JP5348617B2 - Initial purchaser extraction processing method, initial purchaser recommendation method, initial purchaser extraction system, initial purchaser recommendation system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily and reliably extract a consumer corresponding to an initial purchaser. <P>SOLUTION: The initial purchaser extraction processing method includes: step S100 of acquiring sales history information on a plurality of consumers with respect to a plurality of merchandise; step S200 of, on the basis of the sales history information, detecting a fashion peak position P appearing in the temporal transition curve of the number of sales from a market time when merchandise is marketed; step S300 of, when the detected fashion peak position P satisfies the temporal conditions and numerical conditions related to the number of sales, determining that the merchandise corresponding to the temporal transition curve is merchandise to be processed; and step S400 of, on the basis of the purchase history information, extracting a purchaser who has purchased the merchandise determined as the merchandise to be processed until the lapse of a predetermined time which is earlier than the fashion peak position P from the market period as the consumer corresponding to the initial purchaser. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to an initial purchaser extraction processing method for extracting consumers corresponding to an initial purchaser who tends to purchase trendy products from a plurality of consumers, an initial purchaser extracted by the method, The present invention relates to an initial purchaser recommendation method for recommending a specific product, an initial purchaser extraction system for executing these, and an initial purchaser recommendation system.

  In general, a consumer group corresponding to an initial purchaser who performs purchase behavior immediately after a product is put on the market until the number of purchasers rapidly increases and becomes popular is known. This initial purchaser has a characteristic of proceeding with his own will and purchasing a new product without being affected by external influences such as advertisements of various media. Therefore, by recommending a product to the initial purchaser, the number of purchasers can be increased at an early stage immediately after the product is put on the market. As a result, the consumer group who purchases the product after the initial purchaser can be motivated to purchase when another person starts to purchase, and the number of purchasers can be increased.

  Therefore, conventionally, a technique for extracting consumers corresponding to initial purchasers based on purchase history information of a plurality of consumers for a plurality of products has been proposed. For example, in the prior art described in Patent Document 1, a product whose number of purchasers is equal to or greater than a preset threshold based on purchase history information is identified as a trendy product, and a certain period of time is selected from the purchasers of the trendy product A consumer who satisfies the condition is extracted as an initial purchaser.

JP 2008-198163 A

  In the above prior art, whether or not a product is a trendy product is identified only by comparing the number of purchasers and the like with a threshold value. For this reason, regardless of the period until the number of purchasers reaches the threshold value or more after the product is put on the market, if the number of purchasers is equal to or more than the threshold value, it is identified as a trendy product. However, for example, in the case of a product with a short period, there is a possibility that the number of purchasers or the like has exceeded the threshold value due to external factors such as advertisements of various media. In such a case, even if an early purchaser is extracted from among the purchasers of the product, it is possible to reliably extract the initial purchaser having the characteristic of purchasing a new product without being affected externally as described above. I could not say.

  An object of the present invention is to provide an initial purchaser extraction processing method, an initial purchaser recommendation method, an initial purchaser extraction system, and an initial purchaser recommendation system that can easily and reliably extract consumers corresponding to the initial purchaser. There is to do.

  In order to achieve the above object, the first invention associates product identification information with purchase history of a plurality of consumers for the product according to operation input information in the first information terminal on the extractor side. Accessing a consumer database storing purchase history information for each product, and acquiring the purchase history information of a plurality of consumers for a plurality of products, and the consumption acquired in the first procedure Based on the purchase history information stored in the consumer database, at least one trend peak position appearing in the time transition curve of the number of purchasers or the number of sales since the product was introduced to the market is detected for each product And the trend peak position of the time transition curve acquired in the second procedure is the predetermined time condition and the number of purchasers or the number of sales in the time transition curve. When both of the predetermined quantitative conditions to be satisfied are satisfied, based on the third procedure for determining the product corresponding to the time transition curve as the processing target product, and the purchase history information acquired in the first procedure, A consumer who purchases a product determined as a product to be processed in the third procedure from the market launch time until a predetermined time earlier than the trendy peak position elapses. And a fourth procedure for extracting as a person.

  In the first invention of this application, in extracting an initial purchaser from purchase trends of a plurality of consumers for a plurality of products, first, data capable of extracting the initial purchaser is found. That is, when the purchase history information of a plurality of consumers is acquired from the consumer database in the first procedure, the trend peak position of the time transition curve is detected for each product in the second procedure. From the characteristics of the purchase behavior of the initial purchaser as described above, it is preferable to obtain data of a product having a certain number of purchasers and the like with a relatively large peak purchaser number. When the number of buyers at the peak is relatively small, it is difficult to distinguish purchase behaviors by enthusiasts from purchase behaviors by initial purchasers or early minority purchasers. Therefore, in the third procedure, first, it is determined whether or not the trend peak position satisfies a predetermined quantitative condition. In addition, when a peak is reached only after a relatively short period of time has passed since the product was put on the market, it is difficult to distinguish between purchase behavior by the initial purchaser and purchase behavior by the early minority purchaser. In such a case, the fashion may be influenced by external factors such as various media advertisements. Therefore, in the third procedure, it is also determined whether or not the epidemic peak position satisfies a predetermined temporal condition. If the trend peak position meets both these quantitative and temporal conditions, the time transition curve is considered to represent data that can extract the initial purchaser, and the corresponding products corresponding to the time transition curve. Are determined to be products to be processed.

  Once the processing targets have been narrowed down as described above, in the fourth procedure, consumers who have purchased the product are extracted as initial purchasers from the market launch time until a predetermined time earlier than the trend peak position has elapsed. To do.

  As described above, new products that are expected to become popular in the future will be introduced to the market easily and reliably, regardless of external factors such as advertisements, among the purchasing trends of multiple consumers for multiple products. After that, it is possible to find a consumer corresponding to the initial purchaser who tends to purchase early.

In the initial purchaser extraction processing method according to the first aspect of the present invention, the third procedure includes a plurality of time periods from the market launch time to the trendy peak position stored in the consumer database as the time condition. The number of purchasers or the number of sales at the trendy peak position is stored in the consumer database as the quantitative condition, which is longer than the reference period calculated in advance using the purchase history information on the product of The product corresponding to the time transition curve is determined to be a processing target product when the purchase amount information is larger than the first reference quantity calculated in advance using the purchase history information regarding the product.

  Data that is difficult to distinguish between purchase behavior by initial buyers and purchase behavior by early minority purchasers by setting the time period from the market launch time to the peak trend position as a time condition. Can be excluded from. By making the quantity condition that the number of purchasers or sales at the trendy peak position is larger than the first reference quantity, data that is difficult to distinguish between the purchase behavior by the initial purchaser and the purchase behavior by the mania from the processing target Can be eliminated.

The initial purchaser extraction processing method according to a second aspect of the present invention is the process according to the first aspect , wherein the third procedure is a specific product belonging to a specific genre or a specific price range, and the process corresponding to the specific product. The period is longer than the reference period calculated in advance using the purchase history information related to a plurality of products in the specific genre or the specific price range accumulated in the consumer database, and the specific product The number of purchasers or the number of sales at the trendy peak position corresponding to is calculated in advance using the purchase history information related to a plurality of products in the specific genre or the specific price range stored in the consumer database. When the amount is larger than the first reference quantity, the specific product is determined as a processing target product.

  In general, consumers often take a relatively similar common consumption behavior for a plurality of products of the same genre or a plurality of products of the same price range. Therefore, in the third invention of the present application, in order to make the above determination for a product belonging to a specific genre or a specific price range, when calculating the reference period or the first reference quantity, the same specific genre or the above Calculation is performed using purchase histories of a plurality of products belonging to a specific price range. Thereby, the initial purchaser extraction process with higher accuracy can be performed.

The initial purchaser extraction processing method according to a third aspect of the present invention is the first or second aspect of the invention, wherein the purchase history information stored in the consumer database acquired in the first procedure is the genre to which the product belongs or It further includes a fifth procedure that is corrected using a correction coefficient that is determined in advance in association with the price range and stored in the product database, and the second procedure includes the purchase history information corrected in the fifth procedure. The at least one fashion peak position is detected for each product.

  A consumer's purchase trend of a product may be influenced by a genre or a price range to which the product belongs. In the fourth invention of the present application, in view of such circumstances, the correction coefficient is determined in advance in association with the genre and price range to which the product belongs. In the fifth procedure, the purchase history information acquired in the first procedure is corrected using the correction coefficient. In the subsequent second procedure, the trend peak position is detected using the purchase history information corrected as described above. Thereby, the initial purchaser extraction process with higher accuracy can be performed.

