WO2019016891A1 - レコメンド装置 - Google Patents
レコメンド装置 Download PDFInfo
- Publication number
- WO2019016891A1 WO2019016891A1 PCT/JP2017/026107 JP2017026107W WO2019016891A1 WO 2019016891 A1 WO2019016891 A1 WO 2019016891A1 JP 2017026107 W JP2017026107 W JP 2017026107W WO 2019016891 A1 WO2019016891 A1 WO 2019016891A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- product
- owned
- information
- user
- recommended
- Prior art date
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
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
-
- 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
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
-
- 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
-
- 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
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Definitions
- the present invention relates to a recommendation device for selecting an advertisement to be displayed according to a situation in order to display an advertisement of a product that a user really needs or a product that is potentially required as a recommendation (recommended) product. .
- an advertisement of a product to be displayed from the purchase information (including history and schedule) of the user and the characteristics of the product It is common to derive the timing. In addition to this, it is also well known to determine an advertisement of a product to be displayed based on the purchase tendency of other users having similar purchase information.
- the product is only sought for similarity in a wide range of each category.
- the advertisement of the smartphone currently sold is displayed to the customer who purchased the smartphone three years ago.
- the similarity is only found in a narrow range such as a trade name or a model number.
- the same detergent advertisement is displayed to the customer who purchased the detergent one month ago, and one customer purchased the product B together with the product A, so the advertisement of B to the customer who bought another A indicate.
- This invention was made in order to solve this problem, and an object of this invention is to provide the recommendation apparatus which can improve the extraction precision of a recommendation product.
- the recommendation device presents to the target user based on browse information indicating browse information on the user's product and unpurchased product information indicating information on the product on which the user intends to purchase. It includes a product selection unit that extracts recommended product candidates, and a specification correlation calculation unit that extracts products with product specifications that are correlated with the product specifications of recommended product candidates and extracts this product as a recommended product. .
- the recommendation device extracts a product having a product specification that is correlated with a product specification of a recommendation product candidate, and extracts this product as a recommendation product. Thereby, the extraction accuracy of the recommended product can be improved.
- FIG. 1 is a block diagram of a recommendation delivery system including a recommendation device according to the present embodiment.
- the illustrated recommendation delivery system includes a recommendation delivery apparatus 1, a network 2, and user terminals 3-1 to 3-n.
- the recommendation distribution device 1 is a device that distributes advertisement information of a product requested by the user to the user terminals 3-1 to 3-n via the network 2.
- one or more users use the user terminals 3-1 to 3-n having input / output functions such as their own PCs, smartphones, TVs, etc., and recommend distribution devices through the network 2 such as the Internet.
- Information on recommended products tailored to each user such as product advertisement and news is received from 1).
- the information on the recommended product may be displayed together on a part of a screen such as a browser or an application displayed by the user, or may be delivered in the form of an email, a notification, or the like.
- product refers to items that can be purchased and used by the user, such as electrical appliances, clothes, food, furniture, travel tours, and concert tickets.
- the recommendation distribution device 1 includes a recommendation information distribution server 10, a recommendation device 11, a browse information storage unit 12, an unpurchased product information storage unit 13, a owned product information storage unit 14, and a product information storage unit 15.
- the recommendation information distribution server 10 is a server for distributing the recommendation information extracted by the recommendation device 11 to the user terminals 3-1 to 3-n.
- the recommendation device 11 includes a product selection unit 110 and a specification correlation calculation unit 111, and is an apparatus for extracting a recommended product.
- the product selection unit 110 is a processing unit that extracts candidates of recommended products for the target user based on the browse information of the browse information storage unit 12 and the unpurchased product information of the unpurchased product information storage unit 13.
- the specification correlation calculation unit 111 refers to the product specification information 150 of the product information storage unit 15 to extract a product of a product specification having a correlation with the product specification of the recommended product candidate extracted by the product selection unit 110, and the product Is a processing unit that outputs as a recommended product.
- the browse information storage unit 12 is a storage unit that stores browse information which is information on the type and content of a document such as a web page, an application screen, and a PDF that the user has displayed so far.
- the unpurchased product information storage unit 13 is a storage unit that stores the unpurchased product information, which is information on a product that indicates the intention to purchase such that the user has been paying attention or desires to purchase so far. is there. Unpurchased product information is different from browse information, and the user explicitly expresses information such as the non-ownership or purchase intention of the product and the degree of it, and is recommended as the simplest example of acquiring unpurchased product information.
- the owned product information storage unit 14 is a storage unit for storing owned product information which is information on a product purchased by the user through online shopping or the like and a product owned by the user.
- the browse information, the unpurchased product information and the owned product information may include information acquired from the outside of the recommendation distribution device 1 in addition to the information registered through the recommendation distribution device 1. As an example acquired from the outside, browsing information may be collected from external services such as SNS (social network service), browser, search engine, etc.
- the product information storage unit 15 is a storage unit that stores product information classified by category, and has product specification information 150 as one of the product information.
