WO2012049987A1 - Product recommendation system and product recommendation method - Google Patents

Product recommendation system and product recommendation method Download PDF

Info

Publication number
WO2012049987A1
WO2012049987A1 PCT/JP2011/072691 JP2011072691W WO2012049987A1 WO 2012049987 A1 WO2012049987 A1 WO 2012049987A1 JP 2011072691 W JP2011072691 W JP 2011072691W WO 2012049987 A1 WO2012049987 A1 WO 2012049987A1
Authority
WO
WIPO (PCT)
Prior art keywords
product
profit
user
recommended
action
Prior art date
Application number
PCT/JP2011/072691
Other languages
French (fr)
Japanese (ja)
Inventor
亨太 菅野
Original Assignee
日本電気株式会社
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 日本電気株式会社 filed Critical 日本電気株式会社
Priority to JP2012538633A priority Critical patent/JPWO2012049987A1/en
Publication of WO2012049987A1 publication Critical patent/WO2012049987A1/en

Links

Images

Classifications

    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • the present invention relates to a product recommendation system and a product recommendation method for recommending products.
  • Patent Document 1 discloses an information providing system that determines information to be distributed to a user based on the user's browsing history.
  • Patent Document 2 discloses a system in which a customer can order a desired product at a desired price from a plurality of stores at a desired price.
  • Patent Document 3 discloses an estimation method capable of appropriately performing estimation work.
  • Patent Document 4 discloses a product recommendation system capable of recommending a product that accurately reflects the needs of a user who intends to purchase the product.
  • Patent Document 5 enables customers who can obtain high profits to be preferentially targeted, and can transmit knowledge information corresponding to the targeted customers and sales support information to be used when visiting customers.
  • a customer targeting system is disclosed.
  • An object of the present invention is to provide a product recommendation system and a product recommendation method capable of recommending a product so that the profit of the operator is increased.
  • the product recommendation system includes a product profit storage means for storing the profit of the product, the probability that the user will perform a specific action when the product is recommended, and the stored memory. Based on the profit of the product, the profit expected value calculation means for calculating the expected value of the profit due to the recommended product, and the product recommended for the user is selected based on the calculated expected value And recommended product selection means. Further, the product recommendation processing method according to another aspect of the present invention stores the profit of the product, and based on the probability of the user performing a specific action when the product is recommended and the stored profit of the product Then, an expected value of profit due to the product being recommended is calculated, and a product recommended for the user is selected based on the calculated expected value.
  • the product recommendation processing program stored in the recording medium stores the profit of the product, and the probability that the user performs a specific action when the product is recommended, and the stored product Based on the profit, an expected value of profit due to the recommendation of the product is calculated, and based on the calculated expected value, a product for selecting a product recommended for the user is executed by a computer.
  • the effect of the present invention is that a product can be recommended so as to increase the profit of the operator.
  • FIG. 5 is a block diagram showing a configuration combining the first to fifth embodiments of the present invention.
  • FIG. 1 is a hardware configuration diagram of a product recommendation system 1 according to the first embodiment of the present invention.
  • a product recommendation system 1 includes a CPU 10, a memory 12, a hard disk drive (HDD) 14, a communication interface (IF) 16 that performs data communication via a network (not shown), a display device 18 such as a display, and the like.
  • the input device 20 includes a pointing device such as a keyboard and a mouse. These components are connected to each other through the bus 22 and input / output data to / from each other.
  • FIG. 2 is a block diagram illustrating a configuration example of the product recommendation system according to the first embodiment.
  • the product recommendation system includes an expected profit value calculation unit 102, a recommended product selection unit 104, and a product profit storage unit 202.
  • the CPU 10 in FIG. 1 executes various processes as the expected profit calculation unit 102 and the recommended product selection unit 104 in FIG.
  • a program for controlling processing executed by the CPU 10 is stored in the HDD 14.
  • the merchandise profit storage unit 202 stores the profit of the merchandise when the merchandise is sold for each merchandise.
  • the profit of the product is, for example, a value obtained by subtracting the purchase amount from the sales amount per product.
  • the profit of the product may be a profit rate.
  • the profit ratio is a value obtained by subtracting the purchase amount from the sales amount and dividing by the sales amount.
  • the profit of the product may be other values.
  • the merchandise profit storage unit 202 may be realized by either the memory 12 or the HDD 14 (FIG. 1). Based on the profit of the product stored in the product profit storage unit 202 and the probability that the user will perform a specific action when recommending the product (hereinafter referred to as “behavior probability”), An expected value of profits by recommending a product is calculated for each product.
  • the action probability is, for example, a purchase probability of a product or a browsing probability of a page on the Internet on which product information is posted. In the following description of the present embodiment, it is assumed that the action probability is a purchase probability for convenience of description.
  • the expected profit calculation unit 102 calculates the expected value of profit of the product i, for example, by the following method.
  • the expected profit calculation unit 102 receives a purchase probability Pi in a user group that has recommended a product i and a purchase probability Pi ′ in a user group that has not recommended the product i.
  • a component not shown in FIG. 2 may calculate Pi and Pi ′, and input them to the profit expectation calculation unit 102.
  • FIG. 3 is an example of Pi, Pi ′, ⁇ Pi input to the expected profit calculation unit 102.
  • the expected profit calculation unit 102 refers to the product profit storage unit 202 and acquires the profit Qi of the product of the product i.
  • FIG. 4 is an example of merchandise profits stored in the merchandise profit storage unit 202.
  • the recommended product selection unit 104 selects a product to be recommended to the user based on the expected profit value of each product calculated by the expected profit value calculation unit 102.
  • the recommended product selection unit 104 may select a recommended product from the products stored in the product profit storage unit 202 or the storage unit (not shown).
  • the “storage unit not shown” may be, for example, a storage unit on a server of a company that operates a mail order site on a network.
  • the recommended product selection unit 104 selects, for example, the top N products with high profit expectation values as products recommended to the user. Further, the recommended product selection unit 104 may select a product whose expected profit value is higher than a certain value as a product recommended to the user.
  • the recommended product selection unit 104 may output the selected product to the display device 18 (FIG. 1).
  • the expected profit value is calculated by multiplying ⁇ Pi and Qi, but is not limited to this calculation method.
  • the expected profit value calculation unit 102 may calculate the expected profit value by simply multiplying Pi and Qi without considering the difference in purchase probability.
  • the expected profit value calculation unit 102 may calculate the expected profit value by a method other than multiplication of both based on Pi and Qi.
  • FIG. 5 is a flowchart showing an example of the operation of the first embodiment.
  • the product recommendation system starts the following operation when receiving a user instruction or voluntarily transmitting product information.
  • the expected profit calculation unit 102 acquires the action probability of each product (step 11).
  • the expected profit value calculation unit 102 refers to the product profit storage unit 202 and acquires the profit of each product (step 12).
  • the expected profit value calculation unit 102 calculates the expected profit value of each product (step 13).
  • the recommended product selection unit 104 selects a product recommended for the user based on the expected profit value of each product (step 14).
  • the product can be recommended so that the profit of the operator is increased.
  • the expected profit value calculation unit 102 calculates an expected value of profits by recommending a product
  • the recommended product selection unit 104 selects a recommended product based on the calculated expected value.
  • the expected profit calculation unit 102 calculates the expected profit based on the difference between the purchase probability and the case where the recommendation is not recommended, the expected profit of the product that can be sold without recommendation is small. Therefore, the product is not recommended.
  • FIG. FIG. 6 is a block diagram illustrating a configuration example of the product recommendation system according to the second embodiment.
  • the product recommendation system according to the second embodiment further includes an action probability calculation unit 106 and an action history storage unit 204 in addition to the configuration of the first embodiment.
  • the action history storage unit 204 stores a user action history.
  • the action history may be, for example, a history of operations performed by a user on a predetermined web page on the Internet.
  • the action history may be a browsing history of a page on the Internet on which information on a certain product is posted, a purchase history of a product through the Internet, or the like.
  • the action history storage unit 204 may be realized by at least one of the memory 12 and the HDD 14 (FIG. 1).
  • the action history may be stored in either the memory 12 or the HDD 14 (FIG. 1) by the CPU 10 (FIG. 1), for example.
  • the action probability calculation unit 106 calculates an action probability based on the action history stored in the action history storage unit 204.
  • the behavior probability calculation unit 106 refers to the behavior history storage unit 204 to calculate the purchase probability Pi in the user group that recommended the product i and the purchase probability Pi ′ in the user group that did not recommend the product i.
  • a product can be recommended using an action probability with high accuracy. This is because the action probability calculation unit 106 calculates an action probability using an appropriate action history. For example, assume a case where a business operator recommends a product to a specific user layer. The action probability calculation unit 106 calculates an action probability based only on the action history of a specific user layer. Then, it is possible to obtain an action probability with higher accuracy than when action probabilities for all users are determined in advance for each product.
  • FIG. 7 is a block diagram illustrating a configuration example of the product recommendation system according to the third embodiment.
  • the product recommendation system according to the third embodiment further includes a recommended product display unit 108 and an action acquisition unit 110 in addition to the configuration of the second embodiment.
  • the recommended product display unit 108 and the action acquisition unit 110 may be realized by the CPU 10 (FIG. 1).
  • the recommended product display unit 108 displays the product selected by the recommended product selection unit 104 on the display device 18 (FIG. 1).
  • the behavior acquisition unit 110 acquires the presence / absence of recommendation of a certain product to the user and the behavior history, and stores the information in the behavior history storage unit 204.
  • the action acquisition unit 110 acquires information indicating that the product i has been recommended and that the product i has been purchased from the operation history via the input device 20 (FIG. 1). Furthermore, the action acquisition unit 110 stores information “product i, recommended, purchased” in the action history storage unit 204. The action acquisition unit 110 may acquire the action history from another system such as a sensor instead of the operation history via the input device 20. Next, the effect of this embodiment will be described. In the present embodiment, a product can be recommended using an action probability corresponding to a temporal change in preference.
  • FIG. FIG. 8 is a block diagram illustrating a configuration example of a product recommendation system according to the fourth embodiment.
  • the product recommendation system according to the fourth embodiment further includes a preference product selection unit 112 in addition to the configuration of the second embodiment.
  • the preference product selection unit 112 selects a product that the user likes based on the behavior history stored in the behavior history storage unit 204.
  • the expected profit value calculation unit 102 calculates an expected profit value for the product selected by the preference product selection unit 112.
  • the preference product selection unit 112 selects a product having a tendency similar to the product purchased by the user from the products stored in the product profit storage unit 202.
  • the preference product selection unit 112 selects a product purchased by another user based on the behavior history of another user who has a preference tendency similar to that of the specific user stored in the behavior history storage unit 204. You may select as goods which a specific user likes.
  • the effect of this embodiment will be described.
  • FIG. FIG. 9 is a block diagram illustrating a configuration example of a product recommendation system according to the fifth embodiment.
  • the product recommendation system according to the fifth embodiment further includes a product profit setting unit 114 in addition to the configuration of the first embodiment.
  • the merchandise profit setting unit 114 calculates the profit of the merchandise and stores it in the merchandise profit storage unit 202.
  • the product profit setting unit 114 calculates the profit of the product using the inventory amount information.
  • the profit of the product is made larger than the current value.
  • the profit of the product is made smaller than the current value.
  • the product profit setting unit 114 may calculate the profit of the product for each combination of the product and the user group. An example of the profit of the product is shown in FIG. In the table of FIG. 10, when the product 1 is a product that is not recommended for those under the age of 18, the profit of the product is set to 0 for a user under the age of 18.
  • the profit of the product can be set more appropriately.
  • the profit of a product is set using the information on the inventory amount, the product can be recommended while appropriately controlling the inventory amount.
  • the profit of a product for each combination of a product and a user group it is possible to recommend the product in consideration of legal and social ethical benefits and constraints.
  • the configurations of the respective embodiments can be arbitrarily combined.
  • FIG. 11 is a block diagram illustrating a configuration example of a product recommendation system in which all the embodiments are combined.
  • the “product” in the present invention is not limited to a tangible object, but is a concept including a moving image distributed through a network, a service provided by a service provider, and the like. Further, the “product” does not necessarily have to be traded for a fee. Furthermore, “benefit” is a concept that includes not only economic benefits, but also legal and social ethics benefits and favorability. Therefore, for example, a moving image recommendation system in a moving image distribution site performed by a network operator is also included in the scope of the present invention. Furthermore, a system for recommending a service that increases the favorability with respect to a service provided by a certain company for free is also included in the scope of the present invention.
  • the program may be any program that causes the computer to execute each operation described above. While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2010-229227 for which it applied on October 12, 2010, and takes in those the indications of all here.

