WO2018055660A1 - Purchase recommendation system, purchase recommendation method, and program - Google Patents

Purchase recommendation system, purchase recommendation method, and program Download PDF

Info

Publication number
WO2018055660A1
WO2018055660A1 PCT/JP2016/077630 JP2016077630W WO2018055660A1 WO 2018055660 A1 WO2018055660 A1 WO 2018055660A1 JP 2016077630 W JP2016077630 W JP 2016077630W WO 2018055660 A1 WO2018055660 A1 WO 2018055660A1
Authority
WO
WIPO (PCT)
Prior art keywords
purchase
time
recommendation
purchaser
purchased product
Prior art date
Application number
PCT/JP2016/077630
Other languages
French (fr)
Japanese (ja)
Inventor
俊二 菅谷
Original Assignee
株式会社オプティム
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社オプティム filed Critical 株式会社オプティム
Priority to PCT/JP2016/077630 priority Critical patent/WO2018055660A1/en
Publication of WO2018055660A1 publication Critical patent/WO2018055660A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions

Definitions

  • the present invention relates to a purchase recommendation system, a purchase recommendation method, and a program for appropriately recommending purchased products to a purchaser.
  • the purchase history associated with the purchaser is stored in the database of the purchased WEB page. It is known that, based on this purchase history, when the purchaser is browsing a WEB page, the purchase of a similar product is promoted by displaying a product similar to this product.
  • Patent Document 1 discloses a method for recommending appropriate products and services to the user based on purchase history information of products purchased by the user and browsed product information regarding products viewed by the user.
  • Patent Document 1 only recommends a product to be purchased (recommendation), and does not recommend the purchase act itself.
  • the consumer goods to be purchased regularly are determined for the products to be purchased, and what the purchaser wants to notify is not the products to be purchased, but the quantity to be purchased, the timing of purchase, If you buy the same product, it is time to buy it at a low price. For example, if you are purchasing ballpoint pens or plastic bottles that are consumed regularly in the office, it is desirable to let you know that you should purchase these items before they run out of stock.
  • the present invention provides a purchase recommendation system, a purchase recommendation method, and a program in which a computer system learns a purchaser's tendency of purchased products and recommends purchasers to the purchaser at a predetermined timing based on the learned result. Objective.
  • the present invention provides the following solutions.
  • the invention according to the first feature is a data extraction means for extracting the purchase number and purchase time of the purchaser for the purchased product,
  • Purchase standard data generating means for generating purchase standard data as a standard for encouraging the purchaser to purchase at a predetermined time by learning the number of purchases and the purchase time;
  • purchase time calculation means for calculating the next purchase time,
  • a purchase recommendation system comprising recommendation means for recommending that the purchaser purchase the purchased product at a timing prior to the calculated purchase time.
  • the purchase standard data which serves as a reference for prompting the purchaser to purchase at a predetermined time, is obtained by extracting the purchase number and purchase time of the purchaser for the purchased product.
  • the number of purchases and the purchase time are learned and generated.
  • the purchase number and purchase time of the previous purchase are referred to the purchase product, the next purchase time is calculated, and the purchase product is calculated.
  • Encourage buyers to purchase at an earlier time is obtained by extracting the purchase number and purchase time of the purchaser for the purchased product.
  • the invention according to the first feature is a system category, but in other categories such as a method and a program, the same actions and effects corresponding to the category are exhibited.
  • the invention according to the second feature is the invention according to the first feature, wherein a purchase budget receiving means for accepting registration of a budget for purchasing a predetermined purchase product, a price and the number of purchases of the purchase product And a subtracting means for subtracting the product of the purchase from the budget, and the recommending means recommends a purchase recommending system that recommends the purchaser to purchase with the number of purchases within the range of the subtracted budget. provide.
  • the registration of a budget for purchasing a predetermined purchased product is accepted, and the product of the price and the number of purchases of the purchased product is subtracted from the budget and subtracted. Encourage buyers to purchase at the number of purchases within their budget.
  • the invention according to a third feature is the invention according to the first feature, wherein the recommendation means recommends the purchase product at a timing earlier than the calculated purchase time. However, when it is determined that the price of the purchased product is low, a purchase recommendation system that recommends the purchased product at the price is provided.
  • the timing is determined at a timing when the price of the purchased product is determined to be low. Recommend the purchased product at the price.
  • the invention according to the fourth feature is the invention according to the third feature, wherein the recommendation means recommends a purchased product at a timing earlier than the calculated purchase time.
  • a recommendation grace period is provided, and a purchase recommendation system that recommends the purchased product at the price at the timing when the price of the purchased product is determined to be low within the recommended grace period is provided.
  • a recommendation grace period is provided, which is a period for making a recommendation, and the recommended grace period.
  • the purchase product is recommended at the price when it is determined that the price of the purchase product is low.
  • the computer system learns the tendency of the purchase item of the purchaser, and the purchaser can receive a recommendation for purchase at an appropriate timing from the learned result.
  • FIG. 1 is a diagram showing functional blocks of the purchase recommendation system 1.
  • FIG. 2 is a flowchart showing learning and recommendation processing executed by the purchase recommendation computer 100.
  • FIG. 3 is a flowchart showing the purchase number calculation process executed by the purchase recommendation computer 100.
  • FIG. 4 is a graph of price and time as an example for determining the recommendation time.
  • FIG. 5 is an example of a data table of sales data.
  • FIG. 6 is a specific notification example of a recommendation message.
  • FIG. 1 is a diagram showing a system configuration of a purchase recommendation system 1 which is a preferred embodiment of the present invention.
  • the purchase recommendation system 1 includes at least a purchase recommendation computer 100 and includes a sales computer 200 and a sales data database 50 depending on the system configuration.
  • the purchase recommendation system 1 will be described in the case where the sales computer 200 and the sales data database 50 are included as individual hardware. However, the sales computer 200 and the sales data database 50 are separated from the purchase recommendation computer 100 and the individual hardware. These functions may be included in the purchase recommendation computer 100 instead of the computer.
  • the purchase recommendation computer 100 and the sales computer 200, and the purchase recommendation computer 100 and the sales data database 50 are communicably connected via a public line network or a dedicated line, and data communication is performed.
  • the purchase recommendation computer 100 and the sales computer 200 may be computers and servers accessible from computer terminals used by recruiters.
  • the sales data database 50 is a database accessible by the purchase recommendation computer 100.
  • the purchase recommendation computer 100 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and the like as the control unit 110, and the communication unit 120 for enabling communication with other devices.
  • a device that can be connected to a wired / wireless LAN a device that supports IEEE 802.11 (Wireless Fidelity), a device that supports wired connection such as USB and HDMI (registered trademark), and the like.
  • the communication unit 120 enables data communication with the sales computer 200, the sales data database 50, and a computer terminal used by a person in charge of determining a purchase budget.
  • the control unit 110 when the control unit 110 reads a predetermined program, the data extraction module 150, the purchase reference data generation module 160, the purchase time calculation module 170, in cooperation with the communication unit 120 and other hardware, A purchase recommendation module 180, a purchase budget receiving module 190, and a budget reduction module 192 are realized.
  • the sales computer 200 may be a WEB server for allowing a user who is a purchaser to purchase a product via a network such as the Internet.
  • a control unit (not shown) reads a predetermined program, In cooperation with the product sales module 210 is realized.
  • the merchandise sales module 210 accepts access from a purchaser's terminal via a browser, presents purchased merchandise information to the purchaser, accepts selection and determination of purchased merchandise, and realizes settlement of accounts at the presented price.
  • the sales computer 200 When the sales computer 200 sells the purchased product to the purchaser, the sales computer 200 transmits data (sales data) such as the purchased product, purchaser, purchase time, and number of purchases to the sales data database 50 in association with the purchaser. .
  • data sales data
  • the purchaser is stored in the sales computer 200 and the sales data database 50 in association with an ID for identifying the purchaser, an e-mail address as a notification destination for recommending the purchaser, and the like.
  • the purchase recommendation computer 100 may be connected to a plurality of sales computers 200 with different operating entities.
  • the purchase recommendation computer 100 and the sales data database 50 receive sales data from each of the plurality of sales computers 200 and store them in the sales data database 50. At this time, from which sales computer 200 the purchased product is purchased may be stored in association with each other.
  • the sales data database 50 receives sales data (purchased products, number of purchases, purchase time) for each purchaser (user) transmitted from the sales computer 200 and stores them in a data table as shown in FIG. Further, as will be described later, the purchaser's budget for the purchased product (budget within a predetermined period such as the date) is received from the terminal of the person in charge and stored in association with the purchased product.
  • FIG. 2 is a flowchart of learning and recommendation processing executed by the purchase recommendation computer 100. The processing executed by the modules of each device described above will be described together with this processing.