Initial purchaser extraction processing method of the fourth invention, in any one of the first to fourth invention, by the second procedure, when a plurality of the epidemic peak position is detected, the third procedure, the The first epidemic peak position does not satisfy at least one of the temporal condition and the quantitative condition, and the second epidemic peak position satisfies both the temporal condition and the quantitative condition; In addition, the number of purchasers at the first trendy peak position and the decline peak position where the number of purchasers or the number of sales is minimized between the first trendy peak position and the second trendy peak position. Alternatively, when the difference in the number of sales is larger than the second reference quantity, the product corresponding to the time transition curve is determined as the product to be processed, and the fourth procedure includes the second epidemic from the decline peak position. Earlier than peak position Consumers purchased until a predetermined time elapses, and extracts as a consumer corresponding to the initial purchaser.

  Depending on the product, after being put on the market, the number of buyers or sales increased rapidly due to various media advertisements, etc., and after that, after the fashion ended and the fashion ended, purchases were made by consumers who are not influenced by the fashion. There are some cases where the number of purchasers increases again and reaches the second peak. In the case of a product in which two trendy peak positions are formed in this way, consumers who purchased the product during the period from the market launch time until a predetermined time earlier than the first trendy peak position has passed. From the decline peak position where the number of purchasers etc. is minimized between the first fashion peak position and the second fashion peak position until a predetermined time elapses from the second fashion peak position A consumer who has purchased the product in the meantime corresponds to an initial purchaser.

Accordingly, in the fourth invention of the present application, consumers who have purchased the product are extracted from the decline peak position until a predetermined time earlier than the second trendy peak position elapses. Thereby, it is possible to easily and reliably find a consumer corresponding to the initial purchaser from among purchase trends of a plurality of consumers for a plurality of products. In addition, since the difference in the number of purchasers between the first trendy peak position and the decline peak position is larger than the predetermined second reference quantity, a mere temporary decrease in the number of purchasers, etc. Is excluded from the processing target, and a product that has been purchased by a consumer group that is not affected by the fashion after the end of the fashion due to external influences such as media advertisements can be reliably processed.

The initial purchaser extraction processing method according to a fifth aspect of the present invention is the method according to any one of the first to fourth aspects, wherein the trendy peak position satisfies the temporal condition, but the quantitative condition is If not satisfied, and the cumulative value of the number of purchasers or sales from the market launch time to the trendy peak position is greater than the third reference quantity, the product corresponding to the time transition curve is defined as the product to be processed. It is characterized by determining.

  Even if the trend peak position does not satisfy the quantitative condition, it has a certain degree of trend if viewed from the cumulative value of the number of purchasers or sales, and may be regarded as a target product if the time condition is satisfied. .

Therefore, in the fifth invention of the present application, even when the trend peak position does not satisfy the quantitative condition, if the cumulative value such as the number of purchasers from the market launch time to the trend peak position is larger than the third reference quantity, the time concerned The product corresponding to the transition curve is determined as the processing target product. Thereby, the quantitative conditions can be relaxed, and the product as described above can be determined as the product to be processed.

In order to achieve the above object, the initial purchaser recommending method according to the sixth aspect of the invention is based on the operation input information in the first information terminal on the extractor side, the product identification information and the purchase history of a plurality of consumers for the product. Are acquired in the first procedure, the consumer history database for each product is accessed, and the purchase history information of a plurality of consumers for a plurality of products is acquired. Based on the purchase history information accumulated in the consumer database, at least one fashion peak position appearing in a time transition curve of the number of purchasers or the number of sales from the time when the product was introduced to the market, The second procedure to be detected for each product, and the trend peak position of the time transition curve acquired in the second procedure are the predetermined temporal condition and the purchaser in the time transition curve. Alternatively, when both predetermined quantitative conditions regarding the number of sales are satisfied, a third procedure for determining a product corresponding to the time transition curve as a processing target product, and the purchase history information acquired in the first procedure. The consumer who purchased the product determined as the product to be processed in the third procedure from the market launch time until the predetermined time earlier than the trendy peak position has passed to the initial purchaser. The fourth procedure to be extracted as a corresponding consumer, and the second information terminal on the specific consumer side extracted as corresponding to the initial purchaser in the fourth procedure at a predetermined recommended time a sixth procedure for outputting the recommendation information items, and the third step, as the temporal condition, the elapsed period from the time-to-market to the epidemic peak position,蓄the consumer database The number of purchasers or sales at the trendy peak position is accumulated in the consumer database as the quantitative condition, which is longer than a reference period calculated in advance using the purchase history information on a plurality of products that have been made. When there is more than the first reference quantity calculated in advance using the purchase history information relating to a plurality of products, the product corresponding to the time transition curve is determined as a processing target product .

In the sixth invention of the present application, by recommending a product to a consumer group positioned as an initial purchaser, the number of purchasers or sales can be increased at an early stage immediately after the product is put on the market. As a result, it is possible to stimulate the willingness to purchase the consumer group corresponding to the early minority purchaser after another person, and to increase the number of purchasers.

In order to achieve the above object, the initial purchaser extraction system according to the seventh invention stores, for each product, purchase history information in which product identification information is associated with purchase histories of a plurality of consumers for the product. Based on the consumer database and the purchase history information accumulated in the consumer database, at least one epidemic peak position appearing in a time transition curve of the number of purchasers or the number of sales from the time when the product was introduced to the market The trend peak position detection means for detecting each product, and the trend peak position of the time transition curve detected by the trend peak position detection means is a predetermined temporal condition and the number of purchasers in the time transition curve. Alternatively, when both of the predetermined quantitative conditions regarding the number of sales are satisfied, the processing pair that determines the product corresponding to the time transition curve as the processing target product. Based on the product determination means and the purchase history information, the product determined as the processing target product by the processing target product determination means from the market launch time until a predetermined time elapses from the trendy peak position The initial purchaser extracting means for extracting the consumer purchased as a consumer corresponding to the initial purchaser, the processing target product determination means as the time condition from the market launch time to the fashion An elapsed period until the peak position is longer than a reference period calculated in advance using the purchase history information related to a plurality of products stored in the consumer database, and as the quantitative condition, at the trendy peak position A first reference in which the number of purchasers or the number of sales is calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database If greater than the amount, and wherein the determining the products corresponding to the time deviation curve processed product.

In the seventh invention of this application, when the purchase history information of a plurality of consumers is acquired from the consumer database, the trend peak position of the time transition curve is detected for each product by the trend peak position detection means. Then, the processing target product determination means determines whether the trend peak position satisfies a predetermined quantitative condition, and also determines whether the trend peak position satisfies a predetermined temporal condition. If the trend peak position meets both these quantitative and temporal conditions, the time transition curve is considered to represent data that can extract the initial purchaser, and the corresponding products corresponding to the time transition curve. Are determined to be products to be processed. After narrowing down the processing target in this way, the initial purchaser extracting means determines the consumer who has purchased the product from the time of market entry until the elapse of a predetermined time earlier than the trend peak position. Extract as

  As described above, new products that are expected to become popular in the future will be introduced to the market easily and reliably, regardless of external factors such as advertisements, among the purchasing trends of multiple consumers for multiple products. After that, it is possible to find a consumer corresponding to the initial purchaser who tends to purchase early.

In order to achieve the above object, the initial purchaser recommendation system according to the eighth aspect of the present invention stores purchase history information that associates product identification information with purchase histories of a plurality of consumers for the product for each product. Based on the consumer database and the purchase history information accumulated in the consumer database, at least one epidemic peak position appearing in a time transition curve of the number of purchasers or the number of sales from the time when the product was introduced to the market The trend peak position detection means for detecting each product, and the trend peak position of the time transition curve detected by the trend peak position detection means is a predetermined temporal condition and the number of purchasers in the time transition curve. Alternatively, when both of the predetermined quantitative conditions regarding the number of sales are satisfied, the processing pair that determines the product corresponding to the time transition curve as the processing target product. Based on the product determination means and the purchase history information, the product determined as the processing target product by the processing target product determination means from the market launch time until a predetermined time elapses from the trendy peak position An initial purchaser extracting means for extracting a consumer purchased as a consumer corresponding to the initial purchaser, and a specific consumer side extracted by the initial purchaser extracting means as corresponding to the initial purchaser 2 with respect to the information terminal, the recommended information output means for outputting the recommended information of the predetermined product at the predetermined recommended time, and the processing target product determination means from the market launch time as the time condition An elapsed period up to the trendy peak position is longer than a reference period calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database; and As the quantitative condition, when the number of purchasers or the number of sales at the trendy peak position is larger than the first reference quantity calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database In addition, a product corresponding to the time transition curve is determined as a processing target product .

In the eighth invention of the present application, the recommended information output means recommends the product to the consumer group positioned as the initial purchaser, thereby increasing the number of purchasers or the number of sales at an early stage immediately after the product is put on the market. be able to. As a result, it is possible to stimulate the willingness to purchase the consumer group corresponding to the early minority purchaser after another person, and to increase the number of purchasers.