Abstract

The present invention provides a product recommendation system capable of recommending a product so that profits of merchants increase. The product recommendation system of the present invention is provided with product profit storage means for storing profit of a product, profit expected value calculation means for calculating an expected value of profit as a result of the recommending of the product on the basis of the probability that a user will perform a specific action if the product is recommended and the profit of the product which has been stored, and recommended product selection means for selecting a product to recommend to the user on the basis of the calculated expected value.

Description

商品推薦システムおよび商品推薦方法Product recommendation system and product recommendation method
　本発明は、商品を推薦する商品推薦システムおよび商品推薦方法に関する。 The present invention relates to a product recommendation system and a product recommendation method for recommending products.
　商品等を推薦するシステムでは、ユーザ個々の嗜好を予測し、それに合わせて情報や商品を提示する技術が存在する。例えば、特許文献１では、ユーザの閲覧履歴に基づいて、そのユーザに配信する情報を決定する情報提供システムが開示されている。
　また、本発明の関連技術として、特許文献２では、顧客が複数店舗中から希望する店舗で希望する商品を希望する価格で注文できるシステムが開示されている。また、特許文献３では、見積業務を適切に遂行可能な見積方法が開示されている。また、特許文献４では、商品を購入しようとするユーザのニーズを的確に反映して商品を推薦することができる商品推薦システムが開示されている。また、特許文献５では、高い利益を取得可能な顧客を優先的にターゲティングできるようにし、そのターゲティングした顧客に応じたナレッジ情報や、販売支援情報を送信して顧客訪問時に活用することを可能とする顧客ターゲティングシステムが開示されている。 In a system for recommending products and the like, there is a technology for predicting individual user preferences and presenting information and products in accordance with the prediction. For example, Patent Document 1 discloses an information providing system that determines information to be distributed to a user based on the user's browsing history.
In addition, as a related technique of the present invention, Patent Document 2 discloses a system in which a customer can order a desired product at a desired price from a plurality of stores at a desired price. Further, Patent Document 3 discloses an estimation method capable of appropriately performing estimation work. Further, Patent Document 4 discloses a product recommendation system capable of recommending a product that accurately reflects the needs of a user who intends to purchase the product. In addition, Patent Document 5 enables customers who can obtain high profits to be preferentially targeted, and can transmit knowledge information corresponding to the targeted customers and sales support information to be used when visiting customers. A customer targeting system is disclosed.
特開２００９−０８７１５５JP2009-087155 特開２００２−１０９３３４JP 2002-109334 A 特開２００２−２５９５２４JP 2002-259524 A 特開２００３−１５０８３５JP 2003-150835 A 特開２００４−２９５５４７JP 2004-295547 A
　しかし、特許文献１~５に開示された技術では、事業者の利益が大きくなるように、商品を推薦できていないという問題点があった。ユーザの嗜好にあう商品が、利益が大きい商品とは限らないためである。
　［発明の目的］
　本発明の目的は、事業者の利益が大きくなるように、商品を推薦することができる商品推薦システムおよび商品推薦方法を提供することにある。 However, the techniques disclosed in Patent Documents 1 to 5 have a problem that a product cannot be recommended so as to increase the profit of the operator. This is because a product that meets the user's preference is not necessarily a product with a large profit.
[Object of invention]
An object of the present invention is to provide a product recommendation system and a product recommendation method capable of recommending a product so that the profit of the operator is increased.
　かかる目的を達成するため本発明の一形態である商品推薦システムは、商品の利益を記憶する商品利益記憶手段と、商品が推薦された場合にユーザが特定の行動を行う確率と、前記記憶された商品の利益とに基づいて、商品が推薦されることによる利益の期待値を算出する利益期待値算出手段と、前記算出された期待値に基づいて、ユーザに対して推薦する商品を選択する推薦商品選択手段とを備える。
　また、本発明の他の形態である商品推薦処理方法は、商品の利益を記憶させ、商品が推薦された場合にユーザが特定の行動を行う確率と、前記記憶された商品の利益とに基づいて、商品が推薦されることによる利益の期待値を算出し、前記算出された期待値に基づいて、ユーザに対して推薦する商品を選択する。
　また、本発明の他の形態である記録媒体が格納する商品推薦処理プログラムは、商品の利益を記憶させ、商品が推薦された場合にユーザが特定の行動を行う確率と、前記記憶された商品の利益とに基づいて、商品が推薦されることによる利益の期待値を算出し、前記算出された期待値に基づいて、ユーザに対して推薦する商品を選択する、処理をコンピュータに実行させる。 In order to achieve such an object, the product recommendation system according to one aspect of the present invention includes a product profit storage means for storing the profit of the product, the probability that the user will perform a specific action when the product is recommended, and the stored memory. Based on the profit of the product, the profit expected value calculation means for calculating the expected value of the profit due to the recommended product, and the product recommended for the user is selected based on the calculated expected value And recommended product selection means.
Further, the product recommendation processing method according to another aspect of the present invention stores the profit of the product, and based on the probability of the user performing a specific action when the product is recommended and the stored profit of the product Then, an expected value of profit due to the product being recommended is calculated, and a product recommended for the user is selected based on the calculated expected value.
The product recommendation processing program stored in the recording medium according to another aspect of the present invention stores the profit of the product, and the probability that the user performs a specific action when the product is recommended, and the stored product Based on the profit, an expected value of profit due to the recommendation of the product is calculated, and based on the calculated expected value, a product for selecting a product recommended for the user is executed by a computer.
　本発明の効果は、事業者の利益が大きくなるように、商品を推薦できることである。 The effect of the present invention is that a product can be recommended so as to increase the profit of the operator.
本発明の第１の実施形態に係るハードウェア構成図である。It is a hardware block diagram concerning the 1st embodiment of the present invention. 本発明の第１の実施の形態の構成を示すブロック図である。It is a block diagram which shows the structure of the 1st Embodiment of this invention. 本発明の第１の実施の形態における購買確率などの例である。It is an example, such as the purchase probability in the 1st Embodiment of this invention. 本発明の第１の実施の形態における商品の利益の例である。It is an example of the profit of the goods in the 1st Embodiment of this invention. 本発明の第１の実施の形態の動作を示す流れ図である。It is a flowchart which shows the operation | movement of the 1st Embodiment of this invention. 本発明の第２の実施の形態の構成を示すブロック図である。It is a block diagram which shows the structure of the 2nd Embodiment of this invention. 本発明の第３の実施の形態の構成を示すブロック図である。It is a block diagram which shows the structure of the 3rd Embodiment of this invention. 本発明の第４の実施の形態の構成を示すブロック図である。It is a block diagram which shows the structure of the 4th Embodiment of this invention. 本発明の第５の実施の形態の構成を示すブロック図である。It is a block diagram which shows the structure of the 5th Embodiment of this invention. 本発明の第５の実施の形態における商品の利益の例である。It is an example of the profit of the goods in the 5th Embodiment of this invention. 本発明の第１~第５の実施の形態を組み合わせた構成を示すブロック図である。FIG. 5 is a block diagram showing a configuration combining the first to fifth embodiments of the present invention.
　次に、発明を実施するための最良の形態について図面を参照して詳細に説明する。
　実施の形態１．
　図１は、本発明の第１の実施形態に係る商品推薦システム１のハードウェア構成図である。図１に示すように、商品推薦システム１は、ＣＰＵ１０、メモリ１２、ハードディスクドライブ（ＨＤＤ）１４、図示しないネットワークを介してデータの通信を行う通信インタフェース（ＩＦ）１６、ディスプレイ等の表示装置１８およびキーボードやマウス等のポインティングデバイスを含む入力装置２０を有する。これらの構成要素は、バス２２を通して互いに接続されており、互いにデータの入出力を行う。