  • the purchase recommendation computer 100 identifies a purchaser and a purchased product in order to recommend a predetermined purchased product (step S10). Then, the data extraction module 150 extracts the past purchase time and the number of purchases in the purchase product of the purchaser from the sales data database 50 based on the identified purchaser and purchase product (step S11).
  • the purchase criteria data generation module 160 learns the relationship between the purchase time and the number of purchases in the past purchase product extracted by the purchaser, and the criteria for prompting the purchaser to purchase at a predetermined time.
  • the purchase reference data is generated (step S12). That is, the purchase reference data generation module 160 learns the past purchase time and the number of purchases, and generates purchase reference data for determining the next purchase time and the number of purchases.
  • This learning process may simply be a process of deriving an approximate line or approximate curve (least square method) from the purchase time and the number of purchases.
  • an approximate line or an approximate curve is the purchase reference data.
  • This learning process may be so-called machine learning.
  • a nearest neighbor method, a naive Bayes method, a decision tree, or a support vector machine may be used.
  • deep learning may be used in which a characteristic amount for learning is generated by using a neural network. The past purchase time and the number of purchases may be learned as supervised data.
  • the nearest neighbor method or the k-nearest neighbor method past examples of purchase time are placed in the feature space, and when the data to be newly determined is given, the past example that is closest in the feature space (1 or k) classes (specific date and time of purchase) are used as the prediction results.
  • the data components include “Increase rate of number of people using purchased products” and “Substitutes for purchased products” It is conceivable to use “the number of products purchased” or the like as a feature amount. A feature space is generated with these feature amounts, and purchase reference data is generated.
  • the probability of the recommended purchase period is calculated for each of the above feature quantities, and the score of the specific period is assigned to each feature quantity, and this score is added together.
  • the purchase time is determined based on the score.
  • a function for this determination is purchase reference data.
  • the logarithm of the calculated probability may be used.
  • the purchase time calculation module 170 of the purchase recommendation computer 100 calculates the next purchase time with reference to the purchase number and the purchase time of the previous purchase based on the purchase standard data for the purchased product (step S13). That is, for example, if the purchase standard data is a simple first-order approximation and 100 items are purchased once every six months, the number of purchases made last time is 100 and the last purchase time is 3 months ago The purchase time is three months later, and the number of purchases is 100.
  • the purchase recommendation module 180 determines whether the current time is the recommendation time (recommended timing) (step S14), and if it is not the recommendation time (S14: “NO”), the process loops. On the other hand, if it is the recommendation time (S14: “YES”), the process proceeds to step S15.
  • the recommendation time is preferably a time slightly before the calculated purchase time, and may be, for example, several days or months before the calculated purchase time.
  • the purchase recommendation module 180 recommends the purchaser so that the purchaser purchases the purchased product at a timing before the calculated purchase timing.
  • the timing before the purchase time may be a predetermined period (2 to 3 days, 1 week, 1 month, etc.), and as shown in FIG. It may be an appropriate timing within the period.
  • the specific content recommended here may be to send a recommendation message to a notification destination such as an e-mail address associated with the purchaser.
  • the recommendation is made at the price of the purchased product at this timing (the price sold by the sales computer 200).
  • the purchase recommendation computer 100 determines whether or not the purchaser has purchased the purchased product after actually making a recommendation to the purchaser (step S16).
  • the sales computer 200 transmits sales data to the sales data database 50 each time the purchaser purchases a product by the product sales module 210. Therefore, the purchase recommendation computer 100 determines whether or not the purchaser has purchased the purchased product by determining whether or not the received sales data is a recommended purchased product.
  • step S16 determines that the purchaser has purchased the purchased product. If the purchase recommendation computer 100 determines that the purchaser has purchased the purchased product (step S16: “YES”), the purchase recommendation data is again obtained based on the date and time of purchase (currently) and the number of purchases. Generate (regenerate) (step S17). Then, the purchase recommendation computer 100 waits until the next recommendation time comes. However, regardless of whether or not the next recommendation time has come, if the purchaser purchases the purchased product for some reason again, the process of step S17 is performed again to generate purchase reference data.
  • step S16 determines that the purchaser has purchased the purchased product.
  • step S16 if it is determined that the purchaser has not purchased the purchased product (step S16: “NO”), a standby process of a predetermined date and time is performed assuming that there is no reaction from the purchaser. If there is still no purchase, the time for the recommendation process is determined again.
  • the determination of the recommendation time may be arbitrarily determined by the system, for example, three days later from the day after the standby process, or one week later, one month later. Then, the process of step S14 is executed depending on whether or not the determined recommendation time has been reached.
  • the purchase budget receiving module 190 accepts registration of a budget for a predetermined purchased product from a terminal of a person in charge of purchase at an arbitrary timing (step S20).
  • This budget is a budget that can be purchased within a predetermined period of days, months, and years.
  • the budget reduction module 192 of the purchase recommendation computer 100 determines that the predetermined purchase product has been purchased at the timing when the sales data is received from the sales computer 200, and the product of the number of purchases and the price (that is, the purchase payment amount). ) Is subtracted from the budget amount (step S21).
  • step S15 of learning and recommendation processing when the recommendation is executed, the number of purchases is calculated by dividing the subtracted budget amount by the price of the purchased product at the time of execution.
  • the subtracted amount is divided by 3 which is the remaining quarter.
  • the number of purchases is calculated as the budget amount obtained by subtracting the amount divided by.
  • the subtracted budget amount is simply subtracted from the number of purchases already made by subtracting the number of purchases expected (4 times in this case) from the budget minus the purchase payment amount. It is obtained by dividing by a numerical value (in this case, 3) (in this case, once).
  • FIG. 4 is a graph showing the relationship between the price of the purchased product and the date.
  • the purchase recommendation computer 100 appropriately receives the price of the purchased product from the sales computer 200 and generates a judgment standard such as a schematic graph. As shown in the figure, the price of the purchased product is constant, and suddenly drops at the time point A. Therefore, it is desirable for the purchaser to purchase after A.
  • step S15 of the learning and recommendation process described above the recommendation is performed at a timing before the purchase time, but this may be a timing between the recommendation grace period. That is, a period before the purchase period is set as a recommendation grace period, and during this period, the purchase recommendation module 180 monitors the price of the purchased product. Then, at the timing when the predetermined amount is reduced, it is determined that the purchase should be made, and at that time, the recommendation in step S15 is executed.
  • the person in charge may set a range of the amount of money to be lowered in advance as to whether the purchase recommendation module 180 determines to be the recommendation time when the amount is reduced at the time of A. With this process, when purchasing a purchased product, it becomes possible to receive a recommendation when the price drops.
  • FIG. 5 is an example of a data table of purchase data stored in the sales data database 50. As shown in the figure, a budget, a time, and a number (the number of purchases) are stored in association with each purchaser and each purchased product (X, Y).
  • FIG. 6 is a screen image diagram of a message notified to the purchaser when the purchase recommendation computer 100 makes a recommendation.
  • the purchase recommendation computer 100 displays a message asking whether or not to purchase a purchased product, a price, the number of purchases, and now.
  • the purchaser can set the recommended grace period described with reference to FIG. 4 by purchasing when the price drops within two months.
  • the means and functions described above are realized by a computer (including a CPU, an information processing apparatus, and various terminals) reading and executing a predetermined program.
  • the program may be, for example, in a form (SaaS: Software as a Service) provided from a computer via a network, or a flexible disk, CD (CD-ROM, etc.), DVD (DVD-ROM, DVD). It may be provided in a form recorded in a computer-readable recording medium such as a RAM.
  • the computer reads the program from the recording medium, transfers it to the internal storage device or the external storage device, stores it, and executes it.
  • the program may be recorded in advance in a storage device (recording medium) such as a magnetic disk, an optical disk, or a magneto-optical disk, and provided from the storage device to a computer via a communication line.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

[Problem] To enable a computer system to learn tendencies related to purchased products of a purchaser, and recommend purchases to the purchaser at prescribed timings on the basis of the learning results. [Solution] A purchase recommendation system 1 extracts the past number of purchases and purchase times of a purchaser with respect to a purchased product, learns the number of purchases and the purchase times and generates purchase reference data which serves as a reference for prompting the purchaser to make a purchase at a prescribed time, refers to the number of purchases and the purchase time in the previous purchase of the purchased product, on the basis of the purchase reference data, calculates the next purchase time, and recommends that the purchaser purchase the purchased product at a timing earlier than the calculated purchase time.