  According to the present invention, an initial purchaser who tends to purchase early after a new product expected to be popular in the future is put on the market regardless of external factors such as advertisements easily and reliably. Find the right consumer.

It is a system configuration figure showing notionally the whole composition of the goods recommendation system of this embodiment. It is a figure showing an example of the memory content of a consumer database. It is a figure explaining the detection method of the goods used as a process target. It is a figure showing the classification | category by innovator theory. It is a flowchart showing the control content which the control part of a server performs when performing an initial buyer's extraction process. It is a flowchart showing the detailed content of the sales history information acquisition process of step S100. It is a figure showing an example of the total result performed by sales history information acquisition processing. It is a figure showing an example of the total result performed by sales history information acquisition processing. It is a flowchart showing the detailed content of the epidemic time determination process of step S200. It is a figure showing an example of the period data and sales number data which were smoothed by the epidemic time determination process. It is a figure showing an example of the period data and sales number data which were smoothed by the epidemic time determination process. It is a flowchart showing the detailed content of the process target product determination process of step S300. It is a flowchart showing the detailed content of the process target goods parameter | index calculation process of step S600. It is a flowchart showing the detailed content of the process sales history sample acquisition process of step S700. It is a flowchart showing the detailed content of the initial purchaser extraction process of step S400. It is a figure for demonstrating the method of extracting the initial purchaser in the modification in which two fashionable peak positions are formed. It is a flowchart showing the detailed content of the epidemic time determination process of step S200A. It is a flowchart showing the detailed content of the process target product determination process of step S300A. It is a flowchart showing the detailed content of the initial purchaser extraction process of step S400A. It is a figure for demonstrating the method of relaxing a quantitative condition. It is a flowchart showing the detailed content of the process target product determination process of step S300B. It is a system block diagram which represents notionally the whole structure of the goods recommendation system which implement | achieves the system which extracts an initial purchaser based on the singing trend in karaoke.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  The overall configuration of the product recommendation system 1 according to this embodiment will be described with reference to FIG. As shown in FIG. 1, the product recommendation system 1 includes a credit company 100, a server 200, a base station 300, and a mobile terminal 400 carried by each consumer. A server 200 as a first information terminal is connected to the credit company 100 via a network NW such as the Internet. A display 220, a keyboard 221, and a mouse 222 are connected to the server 200. In addition, the server 200 includes a control unit 201 that controls the entire server 200, a storage unit 210 that includes a hard disk drive (HDD), a communication control unit 202, an output control unit 203, and an input control unit 204. have. The communication control unit 202 controls information communication performed between the credit company 100 and the base station 300 via the network NW. The output control unit 203 performs control related to the output of the video signal to the display 220. The input control unit 204 performs control related to input of information via the keyboard 221 and the mouse 222.

  The server 200 functions as an initial purchaser extraction system and an initial purchaser recommendation system described in the claims.

  The storage unit 210 includes a product database 212, a program storage area 213, and a consumer database 214. The product database 212 stores recommendation information, which is information about products such as product names, product contents, prices, appearance images, etc., for recommending various products. The consumer database 214 stores sales history information for a plurality of consumers for a plurality of products. In the program storage area 213, the details will be described later, but a consumer corresponding to the initial purchaser is extracted from the consumers who have purchased the product determined as the processing target product based on the sales history information. Various programs such as an initial purchaser extraction processing program for causing the control unit 201 to execute an initial purchaser extraction processing method and an initial purchaser recommendation method (see FIGS. 5 to 15 and the like described later) are stored. The initial purchaser here is a consumer group that initiates purchase behavior immediately after a product is introduced to the market and immediately before the number of purchasers or sales increases rapidly. The company is willing to adopt and has characteristics that are not affected by external influences such as advertisements for various media such as TV and commercials.

  The control unit 201 is a so-called microcomputer. In other words, although not shown, the control unit 201 includes a central processing unit such as a CPU, a ROM, and a RAM, and stores them in the program storage area 213 of the storage unit 210 while using the temporary storage function of the RAM. Signal processing is performed in accordance with the various programs.

  The server 200 having the above configuration is connected to the base station 300 via the network NW. As a result, the recommended information of the specific product stored in the product database 212 of the server 200 can be transmitted via wireless communication to the mobile terminal 400 of each consumer corresponding to the extracted initial purchaser via the base station 300. It has become. As the mobile terminal 400, for example, a mobile phone or the like is used. In addition, for example, an information terminal such as a Personal Digital Assistant (PDA) may be used.

  A mobile terminal 400 as a second information terminal includes a display unit 401 such as a liquid crystal display that performs various displays, an operation unit 402 that includes numeric keys and operation buttons for a consumer to perform various operation inputs, and a base station 300 and an antenna 410 for transmitting and receiving information via wireless communication. In addition to the display unit 401, the operation unit 402, and the antenna 410, the mobile terminal 400 includes a wireless communication control unit 403 that controls information communication performed with the base station 300 via wireless communication, And a control unit 404 that controls the entire terminal 400. When the controller 404 receives recommended information on a specific product from the server 200 via the network NW and the base station 300, the controller 404 causes the display 401 to display the recommended information.

  An example of the contents stored in the consumer database 214 will be described with reference to FIG.

  As shown in FIG. 2, the consumer database 214 stores, for each product, sales history information or purchase history information in which a product ID, which is product identification information, and a sales history of the product are associated with each other. ing. The sales history includes a sale date and time of a product and a user ID that is identification information of a consumer who has purchased the product. The control unit 201 of the server 200 accesses the consumer database 214 in accordance with operation input information from the keyboard 221 and the mouse 222 on the extractor side, and acquires the sales history information. The sales history information and the purchase history information are substantially the same information.

  The greatest feature of the product recommendation system 1 having the above-described configuration is that the server 200 extracts consumers corresponding to the initial purchaser from among consumers who have purchased products determined as processing target products based on the sales history information. Is in the way. Details of this content will be described below.

  In the present embodiment, in extracting an initial purchaser from purchase trends of a plurality of consumers for a plurality of products, first, a product from which the initial purchaser can be extracted is found. A method for detecting a product to be processed will be described with reference to FIG. That is, when the control unit 201 acquires the sales history information of a plurality of consumers from the consumer database 214, the trend peak position P of the time transition curve representing the time change of the number of purchased products or the number of sold items for each product. Is detected. In the example shown in FIG. 3A, trendy peak positions Pa (Ta, Na) to Pd (Td, Nd)... Related to the products A to D. In this example, the vertical axis of the time transition curve is the number of sales, but it may be the number of purchasers. The same applies to the following description.

  Then, it is determined whether or not the detected trend peak position P satisfies a predetermined quantitative condition. This is because initial buyers tend to purchase trending products early, in other words, because they are consumers who create their own fashions, they acquire a certain number of sales and products with relatively large peak sales. This is because if it is not data, it will be difficult for consumers to include initial purchasers. Also, if the number of sales at the peak is relatively small, it is difficult to distinguish purchase behavior by enthusiasts from purchase behavior by initial purchasers or early minority purchasers. In the present embodiment, the quantitative condition is that the number of sales at the trendy peak position P is greater than the average number of sales Navg that is the average of the number of peak sales related to a plurality of products included in the sales history information. In the example shown in FIG. 3A, the average sales number Navg is the average value of the sales numbers Na, Nb, Nc, Nd... At the trendy peak position P of each product A to D (Na + Nb + Nc + Nd +...). / Total number of products. This average sales number Navg is calculated in advance using sales history information regarding a plurality of products stored in the consumer database 214. The average sales number Navg corresponds to the first reference quantity described in the claims.

  Furthermore, it is also determined whether or not the trend peak position P satisfies a predetermined temporal condition. This is because it is difficult to distinguish purchase behavior by the initial purchaser from purchase behavior by the early minority purchaser when it reaches a peak within a relatively short period of time since the product was put on the market. is there. In such a case, it is possible that the fashion is influenced by external factors such as various media advertisements. The early minority purchasers mentioned here are those consumers who start purchasing behavior by the initial purchaser after a rapid increase in the number of purchasers, etc. It has characteristics that have a great influence on the consumer segment. In the present embodiment, the time condition is that the elapsed period from the market launch time to the fashion peak position P is longer than the average period Tavg, which is the average of the elapsed periods related to a plurality of products included in the sales history information. And In the example shown in FIG. 3A, the average period Tavg is the average value (Ta + Tb + Tc + Td +...) / Total number of products of the elapsed periods Ta, Tb, Tc, Td. ing. This average period Tavg is calculated in advance using sales history information regarding a plurality of products stored in the consumer database 214. The average period Tavg corresponds to the reference period described in the claims.