- the product specification information 150 may be any information as long as it indicates information unique to the product, such as the product size, spec, color, weight, and function.
- the recommendation device 11 includes a processor 101, a memory 102, an input / output interface 103, a storage 104, and a bus 105.
- the processor 101 is a processor for realizing the product selection unit 110 and the specification correlation calculation unit 111 by executing programs corresponding to the functions of the product selection unit 110 and the specification correlation calculation unit 111.
- the memory 102 is a storage unit such as a program memory for storing various programs, a work memory used when the processor 101 performs data processing, and a ROM and a RAM used as a memory for expanding signal data.
- the input / output interface 103 is an interface for exchanging various signals with the browse information storage unit 12 to the product information storage unit 15 and the recommendation information distribution server 10, for example.
- the storage 104 is a storage unit for storing programs corresponding to the respective functions of the product selection unit 110 and the specification correlation calculation unit 111 and for storing various data.
- the bus 105 is a communication path for connecting the processor 101 to the storage 104 to one another. Note that at least one of the product selection unit 110 and the specification correlation calculation unit 111 may be configured by dedicated hardware.
- the product selection unit 110 indicates the user's tendency indicated by the behavior of the user based on the browsing information in the browsing information storage unit 12 and the user indicated by the non-purchased product information in the non-purchased product information storage unit 13 Using the product information and information such as the purchase results based on the owned product information in the owned product information storage unit 14, the product that the target user is potentially interested in is extracted as a candidate for a recommended product . Moreover, as this extraction process, you may analyze together the target user and another user. When the product selection unit 110 extracts a recommended product candidate, the product selection unit 110 outputs a message indicating that these products have not been purchased to the specification correlation calculation unit 111.
- the specification correlation calculation unit 111 uses the product specification information 150 in the product information storage unit 15 to determine a recommended product.
- FIG. 3 is a flowchart showing the operation of the specification correlation calculation unit 111.
- the specification correlation calculation unit 111 receives recommended recommendation products from the product selection unit 110, the specification correlation calculation unit 111 divides the product categories into product categories in terms of price, size, and function (step ST101). At this time, the product information in the product information storage unit 15 may be referred to for categorization.
- the specification correlation calculation unit 111 calculates, for each category (step ST102), the degree of correlation of product specifications between the candidate products for the category-divided candidate products (step ST103), and the degree of correlation is set. It is determined whether the threshold has been exceeded (step ST104).
- step ST104-YES the category in which the strong correlation is recognized is considered to indicate the preference of the user
- the category (condition) is added to the candidate selection condition (step ST105).
- the specification correlation calculation unit 111 determines whether all the categories have been calculated (step ST106). If not (step ST106-NO), the process returns to step ST102 to repeat the processing of steps ST102 to ST105.
- step ST104-NO the degree of correlation is equal to or less than the threshold in step ST104
- the specification correlation calculation unit 111 narrows down candidate products based on the candidate selection conditions so as to meet the conditions based on the candidate selection conditions at the stage where all the candidate selection conditions are present (Step ST106-YES), or otherwise meets the conditions.
- a product is also added as a candidate product from the product information and the like of the product information storage unit 15, and a recommended product is extracted (step ST107).
- the specification correlation calculation unit 111 extracts recommended products based on the candidate selection conditions here, the candidate selection conditions are output to the product selection unit 110, and the product selection unit 110 selects recommended products that meet the conditions. You may Further, as a condition to be added to the recommended product, it may be possible to refer to sales of each product, evaluation from other users, and the like.
- the specification correlation calculation unit 111 can process from the specification of the product selected by the product selection unit 110 independently of the part for narrowing down or adding a product with high relationship, so the existing A wide range of applications can be expected as an extension of the recommendation system.
- the information itself such as the product specification information 150 is necessary as an effect of the extraction of the recommended product itself, but the addition of the information allows the user's preference more finely than ever.
- the browse information indicating the browse information on the product of the user and the unpurchased product information indicating the information on the product on which the user indicates the intention to purchase Based on the product selection unit, which extracts the recommended product candidate to be presented to the target user, and the product specification having a correlation with the product specification of the recommended product candidate, the specification correlation which extracts this product as a recommended product Since the calculation unit is provided, the extraction accuracy of the recommended product can be improved.
- the second embodiment relates to a configuration for estimating a user's next purchase trend from owned product information which is history information of a product owned by the user. Since the configuration on the drawing is the same as that of the first embodiment shown in FIG. 1, it will be described using FIG.
- the product selection unit 110 in Embodiment 2 has a function of acquiring information on the owned products of the target user based on the owned product information of the owned product information storage unit 14. doing.
- the specification correlation calculation unit 111 maintains the product specification of the owned product when extracting a recommended product to be a replacement target of the owned product acquired by the product selection unit 110.
- the product information storage unit 15 stores product group information indicating a combination of products determined based on a predetermined condition.
- the other configuration is the same as that of the first embodiment, and thus the description thereof will be omitted.