Description

購入レコメンドシステム、購入レコメンド方法及びプログラムPurchase recommendation system, purchase recommendation method and program
　本発明は、購入者に購入商品を適切に推奨する購入レコメンドシステム、購入レコメンド方法及びプログラムに関する。 The present invention relates to a purchase recommendation system, a purchase recommendation method, and a program for appropriately recommending purchased products to a purchaser.
　近年、インターネットを利用した商品の購入は一般消費者にもｅコマースとして日常化している。スマートフォンやパソコンから商品を購入するＷＥＢページにアクセスして、商品を閲覧し、購入する商品を選択する。そして、カード等で電子決済を済ませて、商品の発送を、職場や自宅で待つといった購入及び流通方法が一般的である。 In recent years, purchases of products using the Internet have become commonplace for general consumers as e-commerce. Access the WEB page for purchasing products from a smartphone or personal computer, browse the products, and select the products to purchase. In general, the purchase and distribution method is such that electronic payment is completed with a card or the like, and the product is waited for at the workplace or at home.
　ここで、一度、購入者が、商品をそのＷＥＢページで購入すると、その購入したＷＥＢページのデータベースに購入者と関係づけられた購入履歴が記憶される。そして、この購入履歴に基づいて、この購入者がＷＥＢページを閲覧している際に、この商品に類似した商品を表示することで、類似商品の購入を促すことが知られている。 Here, once the purchaser purchases the product on the WEB page, the purchase history associated with the purchaser is stored in the database of the purchased WEB page. It is known that, based on this purchase history, when the purchaser is browsing a WEB page, the purchase of a similar product is promoted by displaying a product similar to this product.
　例えば、特許文献1では、ユーザが購入した商品の購入履歴情報とユーザが視認した商品に関する閲覧商品情報に基づいて、ユーザに適切な商品やサービスを推奨する方法が開示されている。 For example, Patent Document 1 discloses a method for recommending appropriate products and services to the user based on purchase history information of products purchased by the user and browsed product information regarding products viewed by the user.
特開２０１５－１３３０３３号公報JP 2015-133303 A
　しかしながら、特許文献１の構成では、購入すべき商品の推奨（レコメンド）を行うに過ぎず、購入行為自体を推奨するものではない。すなわち、定期的に購入する消費財は、購入すべき商品自体は決定しており、購入者が通知してほしいことは、購入すべき商品ではなく、購入する個数や、購入するタイミング、すなわち、同じ商品を購入するならば、安値で購入するタイミングである。例えば、オフィスで定期的に消費されるボールペンやペットボトルを購入している場合は、これらの商品の在庫がなくなる前に、購入すべきことを知らせてくれることが望ましい。 However, the configuration of Patent Document 1 only recommends a product to be purchased (recommendation), and does not recommend the purchase act itself. In other words, the consumer goods to be purchased regularly are determined for the products to be purchased, and what the purchaser wants to notify is not the products to be purchased, but the quantity to be purchased, the timing of purchase, If you buy the same product, it is time to buy it at a low price. For example, if you are purchasing ballpoint pens or plastic bottles that are consumed regularly in the office, it is desirable to let you know that you should purchase these items before they run out of stock.
　本発明は、コンピュータシステムが、購入者の購入商品の傾向を学習し、その学習した結果から、購入者に購入を所定のタイミングで推奨する購入レコメンドシステム、購入レコメンド方法及びプログラムを提供することを目的とする。 The present invention provides a purchase recommendation system, a purchase recommendation method, and a program in which a computer system learns a purchaser's tendency of purchased products and recommends purchasers to the purchaser at a predetermined timing based on the learned result. Objective.
　本発明では、以下のような解決手段を提供する。 The present invention provides the following solutions.
　第１の特徴に係る発明は、購入商品について、購入者の過去の購入数と購入時期を抽出するデータ抽出手段と、
　所定の時期に前記購入者に購入を促すための基準となる購入基準データを、前記購入数と前記購入時期を学習して生成する購入基準データ生成手段と、
　前記購入商品について、前記購入基準データに基づいて、前回購入した購入数と購入時期を参照し、次回の購入時期を算出する購入時期算出手段と、
　前記購入商品を、前記算出された購入時期よりも前のタイミングで、前記購入者が購入するように推奨するレコメンド手段と、を備える購入レコメンドシステムを提供する。 The invention according to the first feature is a data extraction means for extracting the purchase number and purchase time of the purchaser for the purchased product,
Purchase standard data generating means for generating purchase standard data as a standard for encouraging the purchaser to purchase at a predetermined time by learning the number of purchases and the purchase time;
For the purchased product, based on the purchase standard data, referring to the number of purchases and purchase time purchased last time, purchase time calculation means for calculating the next purchase time,
A purchase recommendation system comprising recommendation means for recommending that the purchaser purchase the purchased product at a timing prior to the calculated purchase time.
　第１の特徴に係る発明によれば、購入商品について、購入者の過去の購入数と購入時期を抽出し、所定の時期に購入者に購入を促すための基準となる購入基準データを、購入数と購入時期を学習して生成し、購入商品について、購入基準データに基づいて、前回購入した購入数と購入時期を参照し、次回の購入時期を算出し、購入商品を、算出された購入時期よりも前のタイミングで、購入者が購入するように推奨する。 According to the invention relating to the first feature, the purchase standard data, which serves as a reference for prompting the purchaser to purchase at a predetermined time, is obtained by extracting the purchase number and purchase time of the purchaser for the purchased product. The number of purchases and the purchase time are learned and generated. Based on the purchase criteria data, the purchase number and purchase time of the previous purchase are referred to the purchase product, the next purchase time is calculated, and the purchase product is calculated. Encourage buyers to purchase at an earlier time.
　第１の特徴に係る発明は、システムのカテゴリであるが、方法及びプログラム等の他のカテゴリにおいても、そのカテゴリに応じた同様の作用・効果を発揮する。 The invention according to the first feature is a system category, but in other categories such as a method and a program, the same actions and effects corresponding to the category are exhibited.
　第２の特徴に係る発明は、第１の特徴に係る発明であって、所定の購入商品を購入するための予算の登録を受付ける購入予算受付手段と、前記購入商品を購入した価格と購入数との積を、前記予算から減算する予算減算手段と、を備え、前記レコメンド手段は、前記減算された予算の範囲内の購入数で、前記購入者に購入するように推奨する購入レコメンドシステムを提供する。 The invention according to the second feature is the invention according to the first feature, wherein a purchase budget receiving means for accepting registration of a budget for purchasing a predetermined purchase product, a price and the number of purchases of the purchase product And a subtracting means for subtracting the product of the purchase from the budget, and the recommending means recommends a purchase recommending system that recommends the purchaser to purchase with the number of purchases within the range of the subtracted budget. provide.
　第２の特徴に係る発明によれば、さらに、所定の購入商品を購入するための予算の登録を受付け、購入商品を購入した価格と購入数との積を、予算から減算し、減算された予算の範囲内の購入数で、購入者に購入するように推奨する。 According to the second aspect of the invention, the registration of a budget for purchasing a predetermined purchased product is accepted, and the product of the price and the number of purchases of the purchased product is subtracted from the budget and subtracted. Encourage buyers to purchase at the number of purchases within their budget.
　第３の特徴に係る発明は、第１の特徴に係る発明であって、前記レコメンド手段は、前記購入商品を、前記算出された購入時期よりも前のタイミングで推奨を行う際に、当該タイミングが、当該購入商品の価格が低いと判断したタイミングで、当該価格で当該購入商品を推奨する購入レコメンドシステムを提供する。 The invention according to a third feature is the invention according to the first feature, wherein the recommendation means recommends the purchase product at a timing earlier than the calculated purchase time. However, when it is determined that the price of the purchased product is low, a purchase recommendation system that recommends the purchased product at the price is provided.
　第３の特徴に係る発明によれば、さらに、購入商品を、算出された購入時期よりも前のタイミングで推奨を行う際に、当該タイミングが、当該購入商品の価格が低いと判断したタイミングで、当該価格で当該購入商品を推奨する。 According to the third aspect of the invention, when recommending a purchased product at a timing prior to the calculated purchase time, the timing is determined at a timing when the price of the purchased product is determined to be low. Recommend the purchased product at the price.
　第４の特徴に係る発明は、第３の特徴に係る発明であって、前記レコメンド手段は、購入商品を、算出された購入時期よりも前のタイミングで推奨を行う場合に、推奨を行う期間であるレコメンド猶予期間を設け、当該レコメンド猶予期間内で、当該購入商品の価格が低いと判断したタイミングで、当該価格で当該購入商品を推奨する購入レコメンドシステムを提供する。 The invention according to the fourth feature is the invention according to the third feature, wherein the recommendation means recommends a purchased product at a timing earlier than the calculated purchase time. A recommendation grace period is provided, and a purchase recommendation system that recommends the purchased product at the price at the timing when the price of the purchased product is determined to be low within the recommended grace period is provided.