　図２は、第１の実施の形態の商品推薦システムの構成例を示すブロック図である。図２に示すように、商品推薦システムは、利益期待値算出部１０２、推薦商品選択部１０４、および商品利益記憶部２０２を有する。ここで、図１のＣＰＵ１０は、図２における利益期待値算出部１０２および推薦商品選択部１０４として各種処理を実行する。ＣＰＵ１０が実行する処理を制御するためのプログラムは、ＨＤＤ１４に記憶される。
　商品利益記憶部２０２は、商品を販売した際の商品の利益を、商品ごとに記憶している。商品の利益は、例えば、商品１個あたりの、売上額から仕入額を引いた値である。また、商品の利益は、利益率であってもよい。利益率とは、売上額から仕入額を引いた値を売上額で割った値である。商品の利益は、その他の価でもよい。なお、商品利益記憶部２０２は、メモリ１２およびＨＤＤ１４（図１）のいずれかにより実現されても良い。
　利益期待値算出部１０２は、商品利益記憶部２０２が記憶している商品の利益と、商品を推薦した場合にユーザが特定の行動を行う確率（以下、「行動確率」）とに基づいて、商品を推薦することによる利益の期待値を、商品ごとに算出する。行動確率とは、例えば、商品の購買確率や、商品の情報が掲載されているインターネット上のページの閲覧確率である。なお、本実施の形態における以降の説明では、説明の便宜のため行動確率は購買確率であるとして説明する。
　利益期待値算出部１０２は、例えば次の方法で商品ｉの利益の期待値を算出する。
　利益期待値算出部１０２には、商品ｉを推薦したユーザ群における購買確率Ｐｉと、商品ｉを推薦しなかったユーザ群における購買確率Ｐｉ’が入力される。ここでＰｉおよびＰｉ’については、図２に図示しない構成部がＰｉおよびＰｉ’を算出し、利益期待算出部１０２に入力しても良い。利益期待値算出部１０２は、その差ΔＰｉ＝Ｐｉ−Ｐｉ’を算出する。なお、利益期待値算出部１０２に、ΔＰｉの値が入力されてもよい。図３は、利益期待値算出部１０２に入力される、Ｐｉ、Ｐｉ’、ΔＰｉ、の例である。
　次に、利益期待値算出部１０２は、商品利益記憶部２０２を参照して、商品ｉの商品の利益Ｑｉを取得する。図４は、商品利益記憶部２０２が記憶している、商品の利益の例である。
　利益期待値算出部１０２は、ΔＰｉとＱｉに基づいて、商品ｉを推薦することによる利益期待値Ｇｉ＝ΔＰｉ×Ｑｉを算出する。
　推薦商品選択部１０４は、利益期待値算出部１０２で算出した各商品の利益期待値に基づいて、ユーザに推薦する商品を選択する。推薦商品選択部１０４は、商品利益記憶部２０２又は図示しない記憶部が記憶している商品の中から推薦する商品を選択しても良い。ここで「図示しない記憶部」とは、例えばネットワーク上の通信販売サイトを運営している企業のサーバ上の記憶部等でも良い。推薦商品選択部１０４は、例えば、利益期待値が高い上位Ｎ個の商品を、ユーザに推薦する商品として選択する。また、推薦商品選択部１０４は、利益期待値が一定値より高い商品を、ユーザに推薦する商品として選択してもよい。
　推薦商品選択部１０４は、選択した商品を表示装置１８（図１）に出力しても良い。
　なお、本実施形態においては、利益期待値を、ΔＰｉ及びＱｉを乗算することで算出したが、この計算方法に限定されない。例えば利益期待値算出部１０２は、購買確率の差分は考慮せず、単純にＰｉ及びＱｉを乗算することで利益期待値を算出しても良い。利益期待値算出部１０２は、Ｐｉ及びＱｉに基づいて、両者の乗算以外の方法で利益期待値を算出しても良い。
　次に、本発明を実施するための第１の実施の形態の動作について詳細に説明する。
　図５は、第１の実施の形態の動作の一例を示す流れ図である。
　商品推薦システムは、ユーザの指示を受け、または自発的に商品の情報を発信したいときに、以下の動作を開始する。商品推薦システムにおいて、利益期待値算出部１０２は、各商品の行動確率を取得する（ステップ１１）。また、利益期待値算出部１０２は、商品利益記憶部２０２を参照して、各商品の利益を取得する（ステップ１２）。次に、利益期待値算出部１０２は、各商品の利益期待値を算出する（ステップ１３）。さらに、推薦商品選択部１０４は、各商品の利益期待値に基づいて、ユーザに推薦する商品を選択する（ステップ１４）。
　次に、本実施の形態の効果について説明する。
　本実施の形態では、事業者の利益が大きくなるように、商品を推薦することができる。利益期待値算出部１０２が、商品を推薦することによる利益の期待値を算出し、推薦商品選択部１０４が、算出された期待値に基づいて推薦する商品を選択しているためである。
　また、特に、利益期待値算出部１０２が、推薦した場合としない場合の購買確率の差に基づいて利益の期待値を算出する場合は、推薦しなくても売れる商品の利益の期待値が小さくなるため、その商品は推薦されない。ここで、商品推薦システムが、推薦しなくても売れる商品を改めて推薦する必要はない。よって、購買確率の差に基づいて利益の期待値を算出する場合、本実施の形態に係る商品推薦システムによれば、より適切に商品を推薦することができる。
実施の形態２．
　図６は、第２の実施の形態の商品推薦システムの構成例を示すブロック図である。第２の実施の形態に係る商品推薦システムは、第１の実施の形態の構成に加えて、行動確率算出部１０６と、行動履歴記憶部２０４をさらに有する。
　行動履歴記憶部２０４は、ユーザの行動履歴を記憶する。行動履歴とは、例えばユーザが、インターネット上の所定のウェブページにおける操作の履歴でも良い。具体的には、行動履歴は、ある商品の情報が掲載されているインターネット上のページの閲覧履歴や、インターネットを通じた商品の購入履歴などでも良い。なお、行動履歴記憶部２０４は、メモリ１２またはＨＤＤ１４（図１）の少なくともいずれかにより実現されても良い。行動履歴は、例えばＣＰＵ１０（図１）によりメモリ１２またはＨＤＤ１４（図１）のいずれかに格納されても良い。
　行動確率算出部１０６は、行動履歴記憶部２０４に記憶された行動履歴に基づいて、行動確率を算出する。例えば、行動確率算出部１０６は、行動履歴記憶部２０４を参照して、商品ｉを推薦したユーザ群における購買確率Ｐｉと、商品ｉを推薦しなかったユーザ群における購買確率Ｐｉ’を算出する。
　次に、本実施の形態の効果について説明する。
　本実施の形態では、精度が高い行動確率を利用して、商品を推薦することができる。行動確率算出部１０６が、適切な行動履歴を利用して行動確率を算出しているためである。例えば、事業者が、特定のユーザ層に対して商品を推薦する場合を想定する。行動確率算出部１０６が、特定のユーザ層の行動履歴のみに基づいて行動確率を算出する。すると、予め、全ユーザについての行動確率が商品ごとに定められている場合より、精度が高い行動確率を求めることができる。
実施の形態３．
　図７は、第３の実施の形態の商品推薦システムの構成例を示すブロック図である。第３の実施の形態に係る商品推薦システムは、第２の実施の形態の構成に加えて、推薦商品表示部１０８と、行動取得部１１０とをさらに有する。推薦商品表示部１０８および行動取得部１１０は、ＣＰＵ１０（図１）により実現されても良い。
　推薦商品表示部１０８は、推薦商品選択部１０４で選択された商品を表示装置１８（図１）に表示する。
　行動取得部１１０は、ユーザへのある商品の推薦の有無及び行動履歴を取得し、行動履歴記憶部２０４にその情報を記憶させる。例えば、商品ｉを推薦商品表示部１０８でユーザに示した後、一定期間内に、ユーザが商品ｉをインターネットを通じて購入した場合を想定する。その場合、行動取得部１１０は、入力装置２０（図１）を介した操作履歴から、商品ｉを推薦したことと、商品ｉを購入したことの情報を取得する。さらに、行動取得部１１０は、「商品ｉ、推薦あり、購入」という情報を、行動履歴記憶部２０４に記憶させる。なお、行動取得部１１０は、行動履歴を、入力装置２０を介した操作履歴でなく、センサー等の他システムから取得してもよい。
　次に、本実施の形態の効果について説明する。
　本実施の形態では、嗜好の時期的変化に対応した行動確率を利用して、商品を推薦することができる。行動取得部１１０が、行動履歴記憶部２０４に、自動的に、最新の行動履歴を追加するためである。
実施の形態４．
　図８は、第４の実施の形態の商品推薦システムの構成例を示すブロック図である。第４の実施の形態に係る商品推薦システムは、第２の実施の形態の構成に加えて、嗜好商品選択部１１２をさらに有する。
　嗜好商品選択部１１２は、行動履歴記憶部２０４に記憶された行動履歴に基づいて、ユーザが好む商品を選択する。そして、利益期待値算出部１０２は、嗜好商品選択部１１２で選択された商品について、利益期待値を算出する。
　例えば、嗜好商品選択部１１２は、商品利益記憶部２０２に記憶された商品の中から、そのユーザが購買した商品と似た傾向の商品を選択する。また、嗜好商品選択部１１２は、行動履歴記憶部２０４に記憶された、特定のユーザと嗜好の傾向が似ている他のユーザの行動履歴に基づいて、他のユーザが購入した商品を、その特定のユーザが好む商品として選択してもよい。
　次に、本実施の形態の効果について説明する。
　本実施の形態では、利益が大きく、かつ、ユーザの嗜好にも適した商品を推薦することができる。嗜好商品選択部１１２が、ユーザが好む商品を絞り込むためである。
実施の形態５．
　図９は、第５の実施の形態の商品推薦システムの構成例を示すブロック図である。第５の実施の形態係る商品推薦システムは、第１の実施の形態の構成に加えて、商品利益設定部１１４をさらに有する。
　商品利益設定部１１４は、商品の利益を算出し、商品利益記憶部２０２に記憶させる。
　例えば、商品利益設定部１１４は、在庫量の情報を用いて、商品の利益を算出する。在庫量が適正水準より高いときは、商品の利益を、現在の値より大きくする。一方、在庫量が適正水準より低いときは、商品の利益を、現在の値より小さくする。
　また、商品利益設定部１１４は、商品とユーザ群の組み合わせ毎に、商品の利益を算出してもよい。商品の利益の例を、図１０に示す。図１０の表では、商品１が１８歳未満に非推奨の商品である場合、１８歳未満のユーザに対しては、その商品の利益を０と設定する。
　次に、本実施の形態の効果について説明する。
　本実施の形態では、商品の利益を、より適切に設定することができる。例えば、在庫量の情報を利用して商品の利益を設定する場合は、在庫量を適切にコントロールしながら、商品を推薦することができる。また、商品とユーザ群の組み合わせ毎に、商品の利益を設定する場合は、法的、社会的倫理に関する利益や制約条件も考慮した、商品を推薦することができる。
　なお、実施の形態１~実施の形態５について、任意に、各実施の形態の構成を組み合わせることができる。例えば、図１１は、全ての実施の形態を組み合わせた商品推薦システムの構成例を示すブロック図である。
　なお、本発明における「商品」とは、有体物に限られるものでなく、ネットワークを通じて配信される動画像や、サービス事業者が提供するサービスなども含む概念である。また、「商品」は、必ずしも有償で取引されなくてもよい。さらに、「利益」とは、経済的利益だけでなく、法的・社会的倫理に関する利益や、好感度も含む概念である。
　よって、例えば、ネットワーク事業者が行う、動画像配信サイトにおける動画像推薦システムも、本発明の範囲に含まれる。さらに、ある事業者が無償で提供するサービスについて、その事業者に対しての好感度が高まるようなサービスを推薦するシステムも、本発明の範囲に含まれる。
　なお、本発明をプログラム及びプログラムを実行するコンピュータによって実現する場合、該プログラムはこれまでに説明した各動作をコンピュータに実行させるものであればよい。
　以上、実施の形態を参照して本願発明を説明したが、本願発明は以上の実施形態に限定されるものではない。本願発明の構成や詳細には、本願発明のスコープ内で同業者が理解し得る様々な変更をすることができる。
　この出願は、２０１０年１０月１２日に出願された日本出願特願２０１０−２２９２２７を基礎とする優先権を主張し、その開示の全てをここに取り込む。 Next, the best mode for carrying out the invention will be described in detail with reference to the drawings.
Embodiment 1 FIG.
FIG. 1 is a hardware configuration diagram of a product recommendation system 1 according to the first embodiment of the present invention. As shown in FIG. 1, a product recommendation system 1 includes a CPU 10, a memory 12, a hard disk drive (HDD) 14, a communication interface (IF) 16 that performs data communication via a network (not shown), a display device 18 such as a display, and the like. The input device 20 includes a pointing device such as a keyboard and a mouse. These components are connected to each other through the bus 22 and input / output data to / from each other.
FIG. 2 is a block diagram illustrating a configuration example of the product recommendation system according to the first embodiment. As illustrated in FIG. 2, the product recommendation system includes an expected profit value calculation unit 102, a recommended product selection unit 104, and a product profit storage unit 202. Here, the CPU 10 in FIG. 1 executes various processes as the expected profit calculation unit 102 and the recommended product selection unit 104 in FIG. A program for controlling processing executed by the CPU 10 is stored in the HDD 14.
The merchandise profit storage unit 202 stores the profit of the merchandise when the merchandise is sold for each merchandise. The profit of the product is, for example, a value obtained by subtracting the purchase amount from the sales amount per product. The profit of the product may be a profit rate. The profit ratio is a value obtained by subtracting the purchase amount from the sales amount and dividing by the sales amount. The profit of the product may be other values. The merchandise profit storage unit 202 may be realized by either the memory 12 or the HDD 14 (FIG. 1).
