JP2000148847A - Method and system for providing commodity information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a system which supplies commodity information through a network to provide such commodity information that is not aware of by users, but is estimated to be valuable. SOLUTION: In a personal data file 22, the number of commodities proposed to each member is stored correspondingly to the membership number of each member. A proposed-commodity list 23 contains the candidates for proposed commodities. When an individual proposed-commodity calculating and processing section 37 receives the membership number of a member through a Mohr server 3, removes the commodities already proposed to the member from the list 23 by referring to the data file 22, and sends the numbers of remaining proposed commodities to the server 3. The server 3 accepts the access from a client PC 14 and sends the membership number of the member to a commodity selection server 2. In addition, the server 3 fetches commodity information from contents 30 in accordance with a proposed-commodity list received from the server 2 and sends the fetched information to the PC 14.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and system for providing merchandise information via a network, and in particular, to provide merchandise information which is not considered by the user and which is estimated to have high value. Methods and systems are provided.

[0002]

BACKGROUND OF THE INVENTION There are various types of relationships between information and users. Looking at the axis of preference, there is a relationship of likes and dislikes. Looking at the axis of necessity,
There is a relationship of interest. The depth of the relationship between the information and the user differs depending on the content of the information and the characteristics of each user. This depth has a positive and a negative direction, meaning that deeper in the positive direction is more valuable to the user, and deeper in the negative direction prevents the user from reaching valuable information, It means feeling uncomfortable.

[0003] Among the mechanisms for providing information using a network, there is a method of providing valuable information to users by selecting information deeper in relation to each user in a positive direction.

[0004] For example, in Japanese Patent Application Laid-Open No. 9-269923, "Information and advertisement distribution system and method", when an information provider provides information according to a classification created by an information provider, a user is interested in a topic or interest. It has a function that can specify the classification and sub-classification of topics without any information, and news articles are selected and distributed for each user accordingly. In this method, the user selects a positive category and a negative category with respect to the information category provided by the information provider, receives the information included in the plus category based on the selection, Information can be received by excluding information included in negative classification items from all information. The information provider selects information having a deeply related relationship, that is, information on the category that the user has determined to be valuable, and provides the information to the user in accordance with an instruction set in advance for the category item by the user. In addition, information that belongs to the category that is deeply related to the user, that is, that is judged to be unworthy or uncomfortable by the user, is excluded from all provided information, and the remaining information that is not excluded is provided to the user. I do.

[0005]

Normally, the information that the user actively obtains by himself is the information that the user himself is conscious of having a positive relationship with himself. The user obtains information that is considered to be positively related by various means such as search, or information in an area that is considered to be positively related. In addition, when acquiring information, users consciously exclude information that they consider to be deeply related to themselves. At the search stage, it is possible to exclude information and fields that are considered to have a negative relationship deeply, or to judge at the stage of finally obtaining the information.

[0006] As described above, the user obtains information that is considered to be deeply related to the user and obtains more valuable information by excluding information that is considered to be deeply related to the user. However, among the information available on the network, there is information that the user is not aware of the depth of the relationship, and the number is based on the vast amount of information existing on the network, It can be said that there is much more than information that is conscious of.

[0007] As described above, regarding information on which the user is conscious of the relationship, the user himself / herself actively collects information.
Although it can be excluded, there may be information that the user recognizes as valuable for information that the user is not aware of the relationship, but there is no way to find out from the user side.

[0008] In the above prior art, an information provider gives an instruction in advance that a user needs or does not need to classify and subclassify information, and selects and provides information according to the instruction. Such a conventional technique has the following problems. That is, it is easy for the user himself / herself to acquire information belonging to the classification that can be designated by the user assuming that the relationship is deeper in the plus direction. In other words, the information provided here can be obtained by a user himself / herself by a method such as a search, and this providing method is not necessarily required. Information that is clearly recognizable as having a deeper relationship in the positive direction, so if the user has not already acquired the information, it is valuable information for the user, but in many cases it has already been acquired in another way, in which case the user The value of this method of providing information is reduced.

[0009] Further, since the unique information classification indicated by the information provider does not always match the classification individually required by each user, unnecessary information is included in the classifications determined to be valuable and selected. It is difficult to select only information that has a deeper relationship with the user in the positive direction, because the information that the user expected to provide is included in a classification different from the selected classification. In other words, there is a possibility that the accuracy of the provided information is low even though the user selects the classification of the information by himself.

As another method for solving such a problem, for each piece of information provided by an information provider, an individual user evaluates the relationship with the information by himself / herself, thereby classifying the user into a unique classification. There is a method that uses the results to select information that has a deep relationship with the user in the positive direction and that the user has not yet obtained, and provides valuable information to users who do not know the user. In this method, a user who is to be provided with information evaluates the relationship with respect to information that can be evaluated among a plurality of pieces of information initially proposed. Users having a similar tendency to the evaluation result are assigned to the same group. Then, from among the information that has not been evaluated by the user who is to be provided with the information, information that has already been evaluated by other users in the same group to have a deeper relationship in the positive direction is selected and provided. In other words, the information provider has a similar tendency for users in the same group to evaluate, so if a user in the same group evaluates that the relationship is deep in the positive direction, the same applies to users who have not yet evaluated in the group. And provide information, predicting that it will be rated “deep in the direction”.