　第４の特徴に係る発明によれば、さらに、購入商品を、算出された購入時期よりも前のタイミングで推奨を行う場合に、推奨を行う期間であるレコメンド猶予期間を設け、当該レコメンド猶予期間内で、当該購入商品の価格が低いと判断したタイミングで、当該価格で当該購入商品を推奨する。 According to the invention according to the fourth feature, when recommending a purchased product at a timing prior to the calculated purchase time, a recommendation grace period is provided, which is a period for making a recommendation, and the recommended grace period. The purchase product is recommended at the price when it is determined that the price of the purchase product is low.
　本発明によれば、コンピュータシステムが、購入者の購入商品の傾向を学習し、その学習した結果から、購入者は、購入についての推奨を、適切なタイミングで受けることができる。 According to the present invention, the computer system learns the tendency of the purchase item of the purchaser, and the purchaser can receive a recommendation for purchase at an appropriate timing from the learned result.
図１は、購入レコメンドシステム１の機能ブロックを示す図である。FIG. 1 is a diagram showing functional blocks of the purchase recommendation system 1. 図２は、購入レコメンドコンピュータ１００が実行する学習及びレコメンド処理を示すフローチャートである。FIG. 2 is a flowchart showing learning and recommendation processing executed by the purchase recommendation computer 100. 図３は、購入レコメンドコンピュータ１００が実行する購入数算出処理を示すフローチャートである。FIG. 3 is a flowchart showing the purchase number calculation process executed by the purchase recommendation computer 100. 図４は、レコメンド時期を決定する一例となる価格と時期のグラフである。FIG. 4 is a graph of price and time as an example for determining the recommendation time. 図５は、販売データのデータテーブルの例である。FIG. 5 is an example of a data table of sales data. 図６は、レコメンドメッセージの具体的な通知例である。FIG. 6 is a specific notification example of a recommendation message.
　以下、本発明を実施するための最良の形態について図を参照しながら説明する。なお、これはあくまでも一例であって、本発明の技術的範囲はこれに限られるものではない。 Hereinafter, the best mode for carrying out the present invention will be described with reference to the drawings. This is merely an example, and the technical scope of the present invention is not limited to this.
　［購入レコメンドシステムのシステム構成］
　図１に基づいて、購入レコメンドシステム１のシステム構成について説明する。図１は、本発明の好適な実施形態である購入レコメンドシステム１のシステム構成を示す図である。購入レコメンドシステム１は、少なくとも、購入レコメンドコンピュータ１００で構成され、システム形態によっては、販売コンピュータ２００、販売データデータベース５０を含む。 [System configuration of purchase recommendation system]
The system configuration of the purchase recommendation system 1 will be described with reference to FIG. FIG. 1 is a diagram showing a system configuration of a purchase recommendation system 1 which is a preferred embodiment of the present invention. The purchase recommendation system 1 includes at least a purchase recommendation computer 100 and includes a sales computer 200 and a sales data database 50 depending on the system configuration.
　以下では、購入レコメンドシステム１は、販売コンピュータ２００、販売データデータベース５０を個別のハードウェアとして含む場合で説明を行うが、販売コンピュータ２００、販売データデータベース５０が購入レコメンドコンピュータ１００と、個別のハードウェアコンピュータでなく、これらの機能が購入レコメンドコンピュータ１００に含まれていてもよい。 In the following description, the purchase recommendation system 1 will be described in the case where the sales computer 200 and the sales data database 50 are included as individual hardware. However, the sales computer 200 and the sales data database 50 are separated from the purchase recommendation computer 100 and the individual hardware. These functions may be included in the purchase recommendation computer 100 instead of the computer.
　購入レコメンドコンピュータ１００と販売コンピュータ２００及び、購入レコメンドコンピュータ１００と販売データデータベース５０は、公衆回線網又は専用線で通信可能に接続されデータ通信が行われる。購入レコメンドコンピュータ１００及び販売コンピュータ２００は、採用担当者等が利用するコンピュータ端末からアクセス可能なコンピュータ、サーバであってよい。販売データデータベース５０は、購入レコメンドコンピュータ１００がアクセス可能なデータベースである。 The purchase recommendation computer 100 and the sales computer 200, and the purchase recommendation computer 100 and the sales data database 50 are communicably connected via a public line network or a dedicated line, and data communication is performed. The purchase recommendation computer 100 and the sales computer 200 may be computers and servers accessible from computer terminals used by recruiters. The sales data database 50 is a database accessible by the purchase recommendation computer 100.
　［各コンピュータと機能の説明］
　図１に基づいて、購入レコメンドシステム１を構成するコンピュータとそのハードウェア構成及び機能について説明する。 [Description of each computer and functions]
Based on FIG. 1, the computer which comprises the purchase recommendation system 1, its hardware constitutions, and a function are demonstrated.
　購入レコメンドコンピュータ１００は、制御部１１０として、ＣＰＵ（Ｃｅｎｔｒａｌ　Ｐｒｏｃｅｓｓｉｎｇ　Ｕｎｉｔ）、ＲＡＭ（Ｒａｎｄｏｍ　Ａｃｃｅｓｓ　Ｍｅｍｏｒｙ）、ＲＯＭ（Ｒｅａｄ　Ｏｎｌｙ　Ｍｅｍｏｒｙ）等を備え、通信部１２０として、他の機器と通信可能にするためのデバイス、例えば、有線・無線ＬＡＮに接続可能なデバイスや、ＩＥＥＥ８０２．１１に準拠したＷｉＦｉ（Ｗｉｒｅｌｅｓｓ　Ｆｉｄｅｌｉｔｙ）対応デバイスやＵＳＢやＨＤＭＩ（登録商標）等の有線接続対応デバイス等を備える。この通信部１２０により、販売コンピュータ２００、販売データデータベース５０や、購入予算を決定する担当者等が利用するコンピュータ端末とのデータ通信が可能となる。 The purchase recommendation computer 100 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and the like as the control unit 110, and the communication unit 120 for enabling communication with other devices. For example, a device that can be connected to a wired / wireless LAN, a device that supports IEEE 802.11 (Wireless Fidelity), a device that supports wired connection such as USB and HDMI (registered trademark), and the like. The communication unit 120 enables data communication with the sales computer 200, the sales data database 50, and a computer terminal used by a person in charge of determining a purchase budget.
　購入レコメンドコンピュータ１００において、制御部１１０が所定のプログラムを読み込むことにより、通信部１２０及びその他のハードウェアと協働して、データ抽出モジュール１５０、購入基準データ生成モジュール１６０、購入時期算出モジュール１７０、購入レコメンドモジュール１８０、購入予算受付モジュール１９０、予算減額モジュール１９２を実現する。 In the purchase recommendation computer 100, when the control unit 110 reads a predetermined program, the data extraction module 150, the purchase reference data generation module 160, the purchase time calculation module 170, in cooperation with the communication unit 120 and other hardware, A purchase recommendation module 180, a purchase budget receiving module 190, and a budget reduction module 192 are realized.
　販売コンピュータ２００は、購入者であるユーザから、インターネット等のネットワーク経由で商品を購入させるためのＷＥＢサーバであってよく、図示していない制御部が所定のプログラムを読み込むことにより、他のハードウェアと協働して、商品販売モジュール２１０を実現する。商品販売モジュール２１０は、購入者の端末からブラウザ経由でアクセスを受付け、購入者に購入商品情報を提示し、購入商品の選択、決定を受付け、提示した価格で決算を実現する。 The sales computer 200 may be a WEB server for allowing a user who is a purchaser to purchase a product via a network such as the Internet. When a control unit (not shown) reads a predetermined program, In cooperation with the product sales module 210 is realized. The merchandise sales module 210 accepts access from a purchaser's terminal via a browser, presents purchased merchandise information to the purchaser, accepts selection and determination of purchased merchandise, and realizes settlement of accounts at the presented price.
　販売コンピュータ２００は、購入者に購入商品を販売した際に、購入商品、購入者、購入時期、購入数等のデータ（販売データ）を、購入者と対応付けて、販売データデータベース５０に送信する。なお、購入者には、購入者を特定するＩＤや購入者に推奨を行うための通知先となるメールアドレス等が対応付けられて、販売コンピュータ２００及び販売データデータベース５０に記憶されている。 When the sales computer 200 sells the purchased product to the purchaser, the sales computer 200 transmits data (sales data) such as the purchased product, purchaser, purchase time, and number of purchases to the sales data database 50 in association with the purchaser. . The purchaser is stored in the sales computer 200 and the sales data database 50 in association with an ID for identifying the purchaser, an e-mail address as a notification destination for recommending the purchaser, and the like.
　なお、図１では販売コンピュータ２００は、一台のみ記載しているが、購入レコメンドコンピュータ１００は、運営主体が異なる複数の販売コンピュータ２００と接続されていてよい。購入レコメンドコンピュータ１００及び販売データデータベース５０は、複数の各々の販売コンピュータ２００から販売データを受信し、販売データデータベース５０に記憶する。この際に、どの販売コンピュータ２００からその購入商品が購入されたかを対応付けて記憶しておいてもよい。 In FIG. 1, only one sales computer 200 is shown, but the purchase recommendation computer 100 may be connected to a plurality of sales computers 200 with different operating entities. The purchase recommendation computer 100 and the sales data database 50 receive sales data from each of the plurality of sales computers 200 and store them in the sales data database 50. At this time, from which sales computer 200 the purchased product is purchased may be stored in association with each other.
　販売データデータベース５０は、販売コンピュータ２００から送信された購入者（ユーザ）毎の販売データ（購入商品、購入数、購入時期）を受信し、図５のようなデータデーブルで記憶する。また、後述するように、その購入者の当該購入商品に対する予算（年月日等の所定期間内の予算）を、担当者の端末から受信して、購入商品と対応付けて記憶する。 The sales data database 50 receives sales data (purchased products, number of purchases, purchase time) for each purchaser (user) transmitted from the sales computer 200 and stores them in a data table as shown in FIG. Further, as will be described later, the purchaser's budget for the purchased product (budget within a predetermined period such as the date) is received from the terminal of the person in charge and stored in association with the purchased product.