- the product selection unit 110 owns the information of the product currently owned by the user from the owned product information of the owned product information storage unit 14 indicating the user's past purchase history and owned products, as in the conventional case.
- the specification correlation calculation unit 111 is notified along with a message indicating that it is a product.
- the specification correlation calculation unit 111 performs processing on the following two patterns: replacement and purchase.
- FIG. 4 shows an operation flowchart of the specification correlation calculation unit 111 in the examination of replacement of the owned product.
- the specification correlation calculation unit 111 selects a candidate product to be replaced in the following process.
- step ST202 next, from the product type, whether the product should maintain or improve the specification of the product at the time of replacement
- the product information of 15 is searched (steps ST203 and ST204). For example, in general, furniture and home appliances should maintain the "size" in the specifications, and other specifications such as performance are required to be improved upon replacement.
- the specification is a specification to be maintained or improved (YES in step ST204)
- the specification is added as a maintenance or improvement condition (step ST205).
- the specification condition is not added to the candidate selection condition. Note that the addition of specifications to be maintained is mandatory, and the addition of specifications to be improved may be optional.
- the specification correlation calculation unit 111 determines whether all the specifications have been processed (step ST206). If there is still a specification to be determined (step ST206-NO), the process returns to step ST203. The processing of steps ST203 to ST205 is repeated.
- step ST206-YES When all the specifications are determined in step ST206 (step ST206-YES), the specification correlation calculation unit 111 extracts a product under the candidate selection condition (step ST207). That is, a product whose replacement is recommended and which should be maintained or improved is selected from the product information. After that, it is determined whether all the products have been processed (step ST208). If there is an unprocessed product remaining (step ST208-NO), the process returns to step ST201 and the above process is repeated to process all products. If it is determined (step ST208-YES), the product replacement determination process is ended.
- the product has a defect based on the information that the user refers to the information such as the failure response of the product in the browse information storage unit 12 as the browse information of the user to determine whether replacement is desirable. You may use the guess that it may have occurred. In addition, information that another user bought the same product for another product may be used. Furthermore, the selection of products that satisfy the conditions may be performed by passing the candidate selection conditions to the product selection unit 110 and the product selection unit 110.
- FIG. 5 shows an operation flowchart of the specification correlation calculation unit 111 in consideration of making up for the lack of owned products. Whether the user is satisfied or not, the specification correlation calculation unit 111 selects a product lacking in the user's owned product as a candidate product for buying and adding. In this case, it is assumed that information of product groups to be combined in advance is registered as the product information of the product information storage unit 15. For example, in the case of home appliances, it is assumed that a refrigerator, a washing machine, a vacuum cleaner, a television and the like are registered as a group. The specification correlation calculation unit 111 first compares the product owned by the user with the product group using such product information (step ST301).
- step ST302-YES when it is determined that there is a missing product, that is, a product that the user has not purchased or does not own (step ST302-YES), among the missing products, a product that the user has not indicated as unnecessary
- the condition of a product which is estimated to be appropriate from the specifications of the user's owned products included in the group is calculated (step ST303). For example, when the user purchases a refrigerator, a washing machine and a vacuum cleaner made by ⁇ ⁇ , a television made by ⁇ ⁇ is required as a condition.
- the specification correlation calculation unit 111 extracts a product meeting the condition from the product information stored in the product information storage unit 15 based on the derived condition, as in the replacement (step ST304).
- step ST305 it is determined whether all the owned products have been compared. If all the products have been determined (step ST305-YES), the process ends. On the other hand, if the owned product as the comparison target remains in step ST305 (step ST305-NO), the process returns to step ST301 and the above process is repeated. Moreover, in step ST302, when there is no shortage product, it ends as it is.
- the present embodiment is more accurate as an extension of the existing system by adding the process using the correlation degree of the product specification to the conventional extraction process of the recommended product for replacement or buy-back. It is possible to select the product that the user is seeking. Specifically, for example, a product for which a recommendation for a long-term purchase of a user, for which the conventional recommendation method is not good, is replaced or bought in consideration of the user's environment (residential space, family structure, economic situation, etc.) There will be products with specifications (size, specifications, prices, etc.) that suit the environment.
- the product selection unit acquires the information of the owned product of the target user based on the owned product information indicating the information of the product owned by the user.
- the specification correlation calculation unit extracts a recommended product to be substituted for the owned product of the target user, it extracts the product specification having a correlation with the product specification to be maintained among the product specifications of the owned product. Because of this, it is possible to improve the extraction accuracy of recommended products at the time of replacement.
- the specification correlation calculation unit extracts the product specification having a correlation with the product specification to be improved among the product specification of the owned product, so that the user It is possible to extract recommended products corresponding to certain products more accurately.
- the product selection unit acquires the information of the owned product of the target user based on the owned product information indicating the information of the product owned by the user, and the specification correlation
- the calculation unit extracts a recommended product for compensating for the lack of the product owned by the target user
- the calculation unit compares the product group indicating the combination of products with the owned product, and lacks the product not included in the product group Since the product specification of the lacking product is obtained from the product specification of the owned product and the product meeting the product specification is extracted, the extraction accuracy of the recommended product at the time of buying and adding can be improved.