However, this method still has a problem.
That is, based on the evaluation of the user who wants to receive the information, the information that the information provider has provided is selected because other users who have a similar tendency to the evaluation consider that the relationship is close to positive. Only information in the vicinity of information evaluated as having a positive relationship by the user who is to receive the information is selected.

The number of pieces of information that a user contacts is limited. In addition, since it is a considerable work load for the user to evaluate each relationship for each piece of information, the number of evaluations is limited. From this, it can be said that the information around the user's consciousness is only a small part of the information that can be provided. There is a possibility that some information outside the vicinity of the user's consciousness is left unprovided even though there is information deep in the positive direction with the user.

An object of the present invention is to provide an information providing system utilizing a network in which a user selects information which is not aware of the relationship between the user and the information, ie, selects information which cannot be actively acquired by the user himself. To provide.

Another object of the present invention is to provide a method of evaluating information complementary to a user's own information in addition to the evaluation of the relationship with the user's own information, thereby reducing the user's work load and the amount of information evaluated. To increase the range / range.

[0015]

SUMMARY OF THE INVENTION The present invention is a system for providing commodity information having a commodity selection server and an information transmitting server, wherein the commodity selection server uses the commodity information in accordance with an identifier of each user. First storage means for storing an identifier of a product proposed to the user, second storage means for storing a list of product candidates providing preset product information, and a user identifier from the information transmission server. Processing means for receiving, from the suggested product list, referring to the first and second storage means, excluding the products already proposed to the user, and transmitting the identifiers of the remaining proposed products to the information transmission server; A transmitting server, storing means for storing product information of the proposed product, processing means for receiving an access from the user's client device, and sending an identifier of the user to the product selection server; Receiving an identifier of the proposed product from goods selected server retrieves the product information of the designated proposed product from the storage means and said system for providing product information and a processing means for transmitting to the client device of the user.

Further, according to the present invention, an identifier of a product proposed to the user and an identifier of a product evaluated negatively by the user among the proposed products are stored in correspondence with the identifier of each user. A first storage unit, and a second storage unit that stores a list of product candidates that provide preset product information.
Storage means for arranging identifiers of other related goods for which a predetermined ratio or more of users have been evaluated negatively in correspondence with the identifiers of the goods evaluated negatively by the user; When providing product information to a user for product proposals, a product identifier that has been proposed to the specific user is retrieved by referring to the first storage unit, and a proposal is made from a proposed product list in the second storage unit. Removed product identifiers,
With reference to the third storage means, the identifier of the related product of the product evaluated by the specific user to be negative is deleted from the proposed product list, and the identifier of the remaining proposed product is replaced with product information for the specific user. It is characterized by a product selection server having a processing means to be a target product to be provided.

[0017] In addition to the products for which the user directly instructs the negative evaluation, the products for which the average access time is less than the predetermined value are regarded as the products for which the user has the negative evaluation, so that the user can unconsciously. The user's evaluation can be read from the action taken.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a system for carrying out Internet shopping according to the present embodiment. This system is roughly composed of three servers, a store server 1, a mall server 3, and a product selection server 2. The store server 1 exchanges information with the POS 7 installed in the actual store, collects sales information of the actual store, and sends necessary data to the product selection server 2. The mall server 3 is a server having a function of transmitting information to the Internet 13 and receiving orders and settlement necessary for shopping related to virtual stores on the Internet 13, and performs processing related to providing and selling information to members.
It also collects information and action histories obtained through exchanges with members and sends necessary data to the product selection server 2. The product selection server 2 performs a process of selecting a product to be proposed to each member based on the information collected from the store server 1 and the mall server 3, and when the product is selected, proposes the product to the member. To the mall server 3.

In FIG. 1, the actual store is a PO with a communication device.
In S7, the sales information of the store is collected. The store server 1 is a computer, and is a PO with a communication device of each physical store.
The communication devices 8 and 9 for performing communication with S7 and the storage device 4 are connected. The store server 1 collects the sales information of the store from the POS 7 with the communication device of the actual store through the communication device 8, and stores the collected information in the store product sales information file 21 on the storage device 4. Furthermore, this store server 1
The product sales information calculation processing unit 31 calculates the sales numerical value of each product from the data of the store product sales information file 21 stored in the storage device 4 and sends it to the product selection server 2 through the communication device 9.

