JP3974407B2 - Recommended item introduction method, recommended item introduction server, recording medium recording recommended item introduction program, recommended item introduction program, and recommended item introduction service provision method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method and system for introducing recommended items to a client via a network such as the Internet.
[0002]
[Prior art]
The conventional recommended item introduction server that introduces recommended items is a client preference estimation factor information for each client identifier of a sufficient number of clients to analyze / estimate all or preference accessing the item provider from the item provider server. When the client identifier is received together with the recommended item introduction request from the item provider server, the client preference is analyzed / estimated based on the client preference estimation factor information list for each client identifier provided, It provided a service to send a list of items that the client would be interested in to an item provider server.
[0003]
FIG. 14 is a configuration diagram of a conventional recommended item introduction system, and FIG. 15 is a flowchart showing its operation.
[0004]
The recommended item introduction system includes an item provider server 62 that acquires preference estimation factor information of the client 63 and a recommended item introduction server 61 that introduces recommended items.
[0005]
Next, the operation of this conventional example will be described with reference to FIG.
[0006]
1. Acquisition of client preference estimation factor information list for each client identifier
(1) Transmission of client preference estimation factor information from the client to the item provider server 62 (step 71)
For example, the item provider server 62, which is a Web page search site, acquires a client access history (configured by a list of client identifiers, item identifiers, and access dates) as client preference estimation factor information.
[0007]
In order to identify the same client among a plurality of item provider servers 62 and provide a service, the client identifier needs to be a unique client identifier across each item provider server.
[0008]
(2) Transmission of client preference estimation factor information list for each client identifier from the item provider server 62 to the recommended item introduction server 61 (step 72)
The item provider server 62 aggregates client preference estimation factor information collected from each client and periodically sends it to the recommended item introduction server 61 as a client preference estimation factor information list for each client identifier. Here, the client preference estimation factor information list for each client identifier is, for example, a map of how many accesses (viewing / purchasing) have been performed for each client identifier and item identifier.
[0009]
(3) Processing of client preference estimation factor information list for each client identifier in the recommended item introduction server 61 (step 73)
The recommended item introduction server 61 processes the client preference estimation factor information list for each received client identifier so as to respond to the recommended item introduction request from the client. For example, the processing result includes a list in which the item identifiers accessed for each client identifier are arranged in order of the number of access times, and the client identifiers of other clients accessing the item identifier for each item identifier. This is a list arranged in the number of accesses.
[0010]
2. Get recommended item introduction request
(4) Send recommended item introduction request from client 63 to item provider server 62 (step 74)
The client 63 transmits a recommended item introduction request to the item provider server 62 together with the client identifier.
[0011]
(5) Sending a recommended item introduction request from the item provider server 62 to the recommended item introduction server 61 (step 75)
The item provider server 62 transmits a client identifier together with a recommended item introduction request to the recommended item introduction server 61.
[0012]
3. Introduction of recommended items
(6) Send recommended item list from recommended item introduction server 61 to item provider server 62 (step 76)
The recommended item introduction server 61 creates a list of introduced item identifiers based on the client identifier received in (4), and transmits it to the item provider server 62 together with the client identifier as a recommended item list. In the transmission of the recommended item list, a fee is paid from the item provider server 62 to the recommended item introduction server operating company by a fixed amount monthly charge or a pay-per-use charge for each information amount.
[0013]
(7) Send recommended item list from item provider server 62 to client 63 (step 77)
The item provider server 62 transmits the recommended item list received in (6) to the client uniquely identified by the client identifier.
[0014]
[Problems to be solved by the invention]
The above-described conventional system has client preference estimator information for each client identifier of a client (all or a sufficient number of clients to analyze / estimate preferences) that many item provider servers access to the item provider server. In many cases, it is considered that the list is valuable know-how, so it is often difficult to have the client preference estimation factor information list for each client identifier provided, and the introduction request is made based on the client identifier. When providing the recommended item introduction service with a plurality of item provider servers, there is a problem that a unique client identifier common to each server must be assigned to identify the client across each system.