　［学習及びレコメンド処理］
　次に、図２に基づいて、購入レコメンドシステム１が実行する学習及びレコメンド処理について説明する。図２は、購入レコメンドコンピュータ１００が実行する学習及びレコメンド処理のフローチャートである。上述した各装置のモジュールが実行する処理について、本処理に併せて説明する。 [Learning and recommendation processing]
Next, learning and recommendation processing executed by the purchase recommendation system 1 will be described with reference to FIG. FIG. 2 is a flowchart of learning and recommendation processing executed by the purchase recommendation computer 100. The processing executed by the modules of each device described above will be described together with this processing.
　はじめに、購入レコメンドコンピュータ１００は、所定の購入商品の推奨を行うために、ある購入者及び購入商品を特定する（ステップＳ１０）。そして、データ抽出モジュール１５０が、特定した購入者と購入商品に基づいて、販売データデータベース５０から、この購入者の購入商品における過去の購入時期、購入数を抽出する（ステップＳ１１）。 First, the purchase recommendation computer 100 identifies a purchaser and a purchased product in order to recommend a predetermined purchased product (step S10). Then, the data extraction module 150 extracts the past purchase time and the number of purchases in the purchase product of the purchaser from the sales data database 50 based on the identified purchaser and purchase product (step S11).
　次に、購入基準データ生成モジュール１６０は、この購入者において、抽出した過去の購入商品における、購入時期と購入数の関係を学習し、所定の時期に、この購入者に購入を促すための基準となる購入基準データを生成する（ステップＳ１２）。すなわち、購入基準データ生成モジュール１６０は、過去の購入時期と購入数を学習し、次の購入時期と購入数を判別するための購入基準データを生成する。 Next, the purchase criteria data generation module 160 learns the relationship between the purchase time and the number of purchases in the past purchase product extracted by the purchaser, and the criteria for prompting the purchaser to purchase at a predetermined time. The purchase reference data is generated (step S12). That is, the purchase reference data generation module 160 learns the past purchase time and the number of purchases, and generates purchase reference data for determining the next purchase time and the number of purchases.
　この学習処理とは、単純に、購入時期と購入数とから近似直線や近似曲線（最小二乗法）を導出する処理であってもよい。この場合、近似直線や近似曲線が購入基準データとなる。また、この学習処理は、いわゆる機械学習であってよい。機械学習の具体的なアルゴリズムとしては、最近傍法、ナイーブベイズ法、決定木、サポートベクターマシンを利用してよい。また、ニューラルネットワークを利用して、学習するための特徴量を自ら生成する深層学習（ディープラーニング）であってもよい。過去の購入時期と購入数を教師ありデータとして学習してもよい。 This learning process may simply be a process of deriving an approximate line or approximate curve (least square method) from the purchase time and the number of purchases. In this case, an approximate line or an approximate curve is the purchase reference data. This learning process may be so-called machine learning. As a specific algorithm for machine learning, a nearest neighbor method, a naive Bayes method, a decision tree, or a support vector machine may be used. Further, deep learning may be used in which a characteristic amount for learning is generated by using a neural network. The past purchase time and the number of purchases may be learned as supervised data.
　例えば、最近傍法やｋ近傍法であれば、購入時期の過去の実例を特徴空間に配置しておき、新しく判定したいデータが与えられた際に、特徴空間上で最も距離が近い過去の実例（1個又はk個）のクラス（購入時期の具体的な日時）を予測結果とする。データの構成要素としては、「購入数」、「購入数の増加率」以外にも、購入者からのデータの入力を受けて「購入商品を使用する人数の増加率」、「購入商品の代替商品の購入数」等を特徴量として使用することが考えられる。これらの特徴量で特徴空間を生成し、購入基準データを生成する。 For example, in the case of the nearest neighbor method or the k-nearest neighbor method, past examples of purchase time are placed in the feature space, and when the data to be newly determined is given, the past example that is closest in the feature space (1 or k) classes (specific date and time of purchase) are used as the prediction results. In addition to “Number of purchases” and “Increase rate of purchases”, the data components include “Increase rate of number of people using purchased products” and “Substitutes for purchased products” It is conceivable to use “the number of products purchased” or the like as a feature amount. A feature space is generated with these feature amounts, and purchase reference data is generated.
　また、ナイーブベイズ法であれば、上記の特徴量毎に購入を推奨する時期の確率を算出し、各特徴量毎に、その具体的な時期のスコアをつけて、このスコアを足し合わせて、スコアの高さで、購入時期を判定する。この判定するための関数が購入基準データとなる。スコアには、算出した確率の対数を用いてよい。 In addition, in the case of the Naive Bayes method, the probability of the recommended purchase period is calculated for each of the above feature quantities, and the score of the specific period is assigned to each feature quantity, and this score is added together. The purchase time is determined based on the score. A function for this determination is purchase reference data. For the score, the logarithm of the calculated probability may be used.
　次に、購入レコメンドコンピュータ１００の購入時期算出モジュール１７０は、購入商品について、購入基準データに基づいて、前回購入した購入数と購入時期を参照し、次回の購入時期を算出する（ステップＳ１３）。すなわち、例えば、購入基準データが、単純な一次近似式で、６ヶ月に１回、１００個購入している場合に、前回購入した購入数が１００個で、前回購入した時期が３ヶ月前だと、購入時期は、３ヶ月後であって、購入数は１００個となる。 Next, the purchase time calculation module 170 of the purchase recommendation computer 100 calculates the next purchase time with reference to the purchase number and the purchase time of the previous purchase based on the purchase standard data for the purchased product (step S13). That is, for example, if the purchase standard data is a simple first-order approximation and 100 items are purchased once every six months, the number of purchases made last time is 100 and the last purchase time is 3 months ago The purchase time is three months later, and the number of purchases is 100.
　次に、実際にその算出した購入時期が来るまでは、時間待ちとなる。そのため、購入レコメンドモジュール１８０は、現在がレコメンド時期（推奨のタイミング）であるかを判断し（ステップＳ１４）、レコメンド時期ではない（Ｓ１４：「ＮＯ」）場合は、処理をループする。一方、レコメンド時期である（Ｓ１４：「ＹＥＳ」）場合は、ステップＳ１５に処理を移す。ここで、レコメンド時期とは、算出した購入時期よりも少し前の時期である方が望ましく、例えば、算出した購入時期の数日若しくは数ヶ月前であってよい。 Next, it will wait until the actual purchase time comes. Therefore, the purchase recommendation module 180 determines whether the current time is the recommendation time (recommended timing) (step S14), and if it is not the recommendation time (S14: “NO”), the process loops. On the other hand, if it is the recommendation time (S14: “YES”), the process proceeds to step S15. Here, the recommendation time is preferably a time slightly before the calculated purchase time, and may be, for example, several days or months before the calculated purchase time.
　このようにして、ステップＳ１５にて、購入レコメンドモジュール１８０は、購入商品を、算出された購入時期よりも前のタイミングで、購入者が購入するように購入者に推奨を行う。ここで、購入時期よりも前のタイミングとは、予め定められた期間（２～３日、１週間、１ヶ月等）であってもよいし、図４にて、後述するように、レコメンド猶予期間内の適切なタイミングであってよい。 In this way, in step S15, the purchase recommendation module 180 recommends the purchaser so that the purchaser purchases the purchased product at a timing before the calculated purchase timing. Here, the timing before the purchase time may be a predetermined period (2 to 3 days, 1 week, 1 month, etc.), and as shown in FIG. It may be an appropriate timing within the period.
　ここでの推奨の具体的な内容は、購入者に対応付けられたメールアドレス等の通知先にレコメンドのメッセージを送信することであってよい。なお、この推奨においては、このタイミングにおける購入商品の価格（この販売コンピュータ２００が販売する価格）で推奨を行う。 The specific content recommended here may be to send a recommendation message to a notification destination such as an e-mail address associated with the purchaser. In this recommendation, the recommendation is made at the price of the purchased product at this timing (the price sold by the sales computer 200).
　次に、購入レコメンドコンピュータ１００は、実際に購入者に対して推奨を行った後に、購入者が購入商品を購入したか否かを判断する（ステップＳ１６）。販売コンピュータ２００は、商品販売モジュール２１０が当該購入者が商品を購入すると、都度、販売データデータベース５０に販売データを送信している。したがって、購入レコメンドコンピュータ１００は、受信した販売データが、推奨した購入商品であるか否かを判断することで、購入者が購入商品を購入したか否かを判断する。 Next, the purchase recommendation computer 100 determines whether or not the purchaser has purchased the purchased product after actually making a recommendation to the purchaser (step S16). The sales computer 200 transmits sales data to the sales data database 50 each time the purchaser purchases a product by the product sales module 210. Therefore, the purchase recommendation computer 100 determines whether or not the purchaser has purchased the purchased product by determining whether or not the received sales data is a recommended purchased product.
　購入レコメンドコンピュータ１００は、購入者が購入商品を購入した（ステップＳ１６：「ＹＥＳ」）と判断した場合は、この購入商品を購入した日時（今現在）と購入数で、再度、購入基準データを生成する（再生成する）（ステップＳ１７）。そして、次のレコメンド時期になるまで、購入レコメンドコンピュータ１００は、待機する。しかしここで、次のレコメンド時期に来たか否かにかかわらず、再度、購入者が購入商品を何らかの都合で、購入した場合は、再度、ステップＳ１７の処理が行われ、購入基準データが生成される。 If the purchase recommendation computer 100 determines that the purchaser has purchased the purchased product (step S16: “YES”), the purchase recommendation data is again obtained based on the date and time of purchase (currently) and the number of purchases. Generate (regenerate) (step S17). Then, the purchase recommendation computer 100 waits until the next recommendation time comes. However, regardless of whether or not the next recommendation time has come, if the purchaser purchases the purchased product for some reason again, the process of step S17 is performed again to generate purchase reference data. The