Based on the profit of the product stored in the product profit storage unit 202 and the probability that the user will perform a specific action when recommending the product (hereinafter referred to as “behavior probability”), An expected value of profits by recommending a product is calculated for each product. The action probability is, for example, a purchase probability of a product or a browsing probability of a page on the Internet on which product information is posted. In the following description of the present embodiment, it is assumed that the action probability is a purchase probability for convenience of description.
The expected profit calculation unit 102 calculates the expected value of profit of the product i, for example, by the following method.
The expected profit calculation unit 102 receives a purchase probability Pi in a user group that has recommended a product i and a purchase probability Pi ′ in a user group that has not recommended the product i. Here, with respect to Pi and Pi ′, a component not shown in FIG. 2 may calculate Pi and Pi ′, and input them to the profit expectation calculation unit 102. The expected profit calculation unit 102 calculates the difference ΔPi = Pi−Pi ′. Note that the value of ΔPi may be input to the expected profit calculation unit 102. FIG. 3 is an example of Pi, Pi ′, ΔPi input to the expected profit calculation unit 102.
Next, the expected profit calculation unit 102 refers to the product profit storage unit 202 and acquires the profit Qi of the product of the product i. FIG. 4 is an example of merchandise profits stored in the merchandise profit storage unit 202.
Based on ΔPi and Qi, expected profit value calculation unit 102 calculates expected profit value Gi = ΔPi × Qi by recommending product i.
The recommended product selection unit 104 selects a product to be recommended to the user based on the expected profit value of each product calculated by the expected profit value calculation unit 102. The recommended product selection unit 104 may select a recommended product from the products stored in the product profit storage unit 202 or the storage unit (not shown). Here, the “storage unit not shown” may be, for example, a storage unit on a server of a company that operates a mail order site on a network. The recommended product selection unit 104 selects, for example, the top N products with high profit expectation values as products recommended to the user. Further, the recommended product selection unit 104 may select a product whose expected profit value is higher than a certain value as a product recommended to the user.
The recommended product selection unit 104 may output the selected product to the display device 18 (FIG. 1).
In the present embodiment, the expected profit value is calculated by multiplying ΔPi and Qi, but is not limited to this calculation method. For example, the expected profit value calculation unit 102 may calculate the expected profit value by simply multiplying Pi and Qi without considering the difference in purchase probability. The expected profit value calculation unit 102 may calculate the expected profit value by a method other than multiplication of both based on Pi and Qi.
Next, the operation of the first embodiment for carrying out the present invention will be described in detail.
FIG. 5 is a flowchart showing an example of the operation of the first embodiment.
The product recommendation system starts the following operation when receiving a user instruction or voluntarily transmitting product information. In the product recommendation system, the expected profit calculation unit 102 acquires the action probability of each product (step 11). Further, the expected profit value calculation unit 102 refers to the product profit storage unit 202 and acquires the profit of each product (step 12). Next, the expected profit value calculation unit 102 calculates the expected profit value of each product (step 13). Further, the recommended product selection unit 104 selects a product recommended for the user based on the expected profit value of each product (step 14).
Next, the effect of this embodiment will be described.
In the present embodiment, the product can be recommended so that the profit of the operator is increased. This is because the expected profit value calculation unit 102 calculates an expected value of profits by recommending a product, and the recommended product selection unit 104 selects a recommended product based on the calculated expected value.
In particular, when the expected profit calculation unit 102 calculates the expected profit based on the difference between the purchase probability and the case where the recommendation is not recommended, the expected profit of the product that can be sold without recommendation is small. Therefore, the product is not recommended. Here, it is not necessary for the product recommendation system to recommend products that can be sold without recommendation. Therefore, when the expected value of profit is calculated based on the difference in purchase probability, the product recommendation system according to the present embodiment can more appropriately recommend the product.
Embodiment 2. FIG.
FIG. 6 is a block diagram illustrating a configuration example of the product recommendation system according to the second embodiment. The product recommendation system according to the second embodiment further includes an action probability calculation unit 106 and an action history storage unit 204 in addition to the configuration of the first embodiment.
The action history storage unit 204 stores a user action history. The action history may be, for example, a history of operations performed by a user on a predetermined web page on the Internet. Specifically, the action history may be a browsing history of a page on the Internet on which information on a certain product is posted, a purchase history of a product through the Internet, or the like. The action history storage unit 204 may be realized by at least one of the memory 12 and the HDD 14 (FIG. 1). The action history may be stored in either the memory 12 or the HDD 14 (FIG. 1) by the CPU 10 (FIG. 1), for example.
The action probability calculation unit 106 calculates an action probability based on the action history stored in the action history storage unit 204. For example, the behavior probability calculation unit 106 refers to the behavior history storage unit 204 to calculate the purchase probability Pi in the user group that recommended the product i and the purchase probability Pi ′ in the user group that did not recommend the product i.
Next, the effect of this embodiment will be described.
In the present embodiment, a product can be recommended using an action probability with high accuracy. This is because the action probability calculation unit 106 calculates an action probability using an appropriate action history. For example, assume a case where a business operator recommends a product to a specific user layer. The action probability calculation unit 106 calculates an action probability based only on the action history of a specific user layer. Then, it is possible to obtain an action probability with higher accuracy than when action probabilities for all users are determined in advance for each product.
Embodiment 3 FIG.