[0015]
An object of the present invention is to obtain a client identifier that is common and unique across a plurality of item provider servers without acquiring a client preference estimation factor information list for each client identifier from each item provider server that wants to use the recommended item introduction service. Recommended item introduction method, recommended item introduction server, recording medium recording recommended item introduction program, recommended item introduction program, and recommended item introduction service provision method that can introduce recommended items without having to grant There is.
[0016]
[Means for Solving the Problems]
In order to solve the above problems, the present invention obtains (a) client preference estimation factor information for each client identifier, which is information used to estimate the preference of each of a plurality of clients, and (b) sends an item via a network. Receiving client preference estimation factor information indicating the preference of the client without specifying the client, together with the recommended item introduction request of the client, from the item provider server provided to a plurality of clients, and the client for each client identifier Provided is a recommended item introduction method characterized in that a list of recommended items is created based on preference estimation factor information and client preference estimation factor information, and a recommended item list is transmitted to the item provider server.
[0017]
In addition, the present invention provides a client preference estimation factor information processing unit that acquires client preference estimation factor information for each client identifier, which is information used to estimate each preference of a plurality of clients, and a plurality of items via a network. Client preference estimation factor information indicating the preference of a certain client is received from the item provider server provided to the client together with the recommended item introduction request of the certain client without specifying the certain client, and the client preference estimation factor for each client identifier A recommended item introduction means for creating a recommended item list based on the information and the client preference estimation factor information and transmitting the recommended item list to the item provider server. To provide a server.
[0018]
The present invention also provides a recording medium on which a recommended item introduction program for causing a computer to execute the recommended item introduction method is recorded.
[0019]
The present invention also provides a recommended item introduction program for causing a computer to execute the recommended item introduction method.
[0020]
Further, the present invention provides a recommended item introduction service from a recommended item introduction server attached to an item provider server to a plurality of item provider servers that provide items to a plurality of clients via a network. In the recommended item introduction server, (a) client preference estimation factor information for each client identifier, which is information on which each client's preference is estimated, is acquired, and (b) from an item provider server When the client preference estimation factor information indicating the preference of the certain client is received without specifying the certain client together with the recommended item introduction request of the certain client, the recommended item introduction server receives the client preference for each client identifier. Create estimated factor information and a list of recommended items based on the client preference estimating factor information, and transmits the recommended item list to the certain item provider server provides a recommended item introduction service providing method characterized by.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
First, the main features of the present invention will be briefly summarized.
[0022]
The present invention introduces a recommended item to a plurality of item providers and servers based on a client preference estimation factor information list for each client identifier provided from a third party company or the like (server preference estimation factor information list for each client identifier). By providing the service, the recommended item introduction service is provided to clients using the item provider server without acquiring the client preference estimation factor information list for each client identifier from each item provider server that wants to use the recommended item introduction service. To do. In addition, since the client identifier is not required as information necessary for the introduction, it is not necessary to assign a client identifier for uniquely identifying the client across multiple item providers and servers. The client preference estimation factor information, which is the client's privacy information, can be prevented from leaking in association with the client, and privacy can be protected.
[0023]
The present invention is particularly effective in an application service provider business that provides recommended item introduction services.
[0024]
The item provider server that is the target of the present invention is used not only for providing services on the Internet, but also for network services that can exchange information (items) in both directions, such as digital TV and information appliances. Also included.
[0025]
Here, terms used in this specification will be described.
[0026]
Item: Advertisement, information, CD / book / movie title and outline, song (music information), etc. Client: User using item provider server (or client preference estimation factor information list providing server for each client identifier). Or, depending on the method of assigning the client identifier, it indicates a client terminal used by the client.
[0027]
Item provider: A company that operates a server that provides items to multiple clients. Introducing / selling items such as CDs / books, EC sites that introduce / sell items, search / directory sites that introduce / search items such as Web contents and information, and integrating multiple EC sites A company that operates an electronic mall site. A server operated by the same company is called an item provider server.
