns on an example of suitable embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Matching server 20 Web server 30 Advertising server 40 Terminal device 110 Purchase status collection part 120 Purchase distribution comparison part 130 Similar distribution extraction part 140 Product information selection part 150 Product advertisement transmission part

Claims (11)

A method for delivering product data to a terminal device to which a server is connected via a network,
Receiving, from the terminal device, product purchase history data including a product category and purchase price of a product purchased by the user in the past;
Based on the product purchase history data received for each of a plurality of users, a graphic composed of cells corresponding to the products purchased by the user among a plurality of cells included in a matrix having the product category and purchase price range as coordinate axes Generating distribution data to be shown;
And the figure indicated by the first distribution data generated on the basis of said received product purchase history data was for the first user, the product purchase history data received for a second user different from the first user If a figure represented by the second distribution data generated based on comparing the image matching, the image similarity is determined to Ru der predetermined or higher, and extracting the second distribution data,
Ri cells Der included in the graphic showing the second distribution data, and, based on the product classification and the purchase price bands corresponding to the cell that is not included in the figure indicating the first distribution data, the Selecting the product data to be distributed to the first user. The method according to claim 1 , wherein the distribution data further includes data indicating the number of purchases in the product category and purchase price range, and indicates a graphic that can distinguish the number of purchases in each cell of the matrix. The distribution data further includes data indicating a purchase time, the product classification, the purchase amount band, and methods of claim 1 or claim 2 showing a three-dimensional figure relating to the purchase time. In the step of selecting the item data, the included in the product purchase history data corresponding to the second distribution data from claim 1 purchase date to select the product data for the new predetermined number to one of claims 3 The method described. The method according to any one of claims 1 to 4 , wherein in the step of selecting the product data, product data relating to a product category that does not exist in the product purchase history data corresponding to the first distribution data is selected. The method according to any one of claims 1 to 5 , wherein the distribution data includes category data indicating a set of a plurality of product categories related to each other. A step of accepting designation of the category data;
The method according to claim 6 , wherein in the generating step, the distribution data is generated based on data related to a product category belonging to the specified category data in the product purchase history data. The method further includes the step of obtaining category data having a high purchase frequency in a predetermined period from the product purchase history data received for the first user,
The method according to claim 6 or 7 , wherein in the step of selecting the product data, product data in the category data having a high purchase frequency is selected. The method further includes a step of obtaining a purchase price range having a high purchase frequency in a predetermined period of the product purchase history data received for the first user,
The method according to any one of claims 1 to 8 , wherein in the step of selecting the product data, product data in a nearby purchase price range including the purchase price range having a high purchase frequency is selected. A server that distributes product data to terminal devices connected via a network,
Means for receiving, from the terminal device, product purchase history data including a product category and purchase price of a product purchased by the user in the past;
Based on the product purchase history data received for each of a plurality of users, a graphic composed of cells corresponding to the products purchased by the user among a plurality of cells included in a matrix having the product category and purchase price range as coordinate axes Means for generating distribution data to be shown;
And the figure indicated by the first distribution data generated on the basis of said received product purchase history data was for the first user, the product purchase history data received for a second user different from the first user If a figure represented by the second distribution data generated based on comparing the image matching, the image similarity is determined to Ru der predetermined or more, it means for extracting the second distribution data,
Ri cells Der included in the graphic showing the second distribution data, and, based on the product classification and the purchase price bands corresponding to the cell that is not included in the figure indicating the first distribution data, the Means for selecting the product data to be distributed to the first user. A program for distributing product data to a terminal device connected via a network,
Receiving, from the terminal device, product purchase history data including a product category and purchase price of a product purchased by the user in the past;
Based on the product purchase history data received for each of a plurality of users, a graphic composed of cells corresponding to the products purchased by the user among a plurality of cells included in a matrix having the product category and purchase price range as coordinate axes Generating distribution data to be shown;
And the figure indicated by the first distribution data generated on the basis of said received product purchase history data was for the first user, the product purchase history data received for a second user different from the first user If a figure represented by the second distribution data generated based on comparing the image matching, the image similarity is determined to Ru der predetermined or higher, and extracting the second distribution data,
Ri cells Der included in the graphic showing the second distribution data, and, based on the product classification and the purchase price bands corresponding to the cell that is not included in the figure indicating the first distribution data, the A program for causing a server to execute the step of selecting the product data to be distributed to a first user.

Images