FIG. 7 is a block diagram illustrating a configuration example of the product recommendation system according to the third embodiment. The product recommendation system according to the third embodiment further includes a recommended product display unit 108 and an action acquisition unit 110 in addition to the configuration of the second embodiment. The recommended product display unit 108 and the action acquisition unit 110 may be realized by the CPU 10 (FIG. 1).
The recommended product display unit 108 displays the product selected by the recommended product selection unit 104 on the display device 18 (FIG. 1).
The behavior acquisition unit 110 acquires the presence / absence of recommendation of a certain product to the user and the behavior history, and stores the information in the behavior history storage unit 204. For example, it is assumed that the user purchases the product i through the Internet within a certain period after the product i is shown to the user by the recommended product display unit 108. In that case, the action acquisition unit 110 acquires information indicating that the product i has been recommended and that the product i has been purchased from the operation history via the input device 20 (FIG. 1). Furthermore, the action acquisition unit 110 stores information “product i, recommended, purchased” in the action history storage unit 204. The action acquisition unit 110 may acquire the action history from another system such as a sensor instead of the operation history via the input device 20.
Next, the effect of this embodiment will be described.
In the present embodiment, a product can be recommended using an action probability corresponding to a temporal change in preference. This is because the behavior acquisition unit 110 automatically adds the latest behavior history to the behavior history storage unit 204.
Embodiment 4 FIG.
FIG. 8 is a block diagram illustrating a configuration example of a product recommendation system according to the fourth embodiment. The product recommendation system according to the fourth embodiment further includes a preference product selection unit 112 in addition to the configuration of the second embodiment.
The preference product selection unit 112 selects a product that the user likes based on the behavior history stored in the behavior history storage unit 204. Then, the expected profit value calculation unit 102 calculates an expected profit value for the product selected by the preference product selection unit 112.
For example, the preference product selection unit 112 selects a product having a tendency similar to the product purchased by the user from the products stored in the product profit storage unit 202. In addition, the preference product selection unit 112 selects a product purchased by another user based on the behavior history of another user who has a preference tendency similar to that of the specific user stored in the behavior history storage unit 204. You may select as goods which a specific user likes.
Next, the effect of this embodiment will be described.
In the present embodiment, it is possible to recommend a product that has a large profit and is suitable for the user's preference. This is because the favorite product selection unit 112 narrows down the products that the user likes.
Embodiment 5 FIG.
FIG. 9 is a block diagram illustrating a configuration example of a product recommendation system according to the fifth embodiment. The product recommendation system according to the fifth embodiment further includes a product profit setting unit 114 in addition to the configuration of the first embodiment.
The merchandise profit setting unit 114 calculates the profit of the merchandise and stores it in the merchandise profit storage unit 202.
For example, the product profit setting unit 114 calculates the profit of the product using the inventory amount information. When the stock quantity is higher than the appropriate level, the profit of the product is made larger than the current value. On the other hand, when the inventory quantity is lower than the appropriate level, the profit of the product is made smaller than the current value.
The product profit setting unit 114 may calculate the profit of the product for each combination of the product and the user group. An example of the profit of the product is shown in FIG. In the table of FIG. 10, when the product 1 is a product that is not recommended for those under the age of 18, the profit of the product is set to 0 for a user under the age of 18.
Next, the effect of this embodiment will be described.
In the present embodiment, the profit of the product can be set more appropriately. For example, when the profit of a product is set using the information on the inventory amount, the product can be recommended while appropriately controlling the inventory amount. In addition, when setting the profit of a product for each combination of a product and a user group, it is possible to recommend the product in consideration of legal and social ethical benefits and constraints.
In addition, regarding Embodiments 1 to 5, the configurations of the respective embodiments can be arbitrarily combined. For example, FIG. 11 is a block diagram illustrating a configuration example of a product recommendation system in which all the embodiments are combined.
The “product” in the present invention is not limited to a tangible object, but is a concept including a moving image distributed through a network, a service provided by a service provider, and the like. Further, the “product” does not necessarily have to be traded for a fee. Furthermore, “benefit” is a concept that includes not only economic benefits, but also legal and social ethics benefits and favorability.
Therefore, for example, a moving image recommendation system in a moving image distribution site performed by a network operator is also included in the scope of the present invention. Furthermore, a system for recommending a service that increases the favorability with respect to a service provided by a certain company for free is also included in the scope of the present invention.
In the case where the present invention is realized by a program and a computer that executes the program, the program may be any program that causes the computer to execute each operation described above.
While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.
This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2010-229227 for which it applied on October 12, 2010, and takes in those the indications of all here.
　１　　商品推薦システム
　１０　ＣＰＵ
　１２　メモリ
　１４　ＨＤＤ
　１６　通信ＩＦ
　１８　表示装置
　２０　入力装置
　２２　バス
　１０２　利益期待値算出部
　１０４　推薦商品選択部
　１０６　行動確率算出部
　１０８　推薦商品表示部
　１１０　行動取得部
　１１２　嗜好商品選択部
　１１４　商品利益設定部
　２０２　商品利益記憶部
　２０４　行動履歴記憶部 1 Product recommendation system 10 CPU
12 Memory 14 HDD
16 Communication IF
18 Display device 20 Input device 22 Bus 102 Expected profit calculation unit 104 Recommended product selection unit 106 Behavior probability calculation unit 108 Recommended product display unit 110 Behavior acquisition unit 112 Preference product selection unit 114 Product profit setting unit 202 Product profit storage unit 204 Behavior History storage