[0028]
Client preference estimation factor information: information that is the basis for estimating client preference. In the present invention, a client's access history to an item, a list of interesting items input by the client, a list of evaluations of interesting items input by the client, and the like. For example, a list of “item identifier and access date” for each client, a list of “item identifier and evaluation value” for each client, a list of “item identifier” for each client, or the like.
[0029]
Client preference estimation factor information list for each client identifier: Client preference estimation factor information collected by a server that collects client preference estimation factor information and listed for each of a plurality of unique client identifiers in the server. Recommended item introduction: A server that implements a service that provides an item estimates client preference using client preference estimation factor information and introduces an item list that the client may be interested in.
[0030]
Next, embodiments of the present invention will be described with reference to the drawings.
[0031]
FIG. 1 is a configuration diagram of a recommended item introduction system according to an embodiment of the present invention. The recommended item introduction system shown in FIG. 1 includes a client preference estimation factor information list providing server 1, a recommended item introduction server 2, and an item provider server 3 for each client identifier.
[0032]
The client preference estimation factor information list providing server 1 for each client identifier collects and aggregates client preference estimation factor information from the client, transmits a client preference estimation factor information list for each client identifier to the recommended item introduction server 2, and responds to this transmission. Receive payment from recommended item introduction server 2.
[0033]
The recommended item introduction server 2 includes a client preference estimation factor information list processing unit 4 and a recommended item introduction service unit 5.
[0034]
The client preference estimation factor information list processing unit 4 receives and aggregates the client preference estimation factor information list for each client identifier from the client identifier client preference estimation factor information list providing server 1, generates a client preference estimation factor information map, and recommends items. A fee is paid for the transmission of the client preference estimation factor information list for each client identifier of the server identifier providing server 1 for each client identifier.
[0035]
When the recommended item introduction service unit 5 receives the client preference estimation factor information list from the client preference estimation factor information list processing unit 4 and receives the recommended item introduction request from the item provider server 3 together with the client preference estimation factor information, A list of introduction items is created based on the preference estimation factor information map and the client preference estimation factor information, and is transmitted to the item provider server 3 as a recommended item list. A fee is paid from the item provider server 3 to the recommended item introduction server 2 for transmission of the recommended item list.
[0036]
The item provider server 3 receives a recommended item introduction request from the client 6, sends it to the recommended item introduction service unit 5 of the recommended item introduction server 2, and receives a recommended item list from the recommended item introduction service unit 5. Is transmitted to the client 6. A fee is paid for transmission of the recommended item list from the recommended item introduction server 2.
[0037]
Next, the operation of this embodiment will be described.
[0038]
1. Acquisition of client preference estimation factor information list (FIGS. 2 and 3)
(1) Transmission of client preference estimation factor information to each client identifier client preference estimation factor information list providing server 1 (step 11)
The client preference estimation factor information list providing server 1 for each client identifier, in the case of a Web page search site or the like, accesses the client 6 access history (item identifier and access for each client identifier (viewed / purchased item of the item provider server 3). ) (Date list) (FIG. 3 (1)), and in the case of an EC site such as a book or CD sale, the client numerically inputs the degree of interest of the client in the book or CD (for each client identifier) Item identifier (product identifier) and a list of evaluation values of the item) (FIG. 3 (2)) or a list of products such as books and CDs that the client is interested in (item identifier for each client identifier ( (Product identifier) list) (Fig. 3 (3)) is obtained from each client That.
[0039]
The client identifier need only be able to identify different clients in the client preference estimation factor information list providing server 1 for each client identifier, and need not be a unique client identifier across each item provider server.
[0040]
(2) Transmission of the client preference estimation factor information list for each client identifier from the client preference estimation factor information list providing server 1 for each client identifier to the recommended item introduction server 2 (client preference estimation factor information list processing unit 4) (step 12)
The client preference estimation factor information list providing server 1 for each client identifier aggregates client preference estimation factor information collected from each client, and the client preference estimation factor information list for each client identifier is recommended in a map format as shown in FIG. 2 (client preference estimation factor information list processing unit 4).