　一方、購入者が購入商品を購入してない（ステップＳ１６：「ＮＯ」）と判断した場合は、購入者からの反応がないとして、予め決められた日時の待機処理をする。そして、それでも購入がない場合は、再度、レコメンド処理をするための時期を決定する。このレコメンド時期の決定は、例えば、その待機処理後の日から、３日後であってよいし、１週間後、１ヶ月後と、システムで任意に定めていてよい。そして、その決定されたレコメンド時期に到達したか否かで、ステップＳ１４の処理を実行する。 On the other hand, if it is determined that the purchaser has not purchased the purchased product (step S16: “NO”), a standby process of a predetermined date and time is performed assuming that there is no reaction from the purchaser. If there is still no purchase, the time for the recommendation process is determined again. The determination of the recommendation time may be arbitrarily determined by the system, for example, three days later from the day after the standby process, or one week later, one month later. Then, the process of step S14 is executed depending on whether or not the determined recommendation time has been reached.
　［購入数算出処理］
　次に、図３に基づいて、購入レコメンドコンピュータ１００が購入数を算出して推奨する処理について説明する。はじめに、購入予算受付モジュール１９０は、任意のタイミングで購入の担当者の端末から、所定の購入商品に対する予算の登録を受付ける（ステップＳ２０）。この予算とは、所定の日数、月数、年数を期間として、その期間内に購入できる予算である。 [Purchase number calculation process]
Next, based on FIG. 3, the process that the purchase recommendation computer 100 calculates and recommends the number of purchases will be described. First, the purchase budget receiving module 190 accepts registration of a budget for a predetermined purchased product from a terminal of a person in charge of purchase at an arbitrary timing (step S20). This budget is a budget that can be purchased within a predetermined period of days, months, and years.
　そして、購入レコメンドコンピュータ１００の予算減額モジュール１９２は、販売コンピュータ２００から販売データを受信したタイミングで、所定の購入商品が購入されたと判断して、その購入数と価格の積（すなわち、購入支払額）を、予算額から減算する（ステップＳ２１）。そして、学習及びレコメンド処理のステップＳ１５において、推奨を実行するときに、減算された予算額を、その実行時の購入商品の価格で除算処理することで、購入数が算出される。 Then, the budget reduction module 192 of the purchase recommendation computer 100 determines that the predetermined purchase product has been purchased at the timing when the sales data is received from the sales computer 200, and the product of the number of purchases and the price (that is, the purchase payment amount). ) Is subtracted from the budget amount (step S21). In step S15 of learning and recommendation processing, when the recommendation is executed, the number of purchases is calculated by dividing the subtracted budget amount by the price of the purchased product at the time of execution.
　この除算処理において、例えば、予算が年額で、購入時期が４ヶ月毎（クオーター）である場合、第１クオーターで購入し、減算された額を、残りのクオーターである３で除算し、その３で除算した額を減算された予算額として購入数を算出する。すなわち、減算された予算額は、単純に、予算から購入支払額を引いた額に、予算を立てている期間で購入が予想される回数（この場合４回）を、既に購入した回数から引いた（この場合１回）数値（この場合３）で除算を行うことで求められる。 In this division processing, for example, when the budget is an annual amount and the purchase time is every four months (quarter), the purchase is made with the first quarter, and the subtracted amount is divided by 3 which is the remaining quarter. The number of purchases is calculated as the budget amount obtained by subtracting the amount divided by. In other words, the subtracted budget amount is simply subtracted from the number of purchases already made by subtracting the number of purchases expected (4 times in this case) from the budget minus the purchase payment amount. It is obtained by dividing by a numerical value (in this case, 3) (in this case, once).
　［レコメンド時期の決定］
　図４は、購入商品の価格と日の関係を示すグラフである。購入レコメンドコンピュータ１００は、販売コンピュータ２００から適宜、購入商品の価格を受信し、この模式的なグラフのような判断基準を生成する。図に示すように、購入商品の価格は一定であり、Ａの時点で、急に下がっている。したがって、購入者はＡの後に購入することが望ましい。 [Decision of recommendation time]
FIG. 4 is a graph showing the relationship between the price of the purchased product and the date. The purchase recommendation computer 100 appropriately receives the price of the purchased product from the sales computer 200 and generates a judgment standard such as a schematic graph. As shown in the figure, the price of the purchased product is constant, and suddenly drops at the time point A. Therefore, it is desirable for the purchaser to purchase after A.
　上述の、学習及びレコメンド処理のステップＳ１５にて、推奨を購入時期より前のタイミングで行うが、これは、レコメンド猶予期間のタイミングであってよい。すなわち、購入時期よりも所定期間前の期間を、レコメンド猶予期間と定め、この間、購入レコメンドモジュール１８０は、購入商品の価格を監視する。そして、所定金額下がったタイミングで、購入すべきタイミングと判断し、その時点で、ステップＳ１５の推奨を実行する。 In step S15 of the learning and recommendation process described above, the recommendation is performed at a timing before the purchase time, but this may be a timing between the recommendation grace period. That is, a period before the purchase period is set as a recommendation grace period, and during this period, the purchase recommendation module 180 monitors the price of the purchased product. Then, at the timing when the predetermined amount is reduced, it is determined that the purchase should be made, and at that time, the recommendation in step S15 is executed.
　なお、Ａの時点でどの程度金額が下がれば、購入レコメンドモジュール１８０が、レコメンド時期と判断するかは、予め下がる金額の幅を担当者が設定しておいてよい。この処理によって、購入商品を購入する際に、価格が下がったタイミングで推奨を受けることが可能になる。 It should be noted that the person in charge may set a range of the amount of money to be lowered in advance as to whether the purchase recommendation module 180 determines to be the recommendation time when the amount is reduced at the time of A. With this process, when purchasing a purchased product, it becomes possible to receive a recommendation when the price drops.
　図５は、販売データデータベース５０が記憶する購入データのデータテーブルの一例である。図に示すように、購入者毎、購入商品（Ｘ，Ｙ）毎に、それぞれ予算、時期、数（購入数）が対応付けられて記憶されている。 FIG. 5 is an example of a data table of purchase data stored in the sales data database 50. As shown in the figure, a budget, a time, and a number (the number of purchases) are stored in association with each purchaser and each purchased product (X, Y).
　図６は、購入レコメンドコンピュータ１００が推奨を行った際に、購入者に通知したメッセージの画面イメージ図である。購入レコメンドコンピュータ１００は、図に示すように、購入商品、価格、購入数、今、購入するか否かを問いかけるメッセージを表示する。なお、購入タイミングについては、２ヶ月以内で値下がりしたら購入する等で、図４で説明したレコメンド猶予期間を購入者が設定することができる。 FIG. 6 is a screen image diagram of a message notified to the purchaser when the purchase recommendation computer 100 makes a recommendation. As shown in the figure, the purchase recommendation computer 100 displays a message asking whether or not to purchase a purchased product, a price, the number of purchases, and now. As for the purchase timing, the purchaser can set the recommended grace period described with reference to FIG. 4 by purchasing when the price drops within two months.
　上述した手段、機能は、コンピュータ（ＣＰＵ、情報処理装置、各種端末を含む）が、所定のプログラムを読み込んで、実行することによって実現される。プログラムは、例えば、コンピュータからネットワーク経由で提供される（ＳａａＳ：ソフトウェア・アズ・ア・サービス）形態であってもよいし、フレキシブルディスク、ＣＤ（ＣＤ－ＲＯＭなど）、ＤＶＤ（ＤＶＤ－ＲＯＭ、ＤＶＤ－ＲＡＭなど）等のコンピュータ読取可能な記録媒体に記録された形態で提供されてもよい。この場合、コンピュータはその記録媒体からプログラムを読み取って内部記憶装置又は外部記憶装置に転送し記憶して実行する。また、そのプログラムを、例えば、磁気ディスク、光ディスク、光磁気ディスク等の記憶装置（記録媒体）に予め記録しておき、その記憶装置から通信回線を介してコンピュータに提供するようにしてもよい。 The means and functions described above are realized by a computer (including a CPU, an information processing apparatus, and various terminals) reading and executing a predetermined program. The program may be, for example, in a form (SaaS: Software as a Service) provided from a computer via a network, or a flexible disk, CD (CD-ROM, etc.), DVD (DVD-ROM, DVD). It may be provided in a form recorded in a computer-readable recording medium such as a RAM. In this case, the computer reads the program from the recording medium, transfers it to the internal storage device or the external storage device, stores it, and executes it. The program may be recorded in advance in a storage device (recording medium) such as a magnetic disk, an optical disk, or a magneto-optical disk, and provided from the storage device to a computer via a communication line.
　以上、本発明の実施形態について説明したが、本発明は上述したこれらの実施形態に限るものではない。また、本発明の実施形態に記載された効果は、本発明から生じる最も好適な効果を列挙したに過ぎず、本発明による効果は、本発明の実施形態に記載されたものに限定されるものではない。 As mentioned above, although embodiment of this invention was described, this invention is not limited to these embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.
　１　購入レコメンドシステム、１００　購入レコメンドコンピュータ、２００　販売コンピュータ 1 purchase recommendation system, 100 purchase recommendation computer, 200 sales computer