Claims (10)

  1. 　商品の利益を記憶する商品利益記憶手段と、
    　商品が推薦された場合にユーザが特定の行動を行う確率と、前記記憶された商品の利益とに基づいて、商品が推薦されることによる利益の期待値を算出する利益期待値算出手段と、
    　前記算出された期待値に基づいて、ユーザに対して推薦する商品を選択する推薦商品選択手段と、
    　を備える商品推薦システム。     Merchandise profit memorizing means for memorizing profits of merchandise;
    Profit expected value calculation means for calculating an expected value of profit by recommending a product based on the probability of the user performing a specific action when the product is recommended and the stored profit of the product;
    Based on the calculated expected value, recommended product selection means for selecting a product recommended for the user;
    Product recommendation system with
  2. 　商品の推薦に対するユーザの行動履歴を記憶する行動履歴記憶手段と、
    　前記記憶された行動履歴に基づいて、前記行動を行う確率を算出する行動確率算出手段と、
    　をさらに備える請求項１に記載の商品推薦システム。     Action history storage means for storing a user's action history for product recommendation;
    Action probability calculating means for calculating a probability of performing the action based on the stored action history;
    The product recommendation system according to claim 1, further comprising:
  3. 　前記選択された商品を表示する推薦商品表示手段と、
    　前記表示された商品に対するユーザの行動を取得し、前記商品と前記ユーザの行動とを対応付けて、前記行動履歴記憶手段に記憶させる行動取得手段と、
    　をさらに備える請求項２に記載の商品推薦システム。     Recommended product display means for displaying the selected product;
    Action acquisition means for acquiring a user's action for the displayed product, associating the product with the user's action, and storing the action in the action history storage means;
    The product recommendation system according to claim 2, further comprising:
  4. 　前記記憶されている特定のユーザの行動履歴に基づいて、前記ユーザの嗜好にあう商品を選択する嗜好商品選択手段をさらに備え、
    　前記利益期待値算出手段は、前記嗜好商品選択手段により選択された商品について、利益の期待値を算出する、
    　請求項２または３に記載の商品推薦システム。     Based on the stored action history of the specific user, further comprising a preference product selection means for selecting a product that meets the user's preference,
    The profit expected value calculation means calculates an expected value of profit for the product selected by the preference product selection means.
    The product recommendation system according to claim 2 or 3.
  5. 　前記利益期待値算出手段は、商品を推薦した場合としない場合の、ユーザが特定の行動を行う確率の差と、前記記憶された商品の利益とに基づいて、商品を推薦することによる利益の期待値を算出する
    　請求項１乃至４のいずれか１項に記載の商品推薦システム。     The profit expectation value calculating means calculates the profit by recommending a product based on the difference in the probability that the user performs a specific action when the product is recommended and when the product is recommended and the stored profit of the product. The product recommendation system according to any one of claims 1 to 4, wherein an expected value is calculated.
  6. 　商品の在庫量に基づいて、前記商品の利益を設定する商品利益設定手段と、
    　をさらに備える請求項１乃至５のいずれか１項に記載の商品推薦システム。     Product profit setting means for setting the profit of the product based on the inventory amount of the product;
    The product recommendation system according to any one of claims 1 to 5, further comprising:
  7. 　ユーザの属性に基づいて、ユーザ毎に、前記商品の利益を設定する商品利益設定手段をさらに備える請求項１乃至５のいずれか１項に記載の商品推薦システム。 The product recommendation system according to any one of claims 1 to 5, further comprising product profit setting means for setting profit of the product for each user based on a user attribute.
  8. 　前記ユーザが特定の行動を行う確率は、ユーザが商品を購買する確率、または、ユーザが商品に関する情報を閲覧する確率の少なくともいずれか一方である
    　請求項１乃至７のいずれか１項に記載の商品推薦システム。     The probability that the user performs a specific action is at least one of a probability that the user purchases a product and a probability that the user views information related to the product. Product recommendation system.
  9. 　商品の利益を記憶させ、
    　商品が推薦された場合にユーザが特定の行動を行う確率と、前記記憶された商品の利益とに基づいて、商品が推薦されることによる利益の期待値を算出し、
    　前記算出された期待値に基づいて、ユーザに対して推薦する商品を選択する、
    　商品推薦方法。     Memorize the profit of the product,
    Based on the probability of the user performing a specific action when the product is recommended and the stored profit of the product, an expected value of profit due to the recommended product is calculated,
    Based on the calculated expected value, a product recommended for the user is selected.
    Product recommendation method.
  10. 　商品の利益を記憶させ、
    　商品が推薦された場合にユーザが特定の行動を行う確率と、前記記憶された商品の利益とに基づいて、商品が推薦されることによる利益の期待値を算出し、
    　前記算出された期待値に基づいて、ユーザに対して推薦する商品を選択する、
    　処理をコンピュータに実行させる商品推薦プログラムを格納する記録媒体。     Memorize the profit of the product,
    Based on the probability of the user performing a specific action when the product is recommended and the stored profit of the product, an expected value of profit due to the recommended product is calculated,
    Based on the calculated expected value, a product recommended for the user is selected.
    A recording medium for storing a product recommendation program that causes a computer to execute processing.
PCT/JP2011/072691 2010-10-12 2011-09-26 Product recommendation system and product recommendation method WO2012049987A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2012538633A JPWO2012049987A1 (en) 2010-10-12 2011-09-26 Product recommendation system, product recommendation method and program