[0041]
(3) Processing of client preference estimation factor information list for each client identifier in recommended item introduction server 2 (client preference estimation factor information list processing unit 4) (step 13)
The recommended item introduction server 2 (client preference estimation factor information list processing unit 4) processes the received client identifier estimation factor information list for each client identifier so as to respond to the recommended item introduction request from the client 6, and estimates the client preference. Generate a factor information map. This client preference estimation factor information map is, for example, a list of item identifiers and evaluation values of items accessed by the client for each client identifier (FIG. 5). Moreover, in an example, it is a list | wrist of the client identifier and evaluation value of the client which is accessing the said item for every item identifier (FIG. 6). Another example is a list of correlation values between correlated items obtained as a result of analyzing client preference estimation factor information. In addition, the case where an evaluation value is not included in FIG. 5 and FIG.
[0042]
The client identifier may be an identifier that can be uniquely identified in the recommended item introduction server 2. It is not necessary for the client identifier to be unique across each item provider / server, and it is not necessary for the client identifier to be transmitted from the client preference estimation factor information list providing server 1 for each client identifier.
[0043]
The recommended item introduction server 2 sends the client preference estimation factor information list for each client identifier to the client preference estimation factor information list for each client identifier by sending a monthly charge or a pay-per-use charge for each information amount. A fee is paid to the operating company of the client preference estimation factor information list providing server 1 for each identifier.
[0044]
(4) Transmission of client preference factor information map from the client preference estimation factor information list processing unit 4 to the recommended item introduction service unit 5 (step 14)
The client preference estimation factor information list processing unit 4 sends the client preference estimation factor information map generated in (3) to the recommended item introduction service unit 5.
[0045]
2. Get recommended item introduction request (Figure 7)
(5) Send recommended item introduction request from client 6 to item provider server 3 (step 21)
The client 6 sends a recommended item introduction request to the item provider server 3 together with a request identifier and client preference estimation factor information. The client preference estimation factor information is, for example, a list of “item identifier and access date / time” for each client, a list of “item identifier and evaluation value for the item” for each client, or a list of “item identifier” for each client. is there. Here, information such as a client identifier for uniquely identifying the client is unnecessary.
[0046]
(6) Sending a recommended item introduction request from the item provider server 3 to the recommended item introduction server 2 (recommended item introduction service section 5) (step 22)
The item provider server 3 transmits a recommended item introduction request together with a request identifier and client preference estimation factor information to the recommended item introduction server 2 (recommended item introduction service unit 5). Here, information such as a client identifier for uniquely identifying the client is unnecessary.
[0047]
3. Introduction of recommended items (Figure 8)
(7) Send recommended item list from recommended item introduction server 2 (recommended item introduction service section 5) to item provider server 3 (step 31)
The recommended item introduction server 2 (recommended item introduction service section 5) creates a list of recommended items based on the client preference estimation factor information map received in (4) and the client preference estimation factor information received in (6). The recommended item list is transmitted to the item provider server 3 together with the request identifier. In the transmission of the recommended item list, a fee is paid from the item provider server 3 to the operating company of the recommended item introduction server 2 by a fixed amount monthly charge or a pay-per-use charge for each information amount.
[0048]
(8) Send recommended item list from item provider server 3 to client 6 (step 32)
The item provider server 3 transmits the recommended item list received in (7) to the client that has received the request in (5).
[0049]
As a modification of the above, there is a method in which the recommended item introduction service unit 5 is installed in the item provider server 3 and the client preference estimation factor information list processing unit 4 and the recommended item introduction service unit 5 are connected via a network. .