  When the detected trend peak position P satisfies both the quantitative condition and the temporal condition, the time transition curve is regarded as representing data from which the initial purchaser can be extracted, and the time transition curve The corresponding product corresponding to is determined as the processing target product. In other words, in the example shown in FIG. 3B, for the product X, the trendy peak position Px is in the region indicated by the diagonal line in the figure, which is a region that is larger than the average sales number Navg and longer than the average period Tavg. Therefore, the product is not determined to be a processing target product. On the other hand, for the product Y, the trendy peak position Py is determined to be a processing target product because it is in the area indicated by the diagonal lines in the figure, which is an area longer than the average sales number Navg and longer than the average period Tavg. The

  After narrowing down the products to be processed as described above, next, a predetermined number of consumers who have purchased the product from the market launch time until a predetermined period earlier than the fashion peak position P elapses Extract as a buyer. In this embodiment, based on innovator theory (Everett M. Rogers, 1962), as the predetermined number of consumers, 2.5% of purchasers whose purchase time is early are extracted as initial purchasers. To do.

  The classification based on the innovator theory will be described with reference to FIG. As shown in FIG. 4, the consumer of the product is classified into one of an innovator, an early adopter, an early majority, a late majority, and a laguard in the order of purchase of the product. In the present embodiment, as described above, 2.5% of purchasers among the entire purchasers, that is, consumers corresponding to innovators are extracted as initial purchasers.

  It should be noted that the extraction method is not limited to the innovator theory as described above, but a predetermined number of consumers are extracted in the order of early purchase, or consumers who have purchased a product within a predetermined period from the market launch time are extracted. May be. In this case, the predetermined number of persons and the predetermined period are determined in advance according to, for example, the genre of the product, the price range of the product, and the like, and are stored in the product database 212.

  Next, the control contents executed by the control unit 201 of the server 200 when performing the above-described initial purchaser extraction process will be described with reference to FIG. It is a flowchart showing the said control content.

  In step S100, the control unit 201 accesses the consumer database 214 and executes sales history information acquisition processing for acquiring sales history information of a plurality of consumers for a plurality of products. This detailed content will be described later with reference to FIG.

  In step S200, the control unit 201 introduces a product into the market for all products included in the sales history information based on the sales history information accumulated in the consumer database 214 acquired in step S100. An epidemic time determination process for detecting the epidemic peak position P appearing in the time transition curve of the number of sales from the period for each product is executed. This detailed content will be described later with reference to FIG.

  In step S300, the control unit 201 determines that the trend peak position P of the time transition curve detected in step S200 for all products included in the sales history information is a predetermined time condition and a predetermined number of sales in the time transition curve. The processing target product determination process for determining whether or not the product corresponding to the time transition curve is a processing target product depending on whether or not both of the quantitative conditions are satisfied. This detailed content will be described later with reference to FIG.

  In step S400, the control unit 201, based on the sales history information acquired in step S100, the consumer corresponding to the initial purchaser among the consumers who have purchased the product determined as the processing target product in step S300. An initial purchaser extraction process is executed to extract. This detailed content will be described later with reference to FIG.

  In step S500, the control unit 201 accesses the product database 212, and acquires recommended information related to the product determined as the processing target in step S300. Then, at a predetermined recommended time, the recommended information is output to the mobile terminal 400 on the specific consumer side extracted as corresponding to the initial purchaser in step S400. Thereby, this flowchart is complete | finished.

  In the above, step S100 corresponds to the first procedure described in the claims, step S200 corresponds to the second procedure and functions as a fashion peak position detecting means, and step S300 corresponds to the third procedure. In addition, it functions as a processing target product determination unit, step S400 corresponds to the fourth procedure and functions as an initial purchaser extraction unit, and step S500 corresponds to the sixth procedure and functions as a recommended information output unit.

  The detailed contents of the sales history information acquisition process in step S100 will be described with reference to FIG.

  In step S110, the control unit 201 accesses the consumer database 214 and acquires all product IDs included in the sales history information stored in the database. In step S120, the control unit 201 uses the acquired product ID as a key. The sales history information regarding all the products associated with the ID is acquired.

  In step S130, the control unit 201 classifies the sales history information related to all the products acquired in step S120 for each product ID, and counts the number of sales for each unit period such as one week for each product. Then, data relating to the total result is stored in an appropriate area of the storage unit 210. Thus, this routine is finished.

  The unit period is not limited to one week, but may be one day, two weeks, or one month.

  FIG. 7 and FIG. 8 show an example of the total results obtained by the sales history information acquisition process. In the example shown in FIG. 7 and FIG. 8, the number of sales for each sales week is tabulated for the product whose product ID is 46594.

  The detailed contents of the epidemic time determination process in step S200 will be described with reference to FIG.

  In step S210, the control unit 201 selects and reads out the period data and the number-of-sale data related to one product from the aggregation result data for all products by the sales history information acquisition process stored in the storage unit 210, The period data and sales number data are corrected. More specifically, the control unit 201 accesses the product database 212 and, based on the product ID of the selected product, the genre or price range to which the product belongs and is stored in the product database 212 in advance in association with the ID. Get information about. Then, a correction coefficient that is predetermined and stored in the product database 212 in association with the information is read, and the period data and the sales number data are corrected using the correction coefficient. Specifically, for example, if the product is a car, the price range is high, so the number of sales is multiplied by 10 times, and if the product is daily necessities, the price range is low, so the number of sales is multiplied by 0.5 Perform data correction such as The correction method is not limited to multiplying the coefficients in this way, but may be performed by adding or subtracting a predetermined correction value to the data, or changing the weight of the data for each genre or price range. Etc. may be performed.

  In step S220, the control unit 201 smoothes the period data and sales number data corrected in step S210 using a known method of smoothing time-series data such as a moving average.

  In step S230, the control unit 201 detects the trendy peak position P where the number of sales is maximized based on the data smoothed in step S220, and sets the maximum number of sales corresponding to the trendy peak position P as the product trend index. Set.

  In step S240, the control unit 201 sets the time corresponding to the fashion peak position P detected in step S230 as the product fashion time. This routine is completed as described above. In addition, the control part 201 performs the process of the said step S210-step S240 about all the goods included in sales history information in the fashion time determination process of the above step S200. At this time, the product trend index data and the product trend time data set for each product are stored in an appropriate area of the storage unit 210.

  The step S210 corresponds to the fifth procedure described in the claims.

  An example of the period data and sales number data smoothed by the fashion period determination process will be described with reference to FIGS. 10 and 11. The examples shown in FIGS. 10 and 11 are obtained by smoothing the number of sales for each sales week for the product whose product ID is 46594 shown in FIGS. 7 and 8 described above. As shown in FIGS. 10 and 11, in this example, data corresponding to the maximum peak value “792.7143” and the sales week “32” is detected as the trend peak position P and corresponds to the trend peak position P. The maximum number of sales “792.7143” is set as the product trend index of the product ID 46594, and the sales week “32” representing the corresponding time is set as the product trend time of the product ID 46594.

  The detailed content of the process target product determination process in step S300 will be described with reference to FIG.

  In step S600, the control unit 201 selects one product from all the products included in the sales history information, and an average period and an average serving as an index for determining whether the product is a processing target product. Processing target product index calculation processing for calculating the number of sales is executed. At this time, the control unit 201 stores the calculated average period data and average sales quantity data for each product in an appropriate area of the storage unit 210. This detailed content will be described later with reference to FIG.

  In step S310, the control unit 201 reads out and acquires the average period data and the average sales quantity data calculated in the processing target product index calculation process in step S600 from the storage unit 210. Then, it is set as a quantitative condition that the product trend index set in the above-described step S230 corresponding to the selected product is larger than the acquired average sales number. Further, it is set as a time condition that the elapsed period from the market launch time of the selected product to the product fashion time set in step S240 is longer than the acquired average period.

  In step S310, it may be set as a temporal condition that the elapsed period is longer than the average period by the standard deviation of the time-series data. Further, it may be set as a quantitative condition that it is longer than the average number of sales by the standard deviation of the time series data.

  In step S320, the control unit 201 determines whether the quantitative condition set in step S310 is satisfied. That is, the control unit 201 acquires product trend index data corresponding to the selected product from the storage unit 210 and compares it with the average sales number data. If the merchandise trend index is equal to or less than the average number of sales, the determination is not satisfied because the quantitative condition is not satisfied, and this routine is terminated. On the other hand, if the product trend index of the product is greater than the average number of sales, the determination is satisfied because the quantitative condition is satisfied, and the process proceeds to step S330.

  In step S330, the control unit 201 determines whether or not the temporal condition set in step S310 is satisfied. That is, the control unit 201 acquires product trend time data corresponding to the selected product from the storage unit 210 and compares it with the average period data. If the elapsed period from the market launch time of the product to the product fashion time is shorter than the average period, the determination is not satisfied because the time condition is not satisfied, and this routine is terminated. On the other hand, when the elapsed period is longer than the average period, the determination is satisfied because the temporal condition is satisfied, and the process proceeds to step S340.