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2010-229227 2010-10-12
JP2010229227 2010-10-12

Publications (1)

Publication Number Publication Date
WO2012049987A1 true WO2012049987A1 (en) 2012-04-19

Family

ID=45938217

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2011/072691 WO2012049987A1 (en) 2010-10-12 2011-09-26 Product recommendation system and product recommendation method

Country Status (2)

Country Link
JP (1) JPWO2012049987A1 (en)
WO (1) WO2012049987A1 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016035424A1 (en) * 2014-09-05 2016-03-10 ソニー株式会社 Information processing device, information processing method and program
JP6286076B1 (en) * 2017-01-18 2018-02-28 株式会社リクルートホールディングス Information management apparatus and program
JP2019144951A (en) * 2018-02-22 2019-08-29 オムロン株式会社 Recommendation information specifying apparatus, recommendation information specifying system, recommendation information specifying method, and program
CN111143678A (en) * 2019-12-27 2020-05-12 第四范式（北京）技术有限公司 Recommendation system and recommendation method
JP2021093216A (en) * 2021-03-11 2021-06-17 オムロン株式会社 Recommendation information specification device, recommendation information specification system, recommendation information specification method, terminal device, and program
JP2022031901A (en) * 2021-03-11 2022-02-22 オムロン株式会社 Recommendation information specification device, recommendation information specification system, recommendation information specification method, terminal device, and program
JP2022058960A (en) * 2021-12-09 2022-04-12 オムロン株式会社 Recommendation information specification apparatus, recommendation information specification system, recommendation information specification method, and program
JP2022062220A (en) * 2021-12-09 2022-04-19 オムロン株式会社 Recommendation information specification apparatus, recommendation information specification system, recommendation information specification method, and program