[0050]
As another modification of the above, as shown in FIG. 9, the client preference estimation factor information list providing server 1 for each client identifier is omitted, and the recommended item introduction server 2 replaces the client preference estimation factor information list processing unit 4 with a client. Alternatively, a client preference estimation factor information group processing unit 4 ′ that directly collects client preference estimation factor information may be provided.
[0051]
FIG. 10 shows a case where the present invention is applied to recommended product (advertisement) introduction such as advertisement display, and FIG. 11 shows an operation in this case.
[0052]
(9) Product (advertisement) information delivery request from the product (advertisement) information delivery source company 7 to the item provider server 3 (step 41)
The product (advertisement) information delivery source company 7 sends the product (advertisement) information to be delivered together with the product (advertisement) information delivery request to the item provider server 3.
[0053]
When delivering product (advertisement) information, flat-rate billing every month, etc., or pay-per-use for each display amount of product (advertisement) information, or pay-per-use for each view of product (advertisement) information, or purchase of products Charges are paid from the merchandise (advertisement) information delivery source company 7 to the operating company of the item provider server 3 with a pay-per-use charge for each number or purchase amount.
[0054]
(7) Send recommended item list from recommended item introduction server 2 (recommended item introduction service section 5) to item provider server 3 (step 42)
The recommended item introduction server 2 (recommended item introduction service section 5) creates a list of recommended items based on the client preference estimation factor information map received in (4) and the client preference estimation factor information received in (6). The recommended item list is transmitted to the item provider server 3 together with the request identifier. In the transmission of the recommended item list, a fee is paid from the item provider server 3 to the operating company of the recommended item introduction server 2 by a fixed amount monthly charge or a pay-per-use charge for each information amount.
[0055]
[8] 'Send recommended item list from item provider server 3 to client 6 (step 43)
The item provider server 3 extracts the product (advertisement) information received in (9) from the recommended item list received in (7) and transmits it to the client that has received the request in (5).
[0056]
As a modified example, the recommended item introduction server 2 identifies / extracts product (advertisement) information recorded in the item provider server 3 and delivers a list of recommended product (advertisement) information to the item provider server 3. It is also possible to do.
[0057]
Next, a method for creating a recommended item list will be described with reference to FIG.
[0058]
As an analysis method for transmitting the recommended item list (7) from the recommended item introduction server 2 to the item provider server 3, a general collaborative filtering technique, a data mining technique, or the like can be used. For example, assume that the following conditions exist.
[0059]
Client preference estimation factor information list for each client identifier sent by sending the client preference estimation factor information list for each client identifier from the server 1 to the recommended item introduction server 2 from the client identifier client preference estimation factor information list providing server 1 (2) Is a list of item identifiers for each client identifier arranged in the order of the most recent access date / time based on the access history (client identifier, item identifier, access date / time) of the item provider server 3.
[0060]
Client preference obtained by processing the client preference estimation factor information list for each client identifier so that the recommended item introduction server 2 that has received the client preference estimation factor information list for each client identifier can respond to the recommended item introduction request from the client 6 A list of item identifiers accessed by the client for each client identifier uniquely reassigned in the recommended item introduction server 2 (referred to as “item list for each client”), and an estimated factor information map for each item identifier This is a list of client identifiers of clients accessing the item (referred to as “client list for each item”).
[0061]
The client preference estimation factor information sent together with the recommended item introduction request is a list of “item identifiers”.
[0062]
In this case, the recommended item list can be created by the following method.
[0063]
<1> Extraction of related client list
In the recommended item introduction server 2, for each item identifier in the list of “item identifier” which is the client preference estimation factor information sent from the item provider server 3 together with the recommended item introduction request, “list of clients for each item” Extract all included client identifiers. The list of client identifiers extracted by this method is called a related client list.
[0064]
<2> Narrow down related client list
For each client identifier of the related client list extracted in <1>, a list of item identifiers included in the “list of items for each client” is extracted, and the item identifiers in the extracted item identifier and “item identifier” list are extracted. Is counted for each client identifier in the related client list extracted in <1>, the client identifiers in the related client list are rearranged in the order of the largest overlap, and the top N1 (positive alignment) of the rearranged related client list. (Numerical value) to the new related client list.