  In step S340, the control unit 201 determines that the product is a processing target product, and sets a processing target product flag corresponding to the product to “TRUE” indicating that the product is a processing target product. The control unit 201 stores the information of the flag in an appropriate area of the storage unit 210 in association with the period data and the sales quantity data of the corresponding product. Thus, this routine is finished. In addition, the control part 201 performs the process of said step S600-step S340 about all the goods contained in sales history information in the process target goods determination process of the above step S300.

  The detailed contents of the processing target product index calculation process in step S600 will be described with reference to FIG.

  In step S700, the control unit 201 selects one product from all the products included in the sales history information, and stores product trend index data and product trend time data relating to a plurality of products belonging to the same genre as the product. The product sales history sample acquisition process acquired from the storage unit 210 is executed. This detailed content will be described later with reference to FIG.

  In step S610, the control unit 201 calculates an average period that is an average value thereof based on the product trend time data related to the plurality of products acquired in step S700, and stores the average period in the storage unit 210 as average period data.

  In step S620, the control unit 201 calculates an average sales number that is an average value based on the product trend index data related to the plurality of products acquired in step S700, and stores the average sales number in the storage unit 210 as average sales number data. To do. This routine is completed as described above. In addition, the control part 201 performs the process of the said step S700-step S620 about all the goods contained in sales history information in the process target goods parameter | index calculation process of step S600.

  The detailed contents of the process sales history sample acquisition process in step S700 will be described with reference to FIG.

  In step S <b> 710, the control unit 201 accesses the product database 212, and stores in the product database 212 that is determined in advance in association with the ID based on the product ID of the product selected from all the products included in the sales history information. The information regarding the genre to which the product concerned belongs is acquired.

  In step S720, the control unit 201 belongs to the same genre as the product genre related to the information acquired in step S710 from the product trend index data and the product trend time data related to all products stored in the storage unit 210. Product trend index data and product trend time data relating to a plurality of products are extracted, and the extracted product trend index data and product trend time data are stored in appropriate areas of the storage unit 210. This routine is completed as described above. Note that the control unit 201 performs the processing of step S710 and step S720 on all the products included in the sales history information in the product sales history sample acquisition processing of step S700 described above.

  In the above, the product trend index data and product trend time data of products of the same genre as the selected product are extracted as index samples. However, the product trend index data and products of all products are not limited to the product genre. Trend data may be used as an index sample. Moreover, you may make it extract the goods of the same price range instead of the same genre, for example, you may extract the goods with the same sex and age group of the consumer of the said goods. In addition, for example, a high-order product in the sales number ranking, a product whose ranking has risen rapidly, or a non-tie-up product may be extracted.

  The detailed contents of the initial purchaser extraction process in step S400 will be described with reference to FIG.

  In step S410, the control unit 201 selects one product from the total result data for all products stored in the storage unit 210, reads the period data and the sales number data, and stores the data in association with the data. It is determined whether or not the processing target product flag is “TRUE”. If the processing target product flag is not “TRUE”, the determination is not satisfied and this routine is terminated. On the other hand, when the processing target product flag is “TRUE”, the determination is satisfied, and the routine goes to Step S420.

  In step S420, the control unit 201 acquires period data and sales number data relating to a product whose processing target product flag is “TRUE” from the storage unit 210, and from the market launch time based on the data from the trend peak position P. A predetermined number of consumers who have purchased the product before the elapse of an early predetermined period are extracted as initial purchasers. In the present embodiment, as described above, based on the innovator theory, 2.5% of purchasers whose purchase time is early among all purchasers are extracted as initial purchasers. Thus, this routine is finished.

  Note that the control unit 201 performs the processing of step S410 and step S420 on all the products included in the sales history information in the above-described initial purchaser extraction processing of step S400.

  According to the embodiment described above, the following effects are obtained. That is, in general, when a new product is introduced into the market, the number of sales of the product increases slowly over time, then increases rapidly at a certain time, and the number of sales reaches a peak at a certain time thereafter. When this peak is reached, the number of items sold increases as the products become more popular. After that, the number of sales of the product gradually decreases, and after a certain period, it almost approaches zero. In the general consumer purchase characteristics expressed in this way, consumers are classified as early purchasers, early minority purchasers, early majority purchasers, late majority purchasers, and final purchasers in the order of purchase of goods. being classified. These correspond to innovators, early adopters, early majority, late majority, and lagard, respectively, in the classification based on the innovator theory shown in FIG.

  A consumer corresponding to an initial purchaser has the characteristic of proceeding with his own will and purchasing a new product without being affected by external influences such as advertisements of various media. Therefore, in the general consumer purchase characteristics expressed as described above, the initial purchaser takes a purchase action from the time immediately after the product is introduced to the market to immediately before the number of sales increases rapidly. . Thereafter, the number of sales starts to increase rapidly, and an early minority purchaser takes a purchase action for a while. Furthermore, early purchasers take action before the increase in sales reaches its peak.

  In this embodiment, when extracting an initial purchaser from purchase trends of a plurality of consumers for a plurality of products, first, data that can extract the initial purchaser is found. That is, when the sales history information of a plurality of consumers is acquired from the consumer database 214 in the sales history information acquisition process in step S100, the trend peak position P of the time transition curve is determined for each product in the trend time determination process in step S200. Is detected. At this time, the data to be processed is preferably data on a product that has obtained a certain number of sales numbers and has a relatively large peak sales number, based on the purchase behavior characteristics of the initial purchaser as described above. This is because if the number of sales at the peak is relatively small, it is difficult to distinguish purchase behavior by enthusiasts from purchase behavior by initial purchasers or early minority purchasers.

  Therefore, in the process target product determination process in step S300, first, it is determined whether or not the trend peak position P satisfies a predetermined quantitative condition. In addition, when a peak is reached only after a relatively short period of time has passed since the product was put on the market, it is difficult to distinguish between purchase behavior by the initial purchaser and purchase behavior by the early minority purchaser. In such a case, the fashion may be influenced by external factors such as various media advertisements. Therefore, in the processing target product determination process, it is also determined whether the trend peak position P satisfies a predetermined temporal condition. When the trend peak position P satisfies both the quantitative condition and the temporal condition, the time transition curve is regarded as representing data from which the initial purchaser can be extracted, and the correspondence corresponding to the time transition curve is taken. The product is determined to be a processing target product.

  After narrowing down the processing target as described above, in the initial purchaser extraction process in step S400, the consumer who has purchased the product from the market launch time until a predetermined period earlier than the fashion peak position P elapses. Extract as an initial purchaser. As described above, new products that are expected to become popular in the future regardless of external factors such as advertisements can be easily and reliably introduced to the market from the purchase trends of multiple consumers for multiple products. Then, it is possible to find a consumer corresponding to an initial purchaser who tends to purchase early.

  In the present embodiment, in particular, in the process target product determination process in step S300, the time condition is that the elapsed period from the market launch time to the product fashion time is longer than the average period. Thereby, it is possible to exclude data that is difficult to distinguish between the purchase behavior of the initial purchaser and the purchase behavior of the early minority purchaser from the processing target. The quantitative condition is that the product trend index at the trend peak position P is larger than the average number of sales. Thereby, it is possible to exclude data that is difficult to distinguish between the purchase behavior of the initial purchaser and the purchase behavior of the mania from the processing target.

  In the present embodiment, in particular, when determining whether or not a specific product is a processing target product, an average serving as an index for the determination using purchase histories of a plurality of products belonging to the same genre as the specific product. Calculate period and average sales. In other words, in general, consumers often take a relatively similar common consumption behavior for multiple products of the same genre or multiple products of the same price range. By doing so, the initial purchaser can be extracted with higher accuracy.

  In particular, in the present embodiment, in consideration of the situation where the purchase trend of a product by a consumer may be influenced by the genre or price range to which the product belongs, a correction coefficient is associated with the genre or price range to which the product belongs in advance. Defined and stored in the product database 212. In step S210, period data and sales quantity data relating to one product are selected from the total result data for all products obtained by the sales history information acquisition process in step S100, and the data is corrected using the correction coefficient described above. To do. In subsequent step S230, the epidemic peak position P is detected using the period data and sales number data corrected as described above. Thereby, an initial purchaser can be extracted with higher accuracy.

  In this embodiment, in particular, a product is recommended for the consumer group that is positioned as the extracted initial purchaser. Thereby, the number of sales can be increased at an early stage immediately after the product is put on the market. As a result, it is possible to stimulate the consumer desire to purchase after the early minority purchaser and further increase the number of sales.

  The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and technical idea of the present invention. Hereinafter, such modifications will be described in order.