- the product selection unit acquires the information of the purchased product of the target user based on the owned product information indicating the information of the product owned by the user.
- the specification correlation calculation unit extracts a product of a product specification that is correlated with the product specification to be maintained among the product specifications of the owned product.
- the product group indicating the combination of products is compared with the owned products to extract the missing products, and the missing products are specified from the product specifications of the owned products. Since the specification is determined and the product matching the product specification is extracted, the extraction accuracy of the recommended product at the time of replacement and addition can be improved.
- the present invention allows free combination of each embodiment, or modification of any component of each embodiment, or omission of any component in each embodiment. .
- the recommendation device relates to a configuration for extracting the recommendation device based on the browse information of the user, the unpurchased product information, and the product specification, and advertising of a product to the user in online shopping Suitable for doing.
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
これらの方法は、適した状況においては非常に効果的な広告を選定できるが、長い間隔で購入する家具や家電、また特定の意図や嗜好を持って商品を選んでいる顧客に対しては、必要としていない商品の広告を表示する可能性が高まるという課題があった。例えば高級腕時計を探している顧客に対して安価な腕時計の広告を表示したり、3年前にゲーム対応PCを購入した顧客にゲーム向きではないPCの広告を表示したりするなど、必ずしも対象とする顧客に適したレコメンド製品ではないという問題があった。
実施の形態1.
図1は、本実施の形態によるレコメンド装置を含むレコメンド配信システムの構成図である。
図示のレコメンド配信システムは、レコメンド配信装置1、ネットワーク2、ユーザ端末3-1~3-nを備える。レコメンド配信装置1は、ユーザが求めている製品の広告情報をネットワーク2を介してユーザ端末3-1~3-nに配信する装置である。この構成では、1人または複数のユーザが、各自の持つPCやスマートフォン、TV等の入出力機能を持つユーザ端末3-1~3-nを用いて、インターネット等のネットワーク2を通じて、レコメンド配信装置1から製品広告やニュース等の各ユーザに合わせたレコメンド製品の情報を受信するものである。この際、レコメンド製品の情報はユーザが表示させているブラウザやアプリなどの画面の一部に一緒に表示させても良いし、メールや通知などの形式で配信しても良い。またここでいう製品とは、電化製品や衣類、食品、家具、旅行ツアー、コンサートチケットなど、ユーザが購入し利用できるものを指す。
レコメンド装置11は、図2に示すように、プロセッサ101、メモリ102、入出力インタフェース103、ストレージ104、バス105を備えている。プロセッサ101は、製品選定部110及び仕様相関計算部111の機能に対応したプログラムを実行することにより、これら製品選定部110及び仕様相関計算部111を実現するためのプロセッサである。メモリ102は、各種プログラムを記憶するプログラムメモリ、プロセッサ101がデータ処理を行う際に使用するワークメモリ及び信号データを展開するメモリ等として使用するROM及びRAM等の記憶部である。入出力インタフェース103は、例えば、閲覧情報格納部12~製品情報格納部15やレコメンド情報配信サーバ10との各種の信号をやり取りするためのインタフェースである。また、ストレージ104は、製品選定部110及び仕様相関計算部111のそれぞれの機能に対応したプログラムを格納すると共に、各種データを蓄積するための記憶部である。バス105は、プロセッサ101~ストレージ104を相互に接続するための通信路である。
なお、製品選定部110及び仕様相関計算部111の少なくともいずれかを専用のハードウェアで構成しても良い。
先ず、製品選定部110は、閲覧情報格納部12の閲覧情報に基づいたユーザのこれまでの行動の傾向と、未購入製品情報格納部13の未購入製品情報で示されるユーザが興味を示した製品の情報と、所有製品情報格納部14の所有製品情報に基づいた購入実績等の情報とを用いて、対象とするユーザが潜在的に興味を持っている製品をレコメンド製品の候補として抽出する。また、この抽出処理としては、対象とするユーザと他ユーザとを合わせて分析しても良い。製品選定部110は、レコメンド製品の候補を抽出すると、これらの製品が未購入であるというメッセージと共に仕様相関計算部111に出力する。