[0065]
<3> Extraction of related item list
All item identifiers included in the “list of items for each client” are extracted for each client identifier in the new related client list. The list of item identifiers extracted by this method is called a related item list.
[0066]
<4> Narrow down related item list
For each item identifier of the related item list extracted in <3>, a list of client identifiers included in the “list of clients for each item” is extracted, and the extracted client identifier and each client identifier of the new related client list Are counted for each item identifier of the related item list extracted in <3>, and the item identifiers of the related item list are sorted in descending order of overlap, and the top two N2 items in the sorted related item list (positive (Integer value) is a new related item list, and this new related item list is a recommended item list to be sent to the client who has requested the recommended item introduction.
[0067]
FIG. 13 shows a configuration example of a computer system that implements the recommended item introduction server 2 in FIG. The communication device 51 is for communicating with the client preference estimation factor information list providing server 1 and the item provider server 3 for each client identifier. The storage device 52 stores a client preference estimation factor information list for each client identifier, a client preference estimation factor information map, and the like. The storage device 53 is a hard disk. The recording medium 54 is a floppy disk, CD-ROM, magneto-optical disk, or the like, in which a recommended item introduction program consisting of processing by the client preference estimation factor information list processing unit 4 and the recommended item introduction service unit 5 is recorded. The data processing device 55 includes a CPU and an interface, reads a recommended item introduction program from the recording medium 54, and executes it.
[0068]
The present invention is expected to be particularly effective in the following business model.
[0069]
For example, it is assumed that the recommended item introduction server of the present invention is operated by a large-scale book EC site (large site). The large site uses the access history in its own site as the client preference estimation factor information, and holds the client preference estimation factor information map.
[0070]
On the other hand, assuming that there are other small book EC sites (small sites), each small site does not have its own recommended item introduction server, or a sufficient amount to create a client preference estimation factor information map. Suppose you do not have an access history.
[0071]
In this case, the large site can provide a recommended item introduction service to the small site group as follows.
[0072]
(1) When there is a recommended item introduction request from a client at a small site, the small site extracts the client preference estimation factor information of this client from the access history in the own site and sends it to the large site.
[0073]
(2) The large site creates a recommended item list (book list) from the received client preference estimation factor information and sends it to the small site.
[0074]
(3) The small site that has received the recommended item list returns this recommended item list to the client.
[0075]
In this business model, the following benefits can be expected.
[0076]
-Even if a small site does not have or cannot have its own recommended item introduction server, it is possible to provide a recommended item introduction service to users of the site.
[0077]
-Since the client preference estimation factor information sent to a large site is not specified to which user, privacy can be protected.
[0078]
-Large sites can offer recommended item introduction services to small sites without collecting the client preference estimation factor information map, which is know-how of large sites, and collect fees from these services.
[0079]
Moreover, the following can be considered as a modification of embodiment of this invention mentioned above.
[0080]
・ The network is not limited to the Internet. There are digital interactive TV, information home appliance network and so on.
[0081]
Clients include PCs (Personal Computers), mobile phones, digital interactive TVs, information appliances, communicable word processors, communicable game machines, communicable watches, PDA (Personal Digital Assistants) and other mobile communicable terminals There is.
[0082]
-Input / output is not limited to Web pages. Using an electronic mail, it is also possible to transmit an access history, a recommended item introduction request, a client identifier, etc. as an electronic mail document or title, or use an electronic mail address instead of the client identifier. The recommended item list can also be sent as an e-mail document.
[0083]
The client identifier can be managed in association with a customer identifier different from the client identifier. The customer identifier includes a unique identifier arbitrarily input on the client side, an ID assigned by the recommended item introduction device separately from the client identifier, and the like.
[0084]
-Introduction of the recommended item list can be displayed, for example, overlaid on the TV screen and over the game display.