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6809148B2 (en) * 2016-11-08 2021-01-06 富士ゼロックス株式会社 Program and combination extraction system

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001184429A (en) * 1999-10-12 2001-07-06 Yuji Saito Method, device, and recording medium for evaluating advertisement effect
JP2002312666A (en) * 2001-04-10 2002-10-25 Ntt Data Corp Advertisement providing method, mail relay server, managing server, and relaying server
JP2004054439A (en) * 2002-07-17 2004-02-19 Nippon Telegr & Teleph Corp <Ntt> Method, device and program for supporting formulation of marketing strategy based on analysis of customer behavior
JP2004240937A (en) * 2003-02-07 2004-08-26 Overture Services Inc Automatic flight management method for online market
JP2008027358A (en) * 2006-07-25 2008-02-07 Silver Egg Technology Kk Recommend device, recommend method and program
JP2009510551A (en) * 2005-09-14 2009-03-12 ジャンプ・タップ，インコーポレイテッド Provision of content to mobile communication facilities
JP2010512604A (en) * 2006-12-12 2010-04-22 ヤフー！ インコーポレイテッド Bid optimization in search engine marketing

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001184429A (en) * 1999-10-12 2001-07-06 Yuji Saito Method, device, and recording medium for evaluating advertisement effect
JP2002312666A (en) * 2001-04-10 2002-10-25 Ntt Data Corp Advertisement providing method, mail relay server, managing server, and relaying server
JP2004054439A (en) * 2002-07-17 2004-02-19 Nippon Telegr & Teleph Corp <Ntt> Method, device and program for supporting formulation of marketing strategy based on analysis of customer behavior
JP2004240937A (en) * 2003-02-07 2004-08-26 Overture Services Inc Automatic flight management method for online market
JP2009510551A (en) * 2005-09-14 2009-03-12 ジャンプ・タップ，インコーポレイテッド Provision of content to mobile communication facilities
JP2008027358A (en) * 2006-07-25 2008-02-07 Silver Egg Technology Kk Recommend device, recommend method and program
JP2010512604A (en) * 2006-12-12 2010-04-22 ヤフー！ インコーポレイテッド Bid optimization in search engine marketing

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11379417B2 (en) 2014-09-05 2022-07-05 Sony Corporation Information processing device, information processing method, and program
JPWO2016035424A1 (en) * 2014-09-05 2017-07-13 ソニー株式会社 Information processing apparatus, information processing method, and program
WO2016035424A1 (en) * 2014-09-05 2016-03-10 ソニー株式会社 Information processing device, information processing method and program
US10496606B2 (en) 2014-09-05 2019-12-03 Sony Corporation Information processing device, information processing method, and program
JP6286076B1 (en) * 2017-01-18 2018-02-28 株式会社リクルートホールディングス Information management apparatus and program
JP2018116492A (en) * 2017-01-18 2018-07-26 株式会社リクルートホールディングス Information management device and program
JP2019144951A (en) * 2018-02-22 2019-08-29 オムロン株式会社 Recommendation information specifying apparatus, recommendation information specifying system, recommendation information specifying method, and program
CN111143678A (en) * 2019-12-27 2020-05-12 第四范式（北京）技术有限公司 Recommendation system and recommendation method
CN111143678B (en) * 2019-12-27 2023-10-17 第四范式（北京）技术有限公司 Recommendation system and recommendation method
JP2021093216A (en) * 2021-03-11 2021-06-17 オムロン株式会社 Recommendation information specification device, recommendation information specification system, recommendation information specification method, terminal device, and program
JP2022031901A (en) * 2021-03-11 2022-02-22 オムロン株式会社 Recommendation information specification device, recommendation information specification system, recommendation information specification method, terminal device, and program
JP7052897B2 (en) 2021-03-11 2022-04-12 オムロン株式会社 Recommended information identification device, recommended information identification system, recommended information identification method, terminal device, and program
JP7127728B2 (en) 2021-03-11 2022-08-30 オムロン株式会社 Recommended information identifying device, recommended information identifying system, recommended information identifying method, terminal device, and program
JP2022058960A (en) * 2021-12-09 2022-04-12 オムロン株式会社 Recommendation information specification apparatus, recommendation information specification system, recommendation information specification method, and program
JP2022062220A (en) * 2021-12-09 2022-04-19 オムロン株式会社 Recommendation information specification apparatus, recommendation information specification system, recommendation information specification method, and program
JP7115655B2 (en) 2021-12-09 2022-08-09 オムロン株式会社 Recommended information identifying device, recommended information identifying system, recommended information identifying method, and program
JP7127749B2 (en) 2021-12-09 2022-08-30 オムロン株式会社 Recommended information identifying device, recommended information identifying system, recommended information identifying method, and program
JP7355181B2 (en) 2021-12-09 2023-10-03 オムロン株式会社 Recommended information identification device, recommended information identification system, recommended information identification method, and program

Also Published As

Publication number Publication date
JPWO2012049987A1 (en) 2014-02-24

Similar Documents

Publication Publication Date Title
WO2012049987A1 (en) Product recommendation system and product recommendation method
US11205218B2 (en) Client user interface activity affinity scoring and tracking
US9811851B2 (en) Automatic product groupings for merchandising
US10289736B2 (en) Systems and methods for ranking results based on dwell time
US11430037B2 (en) Real time item listing modification
US20190362409A1 (en) Real-time recommendation monitoring dashboard
US20140156346A1 (en) Methods and systems for dynamic demand sensing
US20230252546A1 (en) Presenting Search Results in a Dynamically Formatted Graphical User Interface
US20160005096A1 (en) Product recommendation engine based on remaining balance in a stored-value scenario
JP5255728B2 (en) ADVERTISING SYSTEM, ADVERTISING SYSTEM CONTROL METHOD, ADVERTISEMENT CONTROL DEVICE, PROGRAM, AND INFORMATION STORAGE MEDIUM
JP6761068B2 (en) Decision device, decision method and decision program
JP6810184B2 (en) Calculation device, calculation method and calculation program
US20140052502A1 (en) Balanced web analytics scorecard
JP6924611B2 (en) Information processing equipment, information processing methods and programs
US20210133712A1 (en) System and method for monitoring and updating content for an e-commerce platform
JP4993097B2 (en) Company policy decision support device, company policy decision support method and program
JP6748029B2 (en) Calculation device, calculation method, and calculation program
JP2020154879A (en) Determination device, determination method, and determination program
JP7292362B2 (en) Decision device, decision method and decision program
JP7122286B2 (en) Decision device, decision method and decision program
JP4879796B2 (en) RF analysis / RF prediction program
JP2018156335A (en) Content providing device, content providing method, and content providing program
JP2018077636A (en) Program and combination extraction system
JP7114888B2 (en) Recommendation degree determination device, display system, control method and program
JP6034451B1 (en) Information processing apparatus, information processing method, and program

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: 11832424

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2012538633

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11832424

Country of ref document: EP

Kind code of ref document: A1