Claims (6)

  1. 　購入商品について、購入者の過去の購入数と購入時期を抽出するデータ抽出手段と、
    　所定の時期に前記購入者に購入を促すための基準となる購入基準データを、前記購入数と前記購入時期を学習して生成する購入基準データ生成手段と、
    　前記購入商品について、前記購入基準データに基づいて、前回購入した購入数と購入時期を参照し、次回の購入時期を算出する購入時期算出手段と、
    　前記購入商品を、前記算出された購入時期よりも前のタイミングで、前記購入者が購入するように推奨するレコメンド手段と、を備える購入レコメンドシステム。     For the purchased product, a data extraction means to extract the purchase number and purchase time of the buyer in the past,
    Purchase standard data generating means for generating purchase standard data as a standard for encouraging the purchaser to purchase at a predetermined time by learning the number of purchases and the purchase time;
    For the purchased product, based on the purchase standard data, referring to the number of purchases and purchase time purchased last time, purchase time calculation means for calculating the next purchase time,
    A purchase recommendation system comprising: recommendation means for recommending that the purchaser purchase the purchased product at a timing before the calculated purchase time.
  2. 　所定の購入商品を購入するための予算の登録を受付ける購入予算受付手段と、
    　前記購入商品を購入した価格と購入数との積を、前記予算から減算する予算減算手段と、を備え、
    　前記レコメンド手段は、前記減算された予算の範囲内の購入数で、前記購入者に購入するように推奨する請求項１に記載の購入レコメンドシステム。     A purchase budget receiving means for accepting registration of a budget for purchasing a predetermined purchase product;
    Budget subtracting means for subtracting the product of the purchase price and the number of purchases from the budget,
    The purchase recommendation system according to claim 1, wherein the recommendation means recommends the purchaser to purchase with the number of purchases within the range of the subtracted budget.
  3. 　前記レコメンド手段は、前記購入商品を、前記算出された購入時期よりも前のタイミングで推奨を行う際に、当該タイミングが、当該購入商品の価格が低いと判断したタイミングで、当該価格で当該購入商品を推奨する請求項１に記載の購入レコメンドシステム。 When the recommendation means recommends the purchased product at a timing before the calculated purchase timing, the recommendation means determines the purchase price at the timing when it is determined that the price of the purchased product is low. The purchase recommendation system according to claim 1 which recommends goods.
  4. 　前記レコメンド手段は、前記購入商品を、前記算出された購入時期よりも前のタイミングで推奨を行う場合に、推奨を行う期間であるレコメンド猶予期間を設け、当該レコメンド猶予期間内で、当該購入商品の価格が低いと判断したタイミングで、当該価格で当該購入商品を推奨する請求項３に記載の購入レコメンドシステム。 The recommendation means, when recommending the purchased product at a timing prior to the calculated purchase time, provides a recommended grace period during which the recommendation is made, and within the recommended grace period, the purchased product The purchase recommendation system according to claim 3, wherein the purchase product is recommended at the price when it is determined that the price is low.
  5. 　購入商品について、購入者の過去の購入数と購入時期を抽出するステップと、
    　所定の時期に前記購入者に購入を促すための基準となる購入基準データを、前記購入数と前記購入時期を学習して生成するステップと、
    　前記購入商品について、前記購入基準データに基づいて、前回購入した購入数と購入時期を参照し、次回の購入時期を算出するステップと、
    　前記購入商品を、前記算出された購入時期よりも前のタイミングで、前記購入者が購入するように推奨するステップと、を備える購入レコメンド方法。     Extracting the buyer's past purchases and purchase time for the purchase,
    Learning purchase standard data as a reference for prompting the purchaser to purchase at a predetermined time by learning the number of purchases and the purchase time; and
    For the purchased product, referring to the purchase number and purchase time of the previous purchase based on the purchase standard data, calculating a next purchase time;
    And recommending that the purchaser purchase the purchased product at a timing prior to the calculated purchase time.
  6. 　コンピュータシステムに、
    　購入商品について、購入者の過去の購入数と購入時期を抽出するステップ、
    　所定の時期に前記購入者に購入を促すための基準となる購入基準データを、前記購入数と前記購入時期を学習して生成するステップ、
    　前記購入商品について、前記購入基準データに基づいて、前回購入した購入数と購入時期を参照し、次回の購入時期を算出するステップ、
    　前記購入商品を、前記算出された購入時期よりも前のタイミングで、前記購入者が購入するように推奨するステップ、を実行させるためのプログラム。
          Computer system,
    For the purchased product, the step of extracting the purchase number and purchase time of the buyer in the past,
    Learning purchase standard data as a reference for prompting the purchaser to purchase at a predetermined time by learning the number of purchases and the purchase time;
    For the purchased product, referring to the purchase number and purchase time of the previous purchase based on the purchase standard data, calculating a next purchase time,
    A program for executing the step of recommending the purchaser to purchase the purchased product at a timing before the calculated purchase time.
PCT/JP2016/077630 2016-09-20 2016-09-20 Purchase recommendation system, purchase recommendation method, and program WO2018055660A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2016/077630 WO2018055660A1 (en) 2016-09-20 2016-09-20 Purchase recommendation system, purchase recommendation method, and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2016/077630 WO2018055660A1 (en) 2016-09-20 2016-09-20 Purchase recommendation system, purchase recommendation method, and program