[0085]
【The invention's effect】
As described above, the present invention provides a recommended item introduction service to a plurality of item provider servers based on the client preference estimation factor information list for each client identifier provided by a third party company or the like. The recommended item introduction service can be provided to the client using the item provider server without acquiring the client preference estimation factor information list for each client identifier from each item provider server that wants to use the item introduction service.
[0086]
Furthermore, since a client identifier is not required as information necessary for making an introduction, it is not necessary to assign a client identifier for uniquely identifying a client across a plurality of item provider servers. The client preference estimation factor information, which is the privacy information of the client, can be prevented from leaking in association with the client, and privacy can be protected.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration example of a recommended item introduction system according to an embodiment of the present invention.
FIG. 2 is a flowchart of processing for acquiring a client preference estimation factor information list for each client identifier in the recommended item introduction system of FIG. 1;
FIG. 3 is a diagram illustrating an example of client preference estimation factor information that can be used in the recommended item introduction system of FIG. 1;
FIG. 4 is a diagram showing an example of a client preference estimation factor information list that can be used in the recommended item introduction system of FIG. 1;
FIG. 5 is a diagram showing an example of a client preference estimation factor information map that can be used in the recommended item introduction system of FIG. 1;
6 is a diagram showing another example of a client preference estimation factor information map that can be used in the recommended item introduction system of FIG. 1. FIG.
7 is a flowchart of processing for obtaining a recommended item introduction request in the recommended item introduction system of FIG. 1; FIG.
FIG. 8 is a flowchart of recommended item introduction processing in the recommended item introduction system of FIG. 1;
FIG. 9 is a block diagram illustrating another configuration example of the recommended item introduction system according to the embodiment of the present invention.
FIG. 10 is a diagram showing a configuration example when the recommended item introduction system of the present invention is applied to recommended product (advertisement) introduction.
FIG. 11 is a flowchart of recommended product (advertisement) introduction processing in the recommended item introduction system of FIG. 10;
FIG. 12 is a diagram for explaining a recommended item list creation method that can be used in the recommended item introduction system of the present invention;
FIG. 13 is a diagram showing a configuration example of a computer system that implements a recommended item introduction server in the recommended item introduction system of FIG. 1;
FIG. 14 is a diagram showing a configuration example of a conventional recommended item introduction system.
15 is a flowchart showing the operation of the conventional recommended item introduction system of FIG.
[Explanation of symbols]
1 Client preference estimation factor information list providing server for each client identifier
2 Recommended Item Introduction Server
3 Item provider server
4 Client preference estimation factor information list processing section
4 'Client preference estimation factor information group processing section
5 Recommended Item Introduction Service Department
6 Client
7 Product (advertising) information distribution company
51 Communication device
52 Storage device
53 Recording media
54 Data processing equipment
61 Recommended Item Introduction Server
62 Item Provider Server
63 clients

Claims (4)

A recommended item introduction method by a recommended item introduction server that provides a recommended item introduction service to a plurality of item provider servers that provide items to multiple clients via a network,
(A) The recommended item introduction server is
Collect client preference estimation factor information for each client identifier from a plurality of clients From a client preference estimation factor information list providing server for each client identifier, client preference estimation for each client identifier, which is information used to estimate each preference of the plurality of clients Factor information list (item identifier of items browsed / purchased in item provider server for each client identifier and date / time of browsing / purchase of the item, or item browsed / purchased in item provider server for each client identifier And a list of evaluation values indicating the degree of interest of the client with respect to the item identifier or a list of item identifiers of items that the client is interested in for each client identifier). And each client preference estimating factor information list reception step Ianto identifier,
A client preference estimation factor information map processing step for processing the client preference estimation factor information list for each client identifier into a client preference estimation factor information map;
A client preference estimation factor information list for each client identifier and a client preference estimation factor information map storage step for storing the client preference estimation factor information list for each client identifier and a client preference estimation factor information map in a storage device;
(B) The recommended item introduction server is
A client that receives client preference estimation factor information indicating a preference of a certain client together with a recommended item introduction request of the certain client from an item provider server that provides an item to a plurality of clients via a network without specifying the certain client A preference estimation factor information receiving step;