(1) Case where two fashion peak positions are formed In the above embodiment, the case where a single fashion peak position P is formed on the time transition curve has been described as an example. , The number of buyers or sales increased rapidly due to various media advertisements, etc., and after that, after the temporary trend ended, purchases were made by consumers who were not affected by such trends, There are some that the number of purchasers increases again and reaches the second peak. This modification is an example of a product having such a time transition curve.

  With reference to FIG. 16, a method for extracting a consumer corresponding to an initial purchaser from consumers who have purchased a product determined as a processing target product based on sales history information in the present modification will be described.

  Also in this modified example, as in the above-described embodiment, first, a product for which an initial purchaser can be extracted is found. That is, the control unit 201 acquires sales history information of a plurality of consumers from the consumer database 214 in the same manner as in the above-described embodiment, and represents the number of products purchased or the change over time in sales for each product. The trend peak position P of the time transition curve is detected. At this time, when two fashion peak positions exist, both fashion peak positions are detected. In the example shown in FIG. 16, the first trendy peak position P1 (T1, N1) and the second trendy peak position P2 (T2, N2) are detected.

  Then, it is determined whether or not the first trend peak position P1 satisfies the quantitative condition and the temporal condition for each product. If the first epidemic peak position P1 does not satisfy at least one of the quantitative condition and the temporal condition, it is determined whether or not the second epidemic peak position P2 satisfies the quantitative condition and the temporal condition. When the second epidemic peak position P2 satisfies both the quantitative condition and the temporal condition, the first epidemic peak position P1, the first epidemic peak position P1, and the second epidemic peak are further provided. It is determined whether or not the difference in the sales quantity from the decline peak position P3 where the sales quantity is the smallest with respect to the position P2 is larger than a predetermined reference quantity. By making such a determination, products that have been purchased by consumers who are not affected by the fashion after the end of the fashion due to external influences such as media advertisements are excluded from the processing of the temporary decrease in the number of sales. This is because it can be reliably processed. When the decline peak position P3 satisfies the above positional condition, the time transition curve is regarded as representing data from which the initial purchaser can be extracted, and the corresponding product corresponding to the time transition curve is determined as the processing target product. To do.

  In the example shown in FIG. 16, since the sales number N1 corresponding to the first trendy peak position P1 is larger than the average sales number Navg and the elapsed period T1 is shorter than the average period Tavg, the first trendy peak position P1 is a quantity. But it does not meet the time condition. Therefore, it is determined whether or not the sales number N2 at the second trendy peak position P2 is larger than the average sales number Navg, and whether or not the elapsed period T2 at the second trendy peak position P2 is longer than the average period Tavg. To do. In addition, the calculation method of the average sales number Navg and the average period Tavg in this modification is the same as that of the said embodiment. In addition, here, the first epidemic peak position P1 is an example in which the quantitative condition is satisfied but the temporal condition is not satisfied. On the contrary, the temporal condition is satisfied but the quantitative condition is not satisfied. The two conditions may not be satisfied.

  In this example, the second epidemic peak position P2 is in the region indicated by the diagonal line in the figure, which is an area longer than the average sales number Navg and longer than the average period Tavg. P2 satisfies both quantitative and temporal conditions. The difference Ndef in the number of sales between the first trend peak position P1 and the decline peak position P3 where the number of sales is the smallest between the first trend peak position P1 and the second trend peak position P2 is shown in the figure. Is greater than a predetermined reference quantity No, and the decline peak position P3 satisfies the positional condition. Therefore, the product corresponding to the time transition curve shown in FIG. 16 is determined as the secondary trendy product. The reference quantity No corresponds to the second reference quantity described in the claims.

  Once the products to be processed have been narrowed down as described above, in this modification, the predetermined number of people who have purchased the products from the decline peak position P3 to the passage of a predetermined period earlier than the second trendy peak position P2. Are extracted as initial buyers. In the present embodiment, as the predetermined number of consumers, 2.5% of purchasers at the early purchase time among the purchasers at the peak of the second epidemic peak position P2 classified as innovators based on innovator theory. Are extracted as initial purchasers.

  Next, the contents of control executed by the control unit 201 of the server 200 according to this modification when performing the above-described initial purchaser extraction process will be described. Note that the sales history information acquisition process in step S100 and the recommended information output procedure in step S500 are the same as those in the above-described embodiment, and thus the description thereof will be omitted. The determination process and the initial purchaser extraction process will be described.

  With reference to FIG. 17, the detailed contents of the fashion period determination process in step S200A in the present modification will be described.

  Steps S210 to S240 are the same as those in FIG.

  In step S250, the control unit 201 performs correction within the predetermined period after the first trendy peak position P1 detected in step S230 based on the period data and sales number data corrected in step S210 and smoothed in step S220. In addition, it is determined whether or not the second epidemic peak position P2 has been detected. If the second trend peak position P2 is not detected, the determination is not satisfied and this routine is terminated. On the other hand, if the second epidemic peak position P2 is detected, the determination is satisfied, and the routine goes to Step S260.

  In step S260, the control unit 201 sets the number of sales N2 corresponding to the detected second trendy peak position P2 as the second merchandise trend index, and sets the corresponding elapsed period T2 as the second merchandise trend time.

  In step S270, the control unit 201 detects the decline peak position P3 where the sales number is the minimum between the first fashion peak position P1 and the second fashion peak position P2, based on the period data and the sales number data. . This routine is completed as described above.

  In addition, the control part 201 performs the process of said step S210-step S270 about all the goods contained in sales history information in the fashion time determination process of the above step S200A. At this time, the second product fashion index data set for each product, the second product fashion time data, and the data of the elapsed period T3 and the sales number P3 related to the detected decline peak position P3 are appropriately stored in the storage unit 210. Store in the area.

  The detailed content of the process target product determination process in step S300A in this modification will be described with reference to FIG.

  Steps S600 to S340 are the same as those in FIG. On the other hand, if the trend peak position is a product that satisfies the quantitative condition but does not satisfy the temporal condition, the determination in step S330 is not satisfied and the process proceeds to step S350.

  In step S350, the control unit 201 determines whether or not the second trendy peak position satisfies the quantitative condition set in step S310. That is, the control unit 201 obtains the second product trend index data corresponding to the second trend peak position of the selected product from the storage unit 210, compares it with the average sales number data, and the second product trend time. Data is acquired from the storage unit 210 and compared with the average period data. If the second product trend index is greater than the average number of sales and the elapsed period from the market launch time to the second product trend time is longer than the average period, the judgment is satisfied because the quantitative condition and the temporal condition are satisfied. Then, the process proceeds to step S360. On the other hand, if any of the conditions is not satisfied, the determination is not satisfied and this routine is terminated.

  In step S360, the control unit 201 acquires data related to the decline peak position detected in step S270 described above from the storage unit 210, and calculates the difference Ndef in the number of sales between the first trendy peak position and the decline peak position. Then, it is determined whether or not the difference Ndef in the number of sales is greater than a predetermined reference quantity No. If the decline peak position satisfies the positional condition, the process proceeds to step S370. On the other hand, when the decline peak position does not satisfy the positional condition, this routine ends.

  In step S370, the control unit 201 determines that the product is a secondary trendy product, and sets the secondary trend flag corresponding to the product to “TRUE” indicating that it is a secondary trendy product. The control unit 201 stores the information of the flag in an appropriate area of the storage unit 210 in association with the period data and the sales quantity data of the corresponding product. Thereafter, the process proceeds to step S340 described above, the processing target product flag corresponding to the product is set to “TRUE” indicating that the product is a processing target product, and this routine is ended. In addition, the control part 201 performs the process of the said step S600-step S370 about all the products included in sales history information in the process target product determination process of the above step S300A.

  The detailed contents of the initial purchaser extraction process in step S400A in the present modification will be described with reference to FIG.

  Step S410 is the same as that in FIG. 15 described above, and the control unit 201 selects one product from the aggregation result data for all products stored in the storage unit 210, reads the period data and the sales number data, It is determined whether or not the processing target product flag stored in association with the data is “TRUE”. If the processing target product flag is not “TRUE”, the determination is not satisfied and this routine is terminated. On the other hand, if the processing target product flag is “TRUE”, the determination is satisfied, and the routine goes to Step S415.

  In step S415, the control unit 201 determines whether or not the second trend flag stored in association with the period data and sales quantity data of the selected one product is “TRUE”. If the second trendy flag is not “TRUE”, the determination is not satisfied and the process proceeds to step S420, and the processing target product flag is “TRUE” and the second trendy flag is set as in FIG. Period data and sales number data related to products that are not “TRUE” are acquired from the storage unit 210, and based on the data, the products are purchased between the market launch time and a predetermined period earlier than the trendy peak position. A predetermined number of consumers are extracted as initial purchasers. In the present embodiment, as described above, based on the innovator theory, 2.5% of purchasers whose purchase time is early among all purchasers are extracted as initial purchasers.