図3は仕様相関計算部111の動作を示すフローチャートである。仕様相関計算部111は、製品選定部110からレコメンド製品の候補を受け取ると、それらを製品ジャンル毎に、価格や大きさ、機能といった観点でカテゴリ分けを行う(ステップST101)。このとき、製品情報格納部15の製品情報を参考にしてカテゴリ分けを行ってもよい。次に、仕様相関計算部111は、カテゴリ分けされた候補製品に対し、カテゴリ毎に(ステップST102)、候補製品間の製品仕様の相関度を計算し(ステップST103)、相関度が設定された閾値を超えたかを判定する(ステップST104)。これは、同じカテゴリに分類された複数の候補製品が、仕様面でどういった相関を持つかを計算し、より強い相関を持つカテゴリを見つけるための処理である。強い相関が認められたカテゴリ(ステップST104-YES)は、ユーザの嗜好を示すと考えられるため、当該カテゴリ(条件)を候補選定条件に加える(ステップST105)。そして、仕様相関計算部111は、全てのカテゴリを計算したかを判定し(ステップST106)、そうでない場合(ステップST106-NO)はステップST102に戻り、ステップST102~ST105の処理を繰り返す。一方、ステップST104において、相関度が閾値以下であった場合(ステップST104-NO)は、そのままステップST106に移行する。
なお、ここでは仕様相関計算部111が候補選定条件を元にレコメンド製品の抽出を行ったが、候補選定条件を製品選定部110に出力し、製品選定部110が条件に合致するレコメンド製品を選定してもよい。またレコメンド製品に加える条件として、各製品の売れ筋、他ユーザからの評価などを参考にしてもよい。
また、レコメンド製品の抽出自体の効果としては、これまでのレコメンド方式とは異なり、追加で製品仕様情報150といった情報が必要であるが、その情報が加わることでこれまで以上に細かく利用者の嗜好を捉えることが可能となる。例えば、ユーザがある機能を不要と感じており、その機能が付加されていない製品を探している場合、これまでは「○○機能なし」というキーワードで製品を検索することが難しかったため、一つ一つの製品の仕様を確認するほかなかったが、本レコメンド装置を用いることで、○○機能がない製品を重点的にチェックしていることがユーザの行動や意思表示から推測できるため、ユーザは、○○機能が搭載されていない製品のレコメンドを受け、求めている製品を効率よく発見できるようになる。
実施の形態2は、ユーザが所有している製品の履歴情報である所有製品情報から、ユーザの次の購買動向を推定する構成に関するものである。図面上の構成は図1に示した実施の形態1と同様であるため、図1を用いて説明する。
先ず、製品選定部110は、従来と同様に、ユーザのこれまでの購買履歴や所有製品を示す所有製品情報格納部14の所有製品情報から、ユーザが現在所有している製品の情報を、所有製品であるというメッセージと共に、仕様相関計算部111へ通知する。仕様相関計算部111は所有製品の情報を受け取ると、以下に示す買い替えと買い足しの2パターンについて処理を行う。
図4に所有製品の置き換え検討における仕様相関計算部111の動作フローチャートを示す。
所有製品が古くなったあるいは故障したなどの理由で、新しい製品の購入を検討するユーザを想定する場合、仕様相関計算部111は、以下の処理で置き換え候補の製品を選定する。まず、全ての所有製品について、購入時期が判明しているものについては購入時期と買い替え間隔から、判明していないものについては製品の販売期間と買い替え間隔から、当該製品の置き換えが望ましいかを判定する(ステップST201)。これを所有製品毎に行う(ステップST202)。判定対象の所有製品で置き換えが望ましくなかった場合(ステップST202-NO)は、ステップST201に戻って、次の所有製品の置き換え判定を行う。
図5に所有製品の不足を補うことへの検討における仕様相関計算部111の動作フローチャートを示す。
ユーザが満足している、いないにかかわらず、ユーザの所有製品で不足している製品を、仕様相関計算部111は、買い足し候補の製品として選定する。この場合、製品情報格納部15の製品情報として、事前に組み合わさる製品グループの情報が登録されているものとする。例えば生活家電であれば、冷蔵庫、洗濯機、掃除機、テレビなどをグループとして登録されているものとする。このような製品情報を用いて仕様相関計算部111は、先ず、ユーザの所有製品と製品グループとを比較する(ステップST301)。比較の結果、不足製品、すなわちユーザが購入していない、あるいは所有していない製品があると判定した場合(ステップST302-YES)、その不足製品のうち、ユーザが不要と意思表示していない製品について、グループに含まれるユーザの所有製品の仕様から買い足しが適切と推測される製品の条件を計算する(ステップST303)。例えば、ユーザが△△電機製の冷蔵庫と洗濯機と掃除機を購入していた場合、△△電機製のテレビが条件として求められる。次に仕様相関計算部111は、置き換え時と同様に、導出された条件を元に、条件に合致する製品を製品情報格納部15の格納されている製品情報から抽出する(ステップST304)。そして、全ての所有製品を比較したかを判定し(ステップST305)、全てを判定済み(ステップST305-YES)であった場合は終了する。一方、ステップST305において、比較対象としての所有製品が残っていた場合(ステップST305-NO)はステップST301に戻って上記処理を繰り返す。また、ステップST302において、不足製品がなかった場合は、そのまま終了する。
Claims (5)
- ユーザの製品に関する閲覧情報を示す閲覧情報と、ユーザが購入への意思表示をしている製品の情報を示す未購入製品情報に基づき、対象となるユーザに提示するレコメンド製品の候補を抽出する製品選定部と、
前記レコメンド製品の候補の製品仕様と相関のある製品仕様の製品を抽出し、当該製品をレコメンド製品として抽出する仕様相関計算部とを備えたことを特徴とするレコメンド装置。 - 前記製品選定部は、ユーザが所有している製品の情報を示す所有製品情報に基づき、前記対象となるユーザの所有製品の情報を取得し、
前記仕様相関計算部は、前記対象となるユーザの所有製品の置き換え対象となるレコメンド製品を抽出する場合、前記所有製品の製品仕様のうち維持する製品仕様と相関のある製品仕様の製品を抽出することを特徴とする請求項1記載のレコメンド装置。 - 前記仕様相関計算部は、前記所有製品の製品仕様のうち向上させる製品仕様と相関のある製品仕様の製品を抽出することを特徴とする請求項2記載のレコメンド装置。
- 前記製品選定部は、ユーザが所有している製品の情報を示す所有製品情報に基づき、前記対象となるユーザの所有製品の情報を取得し、
前記仕様相関計算部は、前記対象となるユーザの所有製品の不足を補うためのレコメンド製品を抽出する場合、製品の組合せを示す製品グループと前記所有製品とを比較し、前記製品グループで含まれていない製品を不足製品として抽出すると共に、前記所有製品の製品仕様から前記不足製品の製品仕様を求め、当該製品仕様に一致する製品を抽出することを特徴とする請求項1記載のレコメンド装置。 - 前記製品選定部は、ユーザが所有している製品の情報を示す所有製品情報に基づき、前記対象となるユーザの所有製品の情報を取得し、
前記仕様相関計算部は、前記対象となるユーザの所有製品の置き換え対象となるレコメンド製品を抽出する場合、前記所有製品の製品仕様のうち維持する製品仕様と相関のある製品仕様の製品を抽出し、前記所有製品の不足を補うためのレコメンド製品を抽出する場合、製品の組合せを示す製品グループと前記所有製品とを比較して不足製品を抽出すると共に、前記所有製品の製品仕様から前記不足製品の製品仕様を求め、当該製品仕様に一致する製品を抽出することを特徴とする請求項1記載のレコメンド装置。