For each item identifier in the list of item identifiers constituting the client preference estimation factor information, all client identifiers included in the list of client identifiers for each item constituting the client preference estimation factor information map are extracted, and the related client list The first recommended item list creation step to create
For each client identifier in the related client list, an item identifier included in a list of item identifiers for each client constituting the client preference estimation factor information map is extracted, and the extracted item identifier and the client preference estimation factor information are configured. The number of overlaps with the item identifiers of the list of item identifiers to be counted is counted for each client identifier of the related client list, the client identifiers of the related client list are rearranged in descending order of the counted number of overlaps, and A second recommended item list creation step of extracting a first predetermined number from the top and creating a new related client list,
For each client identifier of the new related client list, a third recommended item list creation step of extracting an item identifier included in the list of item identifiers for each client and creating a related item list;
For each item identifier in the related item list, a client identifier included in the list of client identifiers for each item is extracted, and the number of overlaps between each extracted client identifier and each client identifier in the new related client list is calculated. Counting for each item identifier of the related item list, sorting the item identifiers of the related item list in descending order of counting, extracting a second predetermined number from the top of the sorted item identifiers, the recommended item A fourth recommended item list creation step to create a list;
And a recommended item list transmission step of transmitting the recommended item list to the item provider server. A recommended item introduction server that provides recommended item introduction services to multiple item provider servers that provide items to multiple clients over a network,
(A) Collecting client preference estimation factor information for each client identifier from a plurality of clients For each client identifier, which is information used to estimate each preference of the plurality of clients from the client preference estimation factor information list providing server for each client identifier Client preference estimation factor information list (list of item identifiers of items browsed / purchased on item provider server for each client identifier and date / time of browsing / purchase of the item, or browse / purchase on item provider server for each client identifier A list of evaluation items indicating the item identifiers of the selected items and the degree of client interest in the items, or a list of item identifiers of items of interest to the client for each client identifier) A communication device,
A client preference selection factor information list processing unit that processes the client preference estimation factor information list for each client identifier into a client preference estimation factor information map;
A storage device for storing a client preference estimation factor information list and a client preference estimation factor information map for each client identifier;
(B) From the item provider server that provides an item to a plurality of clients via a network, client preference estimation factor information indicating the preference of the client without specifying the client with the recommended item introduction request of the client A communication device for receiving;
For each item identifier in the list of item identifiers constituting the client preference estimation factor information, all client identifiers included in the list of client identifiers for each item constituting the client preference estimation factor information map are extracted, and the related client list And an item identifier included in the list of item identifiers for each client constituting the client preference estimation factor information map for each client identifier of the related client list, and the extracted item identifier and the client The number of overlaps with the item identifiers in the list of item identifiers constituting the preference estimation factor information is counted for each client identifier in the related client list, and the related client resources are listed in descending order of the counted number of overlaps. Means for extracting a first predetermined number from the top of the sorted client identifiers, creating a new related client list, and for each of the extracted client identifiers, an item for each client. Means for extracting item identifiers included in the list of identifiers and creating a related item list;
For each item identifier in the related item list, a client identifier included in the list of client identifiers for each item is extracted, and the number of overlaps between each extracted client identifier and each client identifier in the new related client list is calculated. Counting for each item identifier of the related item list, sorting the item identifiers of the related item list in descending order of counting, extracting a second predetermined number from the top of the sorted item identifiers, the recommended item A recommended item introduction service department having means for creating a list;
A communication device for transmitting the recommended item list to the item provider server;
The recommended item introduction server characterized by having. A recording medium recording a recommended item introduction program for causing a computer to execute the recommended item introduction method according to claim 1. A recommended item introduction program for causing a computer to execute the recommended item introduction method according to claim 1.

Images