  On the other hand, if the secondary trend flag is “TRUE”, the determination is satisfied, and the routine goes to Step S417. In this step S417, the control unit 201 acquires period data and sales number data related to a product whose processing target product flag and second trend flag are both “TRUE” from the storage unit 210, and based on the data, A predetermined number of consumers who have purchased the product during the period from the decline peak position to the passage of a predetermined period earlier than the second trendy peak position are extracted as initial purchasers. In this case as well, based on the innovator theory, 2.5% of purchasers with an early purchase time are extracted as initial purchasers among the purchasers in the mountain portion having the second epidemic peak position at the top. Thus, this routine is finished.

  In addition, in the initial purchaser extraction process in step S400A described above, the control unit 201 performs the processes in steps S410 to S420 for all products included in the sales history information.

  In the above, step S200A corresponds to the second procedure described in the claims and functions as a fashion peak position detection unit, and step S300A corresponds to the third procedure and functions as a processing target product determination unit. Corresponds to the fourth procedure and functions as an initial purchaser extracting means.

  In the modification described above, consumers who have purchased the product are extracted from the decline peak position until a predetermined period earlier than the second trendy peak position elapses. Thereby, it is possible to easily and reliably find a consumer corresponding to the initial purchaser from among purchase trends of a plurality of consumers for a plurality of products. In addition, since the difference in the number of purchasers between the first trendy peak position and the decline peak position is larger than a predetermined reference quantity No, it is possible to reduce the number of purchasers only temporarily. Products that are not subject to processing and that have been purchased by a consumer group that is not affected by the fashion after the end of the fashion due to external influences such as media advertisements can be reliably treated.

(2) When the quantitative conditions are relaxed Even if the epidemic peak position P does not satisfy the quantitative conditions, it has a certain epidemic degree in terms of the cumulative value of the number of sales, and if the temporal conditions are satisfied, it is processed It may be regarded as a target product. This modification corresponds to such a case, and is a modification that relaxes the quantitative condition.

  With reference to FIG. 20, a method for relaxing the quantitative condition of this modification will be described.

  In the example shown in FIG. 20A, the elapsed period Tp corresponding to the epidemic peak position P is longer than the average period Tavg and the sales number Np is less than the average sales number Navg. Is satisfied, but the quantitative conditions are not satisfied. Therefore, as it is, the time transition curve is not considered to represent data from which the initial purchaser can be extracted, and the product is not determined to be a processing target product.

  Therefore, in this modification, as shown in FIG. 20B, the cumulative sales number N ′ is calculated based on the sales history information related to the selected product, and sales related to a plurality of products stored in the consumer database 214 are performed. The average cumulative sales number N′avg of a plurality of products calculated and stored in advance using the history information is acquired from the consumer database 214, and the average cumulative sales number N′avg and the trend peak position P at the time corresponding to the average cumulative sales number N′avg are acquired. The cumulative sales number N′p is compared. When the cumulative sales number N′p is larger than the average cumulative sales number N′avg as shown in FIG. 20B, the time transition curve is regarded as representing data from which the initial purchaser can be extracted. The corresponding product corresponding to the time transition curve is determined as the processing target product. By doing in this way, it becomes possible to set the product data that could not be processed because the initial quantitative condition is not satisfied, and the quantitative condition can be relaxed.

  Note that the average cumulative sales number N′avg is periodically calculated and updated in accordance with the update of the sales history information regarding a plurality of products stored in the consumer database 214. The average cumulative sales number N′avg corresponds to the third reference quantity described in the claims.

  The detailed content of the process target product determination process in step S300B in this modification will be described with reference to FIG.

  Steps S600 and S310 are the same as those in FIG.

  In step S325, the control unit 201 determines whether or not the temporal condition set in step S310 is satisfied. That is, the control unit 201 acquires product trend time data corresponding to the selected product from the storage unit 210 and compares it with the average period data. If the elapsed period from the market launch time to the product fashion time is shorter than the average period, the determination is not satisfied because the time condition is not satisfied, and this routine is terminated. On the other hand, if the elapsed period is longer than the average period, the determination is satisfied because the temporal condition is satisfied, and the routine goes to Step S335.

  In step S335, the control unit 201 determines whether the quantitative condition set in step S310 is satisfied. That is, the control unit 201 acquires product trend index data corresponding to the selected product from the storage unit 210 and compares it with the average sales number data. If the merchandise trend index is greater than the average number of sales, the determination is satisfied because the quantitative condition is satisfied, and the process proceeds to step S340, and the product is determined to be a processing target product as in FIG. The process target product flag corresponding to the product is set to “TRUE” indicating that the product is a process target product. Then, this routine ends. On the other hand, when the commodity fashion index is equal to or less than the average number of sales, the determination is not satisfied because the quantitative condition is not satisfied, and the process proceeds to step S380.

  In step S380, the control unit 201 calculates the cumulative sales number N ′ based on the sales history information related to the selected product.

  In step S <b> 390, the control unit 201 uses the consumer database 214 to calculate the average cumulative sales number N′avg of the plurality of products calculated and stored in advance using the sales history information regarding the plurality of products stored in the consumer database 214. The average cumulative sales number N′avg is compared with the cumulative sales number N′p at the time corresponding to the trendy peak position P calculated in step S380, and the cumulative sales number N′p It is determined whether the number of sales is greater than N′avg. If the product sales number N′p is larger than the average cumulative sales number N′avg, the process proceeds to step S340 described above, and the processing target product flag corresponding to the product is set to “TRUE” indicating that the product is a processing target product. Set and exit this routine. On the other hand, if the product sales number N′p is equal to or less than the average cumulative sales number N′avg, the determination is not satisfied and this routine is terminated.

  In addition, the control part 201 performs the control content similar to the said embodiment except the above step S300B. The step S300B corresponds to a third procedure described in the claims and functions as a processing target product determination unit.

  In the modification described above, even when the trend peak position P does not satisfy the quantitative condition, the cumulative value N′p of the sales number from the market launch time to the trend peak position P is higher than the average cumulative sales number N′avg. If there are many, the product corresponding to the time transition curve is determined as the product to be processed. Thereby, the quantitative conditions can be relaxed, and the product as described above can be determined as the product to be processed.

(3) Case where initial purchaser is extracted based on singing trend in karaoke In the above embodiment, the content of extracting a consumer based on sales history information of a product has been described. Music data sung by consumers is also included. That is, when a consumer who is a karaoke user sings in a karaoke room, the distribution of music data by a music distribution company to a karaoke device is equivalent to the sale of the music data, and for consumers to purchase the music data. Equivalent to. Accordingly, the consumer corresponding to the initial purchaser is extracted from the consumers who have purchased the music determined as the music to be processed based on the sales history information of the music data corresponding to the singing history at such karaoke. May be.

  The overall configuration of the product recommendation system 1 ′ that realizes such a system will be described with reference to FIG.

  In FIG. 22, the product recommendation system 1 ′ includes a karaoke device 100 installed in a karaoke room KR, a server 200, a base station 300, a portable terminal 400 carried by each consumer, and a karaoke device via a network NW. 100 has a music distribution company 500 that distributes music data. The configurations of the server 200, the base station 300, and the mobile terminal 400 are the same as those in the above-described embodiment.

  The karaoke device 100 includes a karaoke device main body 110, a display device 120, and a microphone 115. The karaoke apparatus main body 110 reproduces music data distributed from the music distribution company 500 via the network NW. The display device 120 displays a video related to singing when the music data is reproduced. The microphone 115 is connected to the karaoke apparatus main body 110 and is used for singing by consumers. The karaoke apparatus main body 110 and the server 200 are connected so as to be able to communicate with each other via a network NW. When a consumer reproduces music data in the karaoke room KR, sales history information (purchase history information) of the music data. Is stored in the consumer database 214 of the server 200 for each piece of music data.

  The stored contents of the consumer database 214 in this case are the same as those shown in FIG. That is, in this modification, the product ID shown in FIG. 2 corresponds to the identification information of the music data, and the sales date / time of the product included in the sales history stored for each product ID corresponds to the distribution date / time of the music data. . The user ID is identification information of a consumer who purchased the music data, that is, a karaoke user.

  Therefore, the method of extracting the initial purchaser in this modification is the same as that in the above-described embodiment. That is, if the control part 201 acquires the sales history information equivalent to the singing history in karaoke from the consumer database 214, the control part 201 represents the time change of the sales number of the goods equivalent to the number of distribution of music data for every goods. The trend peak position P of the time transition curve is detected. Then, it is determined whether or not the detected trend peak position P satisfies predetermined quantitative conditions and temporal conditions. When the trend peak position P satisfies both the quantitative condition and the temporal condition, the time transition curve is regarded as representing data from which the initial purchaser can be extracted, and the correspondence corresponding to the time transition curve is taken. The product is determined to be a processing target product. Once the products to be processed are narrowed down in this way, a predetermined number of people who have purchased the products between the market launch time corresponding to the music data distribution start time and a predetermined period earlier than the fashion peak position P have passed. A consumer is extracted as an initial purchaser.