Priority Applications (8)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/621,013 US11538085B2 (en) | 2017-07-19 | 2017-07-19 | Recommendation device |
CN201780093108.1A CN110892438A (zh) | 2017-07-19 | 2017-07-19 | 推荐装置 |
PCT/JP2017/026107 WO2019016891A1 (ja) | 2017-07-19 | 2017-07-19 | レコメンド装置 |
JP2019530286A JP6611998B2 (ja) | 2017-07-19 | 2017-07-19 | レコメンド装置 |
DE112017007652.8T DE112017007652T5 (de) | 2017-07-19 | 2017-07-19 | Empfehlungseinrichtung |
KR1020207000661A KR102170865B1 (ko) | 2017-07-19 | 2017-07-19 | 레커맨드 장치 |
TW106131549A TWI660316B (zh) | 2017-07-19 | 2017-09-14 | Recommended device |
US18/058,119 US20230153888A1 (en) | 2017-07-19 | 2022-11-22 | Recommendation device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2017/026107 WO2019016891A1 (ja) | 2017-07-19 | 2017-07-19 | レコメンド装置 |
Related Child Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US16/621,013 A-371-Of-International US11538085B2 (en) | 2017-07-19 | 2017-07-19 | Recommendation device |
US18/058,119 Continuation US20230153888A1 (en) | 2017-07-19 | 2022-11-22 | Recommendation device |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019016891A1 true WO2019016891A1 (ja) | 2019-01-24 |
Family
ID=65015410
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2017/026107 WO2019016891A1 (ja) | 2017-07-19 | 2017-07-19 | レコメンド装置 |
Country Status (7)
Country | Link |
---|---|
US (2) | US11538085B2 (ja) |
JP (1) | JP6611998B2 (ja) |
KR (1) | KR102170865B1 (ja) |
CN (1) | CN110892438A (ja) |
DE (1) | DE112017007652T5 (ja) |
TW (1) | TWI660316B (ja) |
WO (1) | WO2019016891A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7434469B1 (ja) | 2022-09-02 | 2024-02-20 | 株式会社ジェーシービー | プログラム、情報処理装置、及び情報処理方法 |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11538085B2 (en) * | 2017-07-19 | 2022-12-27 | Trygle Co., Ltd. | Recommendation device |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002123547A (ja) * | 2000-10-12 | 2002-04-26 | Canon Inc | 商品選択支援システム、商品選択支援サーバー、商品選択支援装置および商品選択支援方法 |
JP2004220152A (ja) * | 2003-01-10 | 2004-08-05 | Sharp Corp | 情報推薦装置、情報推薦先の抽出装置、コンピュータプログラム、およびコンピュータ読取可能な記録媒体 |
JP2004348613A (ja) * | 2003-05-26 | 2004-12-09 | Canon Inc | 製品およびサービス推薦装置 |
JP2008276449A (ja) * | 2007-04-27 | 2008-11-13 | Albert:Kk | 希望商品に類似する商品の推薦システム |
WO2013125128A1 (ja) * | 2012-02-22 | 2013-08-29 | 株式会社ニコン | 電子機器 |
JP2015191449A (ja) * | 2014-03-28 | 2015-11-02 | パナソニック インテレクチュアル プロパティ コーポレーション オブアメリカPanasonic Intellectual Property Corporation of America | 情報提供方法、情報提供装置、情報通信機器、情報通信機器の表示方法、並びにプログラム |
Family Cites Families (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4330478B2 (ja) | 2004-03-31 | 2009-09-16 | 富士通株式会社 | 商品情報提供方法および商品情報提供装置 |
JP2006176231A (ja) | 2004-12-21 | 2006-07-06 | Atsunori Kubokawa | サーバシステム、その制御方法及び制御プログラム、情報処理システム及び方法 |
US8214264B2 (en) * | 2005-05-02 | 2012-07-03 | Cbs Interactive, Inc. | System and method for an electronic product advisor |
US20070124265A1 (en) * | 2005-11-29 | 2007-05-31 | Honeywell International Inc. | Complex system diagnostics from electronic manuals |