Publications (1)

Publication Number Publication Date
WO2018055660A1 true WO2018055660A1 (en) 2018-03-29

Family

ID=61689853

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/077630 WO2018055660A1 (en) 2016-09-20 2016-09-20 Purchase recommendation system, purchase recommendation method, and program

Country Status (1)

Country Link
WO (1) WO2018055660A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2022177307A (en) * 2019-07-03 2022-11-30 富士フイルム株式会社 Information proposition system, information proposition method, program, and recording medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001101314A (en) * 1999-08-19 2001-04-13 Internatl Business Mach Corp <Ibm> Method and device for virtually transacting commodity and method and device for transacting commodity
JP2009157421A (en) * 2007-12-25 2009-07-16 Sii Data Service Kk Self order system
JP2011243041A (en) * 2010-05-19 2011-12-01 Sony Corp Coupon selection support device, coupon selection support system, coupon selection support method and program
JP2012014467A (en) * 2010-06-30 2012-01-19 Rakuten Inc Information provision system, information provision device, information provision method, program and information recording medium
WO2012147405A1 (en) * 2011-04-28 2012-11-01 楽天株式会社 Purchasing administration device, purchasing administration method, purchasing administration program, and storage medium
JP2014038460A (en) * 2012-08-15 2014-02-27 Fujitsu Ltd Customer extraction program, customer extractor, and customer extraction method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001101314A (en) * 1999-08-19 2001-04-13 Internatl Business Mach Corp <Ibm> Method and device for virtually transacting commodity and method and device for transacting commodity
JP2009157421A (en) * 2007-12-25 2009-07-16 Sii Data Service Kk Self order system
JP2011243041A (en) * 2010-05-19 2011-12-01 Sony Corp Coupon selection support device, coupon selection support system, coupon selection support method and program
JP2012014467A (en) * 2010-06-30 2012-01-19 Rakuten Inc Information provision system, information provision device, information provision method, program and information recording medium
WO2012147405A1 (en) * 2011-04-28 2012-11-01 楽天株式会社 Purchasing administration device, purchasing administration method, purchasing administration program, and storage medium
JP2014038460A (en) * 2012-08-15 2014-02-27 Fujitsu Ltd Customer extraction program, customer extractor, and customer extraction method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2022177307A (en) * 2019-07-03 2022-11-30 富士フイルム株式会社 Information proposition system, information proposition method, program, and recording medium
JP7360524B2 (en) 2019-07-03 2023-10-12 富士フイルム株式会社 Information proposal system, information proposal method, program and recording medium

Similar Documents

Publication Publication Date Title
JP5918410B1 (en) Product sales forecasting system, product sales forecasting method, and non-transitory computer-readable recording medium
US11816721B2 (en) Recommendation apparatus and method
JP6220996B1 (en) Electronic warranty management server and product management method
JP2016071586A (en) Household account book management device, household account book management method and household account book management program
JP7117254B2 (en) Provision device, provision method and provision program
KR100891930B1 (en) Method, system and computer-readable recording medium for providing popular goods by using evaluation data on goods
CN111127074A (en) Data recommendation method
KR102270332B1 (en) Server for recommending business opportunity information and method thereof
JP2020013447A (en) Determination device, determination method, and determination program
KR101479379B1 (en) Method of sharing customer rate of merchandise based on social network
WO2018055660A1 (en) Purchase recommendation system, purchase recommendation method, and program
JP6486530B1 (en) Providing device, providing method, and providing program
JP6873030B2 (en) Information processing equipment, information processing methods and information processing programs
JP2021114340A (en) Notification device and method for determining notification means
JP6702628B1 (en) Providing device, providing method, and providing program
KR20100083540A (en) Method and system for brokering transaction for preorder goods
JP2011053855A (en) Initial purchaser extraction processing method, initial purchaser recommendation method, initial purchaser extraction system, and initial purchaser recommendation system
JP7469450B1 (en) Information processing device, information processing method, and program
JP2019204208A (en) Information processing apparatus, information processing method, and information processing program
JP7419583B1 (en) Information processing device, information processing method, and information processing program
JP7489950B2 (en) Information processing device, information processing method, and information processing program
JP6742254B2 (en) Providing device, providing method, and providing program
JP7250057B2 (en) LEARNING DEVICE, LEARNING METHOD AND LEARNING PROGRAM
JP6513751B2 (en) Determination apparatus, determination method, and determination program
KR102563130B1 (en) Apparatus and method for providing merchandise sales page

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16916731

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 06.06.2019)

NENP Non-entry into the national phase

Ref country code: JP

122 Ep: pct application non-entry in european phase

Ref document number: 16916731

Country of ref document: EP

Kind code of ref document: A1