  Also by the modified example described above, the same effect as that of the above-described embodiment can be obtained.

  In addition, in the above, the arrow shown in each figure of FIG. 1, FIG. 22 shows an example of the flow of a signal, and does not limit the flow direction of a signal.

  The flowcharts shown in FIGS. 5, 6, 9, etc. do not limit the present invention to the procedures shown in the above-mentioned flow, but add, delete, or order the procedures within the scope of the gist and technical idea of the invention. May be changed.

  In addition to those already described above, the methods according to the above-described embodiments and modifications may be used in appropriate combination.

  In addition, although not illustrated one by one, the present invention is implemented with various modifications within a range not departing from the gist thereof.

200 servers (first information terminal, initial purchaser extraction system, initial purchaser recommendation system)
214 Consumer database 400 Mobile terminal (second information terminal)
Navg average sales volume (first reference quantity)
N'avg average cumulative sales (third reference quantity)
No Standard quantity (second standard quantity)
N'p Cumulative number of sales (cumulative value)
P1 First epidemic peak position P2 Second epidemic peak position P3 Decay peak position Tavg Average period (reference period)

Claims (8)

A consumer database that stores, for each product, purchase history information that associates product identification information with purchase histories of a plurality of consumers for the product in accordance with operation input information in the first information terminal on the extractor side And obtaining the purchase history information of a plurality of consumers for a plurality of products,
Based on the purchase history information accumulated in the consumer database acquired in the first step, at least one appearing in a time transition curve of the number of purchasers or the number of sales from the market launch time when the product was introduced to the market. A second procedure for detecting one fashion peak position for each product;
When the trend peak position of the time transition curve acquired in the second procedure satisfies both a predetermined time condition in the time transition curve and a predetermined quantitative condition regarding the number of purchasers or the number of sales , A third procedure for determining a product corresponding to the time transition curve as a processing target product,
Based on the purchase history information acquired in the first procedure, the product determined as the product to be processed in the third procedure until the predetermined time elapses from the market launch time before the trendy peak position. A fourth step of extracting consumers purchased in the meantime as consumers corresponding to the initial purchaser ;
Have
The third procedure includes
As the time condition, an elapsed period from the market launch time to the trendy peak position is longer than a reference period calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database, In addition, as the quantitative condition, the number of purchasers or the number of sales at the trendy peak position is greater than a first reference quantity calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database. An initial purchaser extraction processing method characterized by determining a product corresponding to the time transition curve as a processing target product when there are many . The third procedure includes
A specific product belonging to a specific genre or a specific price range,
The elapsed period corresponding to the specific product is based on the reference period calculated in advance using the purchase history information related to the specific product of the specific genre or the specific price range stored in the consumer database. Long, and
Using the purchase history information regarding the plurality of products in the specific genre or the specific price range, the number of purchasers or sales at the trendy peak position corresponding to the specific product is stored in the consumer database. If more than a pre-calculated first reference quantity, an initial purchaser extraction processing method according to claim 1, wherein determining that processed products of the specific product. The purchase history information accumulated in the consumer database acquired in the first procedure is determined in advance in association with the genre or price range to which the product belongs, and using a correction coefficient stored in the product database. Further comprising a fifth step of correcting,
The second procedure includes
The initial purchaser extraction processing method according to claim 1 or 2 , wherein the at least one trendy peak position is detected for each product based on the purchase history information corrected in the fifth procedure. . In the case where a plurality of the fashion peak positions are detected by the second procedure,
The third procedure includes
The first epidemic peak position does not satisfy at least one of the temporal condition and the quantitative condition;
The second epidemic peak position satisfies both the temporal condition and the quantitative condition, and the first epidemic peak position, the first epidemic peak position, and the second epidemic peak position. If the difference in the number of purchasers or sales from the decline peak position where the number of purchasers or sales is the smallest is greater than the second reference quantity, the product corresponding to the time transition curve is Judged as a product to be processed
The fourth procedure includes
2. The consumer purchased from the decline peak position until a predetermined time earlier than the second epidemic peak position elapses is extracted as a consumer corresponding to the initial purchaser. The initial purchaser extraction processing method according to any one of claims 3 to 3 . The third procedure includes
When the trend peak position satisfies the time condition but does not satisfy the quantitative condition, the cumulative value of the number of purchasers or sales from the market launch time to the trend peak position is a third criterion. The initial purchaser extraction processing method according to any one of claims 1 to 4 , wherein when the number is larger than the quantity, the product corresponding to the time transition curve is determined as a processing target product. A consumer database that stores, for each product, purchase history information that associates product identification information with purchase histories of a plurality of consumers for the product in accordance with operation input information in the first information terminal on the extractor side And obtaining the purchase history information of a plurality of consumers for a plurality of products,
Based on the purchase history information accumulated in the consumer database acquired in the first step, at least one appearing in a time transition curve of the number of purchasers or the number of sales from the market launch time when the product was introduced to the market. A second procedure for detecting one fashion peak position for each product;
When the trend peak position of the time transition curve acquired in the second procedure satisfies both a predetermined time condition in the time transition curve and a predetermined quantitative condition regarding the number of purchasers or the number of sales , A third procedure for determining a product corresponding to the time transition curve as a processing target product,
Based on the purchase history information acquired in the first procedure, the product determined as the product to be processed in the third procedure until the predetermined time elapses from the market launch time before the trendy peak position. A fourth step of extracting consumers purchased in the meantime as consumers corresponding to the initial purchaser;
A sixth procedure for outputting recommended information of a predetermined product at a predetermined recommended time to the second information terminal on the specific consumer side extracted as corresponding to the initial purchaser in the fourth procedure ;
Have
The third procedure includes
As the time condition, an elapsed period from the market launch time to the trendy peak position is longer than a reference period calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database, In addition, as the quantitative condition, the number of purchasers or the number of sales at the trendy peak position is greater than a first reference quantity calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database. An initial purchaser recommendation method characterized in that if there are many products, a product corresponding to the time transition curve is determined as a product to be processed . A consumer database that stores, for each product, purchase history information that associates product identification information with purchase history of a plurality of consumers for the product;
Based on the purchase history information accumulated in the consumer database, at least one trend peak position appearing in a time transition curve of the number of purchasers or the number of sales from the time when the product was introduced to the market is determined for each product. Fashion peak position detection means to detect in,
When the trend peak position of the time transition curve detected by the trend peak position detection means satisfies both a predetermined temporal condition and a predetermined quantitative condition regarding the number of purchasers or sales in the time transition curve The processing target product determination means for determining the product corresponding to the time transition curve as the processing target product,
A consumer who purchases the product determined as the processing target product by the processing target product determination means based on the purchase history information from the market launch time until a predetermined time earlier than the trendy peak position elapses. Initial buyer extracting means for extracting as a consumer corresponding to the initial buyer ,
Have
The processing target product determination means includes
As the time condition, an elapsed period from the market launch time to the trendy peak position is longer than a reference period calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database, In addition, as the quantitative condition, the number of purchasers or the number of sales at the trendy peak position is greater than a first reference quantity calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database. An initial purchaser extraction system characterized in that if there are many products, a product corresponding to the time transition curve is determined as a product to be processed . A consumer database that stores, for each product, purchase history information that associates product identification information with purchase history of a plurality of consumers for the product;
Based on the purchase history information accumulated in the consumer database, at least one trend peak position appearing in a time transition curve of the number of purchasers or the number of sales from the time when the product was introduced to the market is determined for each product. Fashion peak position detection means to detect in,
When the trend peak position of the time transition curve detected by the trend peak position detection means satisfies both a predetermined temporal condition and a predetermined quantitative condition regarding the number of purchasers or sales in the time transition curve The processing target product determination means for determining the product corresponding to the time transition curve as the processing target product,
A consumer who purchases the product determined as the processing target product by the processing target product determination means based on the purchase history information from the market launch time until a predetermined time earlier than the trendy peak position elapses. Initial buyer extracting means for extracting as a consumer corresponding to the initial buyer,
Recommended information output means for outputting recommended information of a predetermined product at a predetermined recommended time to the second information terminal on the specific consumer side extracted as corresponding to the initial purchaser by the initial purchaser extracting means and,
Have
The processing target product determination means includes
As the time condition, an elapsed period from the market launch time to the trendy peak position is longer than a reference period calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database, In addition, as the quantitative condition, the number of purchasers or the number of sales at the trendy peak position is greater than a first reference quantity calculated in advance using the purchase history information regarding a plurality of products stored in the consumer database. An initial purchaser recommendation system characterized in that if there are many products, a product corresponding to the time transition curve is determined as a product to be processed .

Images