US8126817B2 (en) * | 2008-06-06 | 2012-02-28 | International Business Machines Corporation | Product repair assistance using a virtual universe |
US9189811B1 (en) | 2010-01-07 | 2015-11-17 | Amazon Technologies, Inc. | Electronic marketplace recommendations |
US9092810B2 (en) * | 2011-04-11 | 2015-07-28 | Ebay Inc. | Methods and systems for merchandising products in bundles in an online marketplace |
US8484099B1 (en) * | 2012-02-06 | 2013-07-09 | Amazon Technologies, Inc. | Method, medium, and system for behavior-based recommendations of product upgrades |
US20150339759A1 (en) | 2012-02-06 | 2015-11-26 | Amazon Technologies, Inc. | Detecting product attributes associated with product upgrades based on behaviors of users |
CN102609523B (zh) | 2012-02-10 | 2018-03-06 | 上海视畅信息科技有限公司 | 基于物品分类和用户分类的协同过滤推荐方法 |
JP2014029643A (ja) | 2012-07-31 | 2014-02-13 | Sharp Corp | 情報提供装置、情報提供方法、情報提供システム、およびプログラム |
US20140156423A1 (en) * | 2012-11-30 | 2014-06-05 | Wal-Mart Stores, Inc. | Suggesting Expired Product Replenishment Based On Receipt Data |
KR101663580B1 (ko) | 2013-03-08 | 2016-10-07 | 심플렉스 인터넷 주식회사 | 전자상거래 추천 분석 시스템 및 추천 분석 방법 |
JP5391352B1 (ja) | 2013-04-24 | 2014-01-15 | 株式会社タカマツヤ | システム及び顧客管理サーバ |
US20150120386A1 (en) * | 2013-10-28 | 2015-04-30 | Corinne Elizabeth Sherman | System and method for identifying purchase intent |
US10318543B1 (en) * | 2014-03-20 | 2019-06-11 | Google Llc | Obtaining and enhancing metadata for content items |
US10366438B2 (en) * | 2014-03-26 | 2019-07-30 | Cellco Partnership | Product notification and recommendation technology utilizing detected activity |
TW201541381A (zh) * | 2014-04-29 | 2015-11-01 | Taiwan Mobile Co Ltd | 商品推薦系統 |
WO2015166508A1 (en) * | 2014-04-30 | 2015-11-05 | Hewlett-Packard Development Company, L.P. | Correlation based instruments discovery |
US10748200B2 (en) | 2015-03-03 | 2020-08-18 | Amazon Technologies, Inc. | Search query-based replacement part interface |
CN106709741A (zh) | 2015-07-22 | 2017-05-24 | 中兴通讯股份有限公司 | 一种商品推荐方法及系统 |
CN106708888B (zh) * | 2015-11-17 | 2020-06-26 | 阿里巴巴集团控股有限公司 | 推荐方法及装置 |
US11093906B2 (en) * | 2016-03-09 | 2021-08-17 | Autozone Parts, Inc. | Systems and apparatuses facilitating a do-it-yourself experience-based repair solution |
JP6405343B2 (ja) * | 2016-07-20 | 2018-10-17 | Necパーソナルコンピュータ株式会社 | 情報処理装置、情報処理方法、およびプログラム |
WO2018042982A1 (ja) * | 2016-08-31 | 2018-03-08 | 日立工機株式会社 | 故障診断システム及び管理システム |
JP6762819B2 (ja) * | 2016-09-14 | 2020-09-30 | 株式会社東芝 | 入力支援装置およびプログラム |
US11538085B2 (en) * | 2017-07-19 | 2022-12-27 | Trygle Co., Ltd. | Recommendation device |
-
2017
- 2017-07-19 US US16/621,013 patent/US11538085B2/en active Active
- 2017-07-19 CN CN201780093108.1A patent/CN110892438A/zh active Pending
- 2017-07-19 JP JP2019530286A patent/JP6611998B2/ja active Active
- 2017-07-19 KR KR1020207000661A patent/KR102170865B1/ko active IP Right Grant
- 2017-07-19 WO PCT/JP2017/026107 patent/WO2019016891A1/ja active Application Filing
- 2017-07-19 DE DE112017007652.8T patent/DE112017007652T5/de not_active Ceased
- 2017-09-14 TW TW106131549A patent/TWI660316B/zh active
-
2022
- 2022-11-22 US US18/058,119 patent/US20230153888A1/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002123547A (ja) * | 2000-10-12 | 2002-04-26 | Canon Inc | 商品選択支援システム、商品選択支援サーバー、商品選択支援装置および商品選択支援方法 |
JP2004220152A (ja) * | 2003-01-10 | 2004-08-05 | Sharp Corp | 情報推薦装置、情報推薦先の抽出装置、コンピュータプログラム、およびコンピュータ読取可能な記録媒体 |
JP2004348613A (ja) * | 2003-05-26 | 2004-12-09 | Canon Inc | 製品およびサービス推薦装置 |
JP2008276449A (ja) * | 2007-04-27 | 2008-11-13 | Albert:Kk | 希望商品に類似する商品の推薦システム |
WO2013125128A1 (ja) * | 2012-02-22 | 2013-08-29 | 株式会社ニコン | 電子機器 |
JP2015191449A (ja) * | 2014-03-28 | 2015-11-02 | パナソニック インテレクチュアル プロパティ コーポレーション オブアメリカPanasonic Intellectual Property Corporation of America | 情報提供方法、情報提供装置、情報通信機器、情報通信機器の表示方法、並びにプログラム |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7434469B1 (ja) | 2022-09-02 | 2024-02-20 | 株式会社ジェーシービー | プログラム、情報処理装置、及び情報処理方法 |
Also Published As
Publication number | Publication date |
---|---|
US11538085B2 (en) | 2022-12-27 |
US20200175565A1 (en) | 2020-06-04 |
CN110892438A (zh) | 2020-03-17 |
JP6611998B2 (ja) | 2019-11-27 |
TWI660316B (zh) | 2019-05-21 |
DE112017007652T5 (de) | 2020-03-05 |
JPWO2019016891A1 (ja) | 2019-12-19 |
KR20200008659A (ko) | 2020-01-28 |
TW201909076A (zh) | 2019-03-01 |
KR102170865B1 (ko) | 2020-10-28 |
US20230153888A1 (en) | 2023-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4886749B2 (ja) | 推薦商品選択装置、推薦商品選択プログラムおよび商品検索装置 | |
KR101753481B1 (ko) | 오픈 마켓별 가격변동 자동 모니터링 및 변경 시스템 | |
US20160364759A1 (en) | Generating Advertisements Based On Proximate Computing Devices | |
US20140067530A1 (en) | Systems and methods for precision retailing | |
WO2013126648A1 (en) | Methods and apparatus for recommending products and services | |
JP2013537339A (ja) | 商品推薦の生成 | |
US20230153888A1 (en) | Recommendation device | |
CN105447045A (zh) | 信息排序方法、装置及信息提供方法、系统 | |
KR20170077479A (ko) | 웹 서버 및 마켓팅에 따른 리워드 제공 방법 | |
US20160225021A1 (en) | Method and system for advertisement retargeting based on predictive user intent patterns | |
JP5759084B1 (ja) | 情報処理装置、情報処理方法及びプログラム | |
JP6320258B2 (ja) | 抽出装置、抽出方法及び抽出プログラム | |
JP6674527B1 (ja) | 価額設定装置、価額設定方法、及び価額設定プログラム | |
CN114245165A (zh) | 直播推荐方法、装置、电子设备以及存储介质 | |
WO2017126707A1 (ja) | 商品購入支援システム | |
CN106687947B (zh) | 利用多连接图标的移动信息输出控制系统及其方法 | |
KR20140079752A (ko) | 쇼핑 정보 제공 방법 및 장치 | |
KR101655372B1 (ko) | 상품 정보 제공 장치 및 방법 | |
JP6355151B2 (ja) | オファーリングシステム、オファーリング方法及びプログラム | |
KR101561233B1 (ko) | 사용자 단말 타겟팅 시스템 및 방법 | |
KR101698528B1 (ko) | 온라인 거래 장치 및 온라인 거래 방법 | |
KR20160076374A (ko) | Crm 마케팅 정보를 제공하는 방법 및 장치 | |
JP6059597B2 (ja) | コンテンツ検索結果提供装置、コンテンツ検索結果提供方法、及びコンテンツ検索結果提供システム | |
JP2022060964A (ja) | 情報処理装置及び情報処理プログラム | |
Hackl et al. | Measuring Product Differentiation by Observing Consumer Search |
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: 17918177 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2019530286 Country of ref document: JP Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 20207000661 Country of ref document: KR Kind code of ref document: A |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 17918177 Country of ref document: EP Kind code of ref document: A1 |