The mall server 3 is connected to the Internet 13
It is a computer that has a function as an information transmission server to the customer, a function of receiving orders / settlement, and a function of collecting an access log containing product information. This mall server 3 is a communication device 1
5, 16 and the storage device 6 are connected. Mall server 3
Transmits information to the Internet 13 through the communication device 16. Then, communication is performed with a user who accesses the Internet 13 from the client PC (personal computer) 14 through the communication device 17. The selected product transmission processing unit 38 stores the content 30 stored in the storage device 6 based on the product number sent from the product selection server 2 when the site proposes a product to the member.
Extract the relevant information from and create a file to send to each member. After providing the information to the member, the contents of the access performed by the member are recorded in the access log 28. At regular intervals, an Internet product-specific result 29 is calculated from the access log 28. The access log 28 and the Internet product-specific results 29 stored in the storage device 6 are transmitted from the mall server 3 through the communication device 15 to the product selection server 2.
Sent to

The product selection server 2 connects the storage device 5, the communication devices 11 and 12, and the input device 10, and performs a series of processes for selecting a proposed product for each user. Store product sales information file 21 sent from store server 1
Then, processing for selecting a proposed product is performed based on the data of the access log 28 transmitted from the mall server 3 and the result 29 of the Internet product.

The product master 27 stores initial data input from the input device 10 and information on sales and access sent from the store server 1 and the mall server 3. The ID-ordered access log 26 is a file edited by the access log processing unit 35 to arrange the access logs 28 sent from the mall server 3 in the order of the member numbers. The individual access data 25 stores information on the number of products purchased for each member, Internet access status, and members' evaluation of each product.
The individual access data calculation processing unit 36 is a program for creating the individual access data 25 from the ID-based access log 26. The personal data file 22 stores, for each member, the attribute of the member and the number of a product corresponding to each value parameter. The personal data creation processing unit 33 is a program for creating the personal data file 22 from the individual access data 25. The proposed merchandise list 23 stores the number of sales at an actual store, the number of sales and access status regarding the Internet, the overall evaluation of the merchandise, and the like for each merchandise. The proposed merchandise list creation processing unit 32 stores the merchandise master 27
Is a program for creating the suggested product list 23 based on the The negative product group 24 stores a list of products with negative evaluations that are highly relevant in association with products that have been negatively evaluated by members. The minus product group creation processing unit 34 is a program for creating the minus product group 24 from the product master 27 and the personal data file 22.

The individual proposed product calculation processing section 37 is a program for selecting a product to be proposed for each member based on the personal data file 22, the proposed product list 23, and the negative product group 24. The number of the selected product is
The product is sent to the mall server 3 through the communication devices 12 and 15 and the corresponding product is
7 to the member having the client PC 14.

A program having a processing procedure of any one of the processing units of the product selection server 2 is stored in a storage medium, and is read into a memory of the product selection server 2 via a driving device connected to the product selection server 2, or It can be transmitted to the memory of the product selection server 2 via a drive device of another computer or a network, and can be executed by the product selection server 2.

FIG. 2 shows an example of data constituting the personal data file 22. The attributes of the member number 41, age 42, and gender 43, which are uniquely given on the shopping site and serve as member identifiers, are information provided and input by the user when the user registers as a member to use the shopping site. It is. The plus commodity 44, the minus commodity 45, and the unresponsive commodity 46 store the number of the commodity for which information was provided to the member corresponding to the classification of each value parameter.

The plus item 44 means a favorite item among the items for which the member has already obtained information. Among the products for which information has been acquired, the products are classified as those that have already been purchased or that may be purchased. The minus commodity 45 means a commodity which the member has not obtained yet among the commodities for which information has already been acquired. Among the products for which information has been acquired, the products are classified as those having a low possibility of purchase. Unresponsive product 46
Is a product for which product information has been provided from the site, but the member has not taken an action to determine a positive or negative evaluation. Since there is no response even if the information is provided, the value parameter value is estimated to be negative, but is classified as one that may be proposed again. The proposed merchandise history 47 is a history of merchandise proposed in the past from the site side to this user.

FIG. 3 shows an example of data constituting the proposed merchandise list 23. Each record of the proposed product list 23 includes data items such as a product number which is a product identifier, access count, access time, actual store sales, Internet sales, registration days, and comprehensive evaluation. The number of accesses and the access time
The total number of access times and total access times for that product made through
It has nothing to do with actual store sales. The number of registration days is a registered period (number of days) and is a data item used as an index for determining the priority of each product. The comprehensive evaluation is a point calculated by arbitrarily weighting data items from the number of times of access to the number of days of registration as an index. The proposed product list 23 is a candidate for a proposed product, in which products are rearranged in descending order of the overall evaluation and the top products are taken out.
Used as an original list from which the site selects products to propose to members.

FIG. 4 shows an example of data constituting the minus commodity group 24. Many of the members who have evaluated the value of the value parameter as negative for the product indicated by the product number are grouping products that have also been negatively evaluated as related inappropriate products. It is used to delete in advance from the product information in a range that is not conscious of the user, related product information in which the value of the value parameter is expected to be negative for the user.

FIG. 5A is a flowchart showing the flow of processing of the selected commodity transmission processing section 38 of the mall server 3, and FIG.
b is a personalized product calculation processing unit 3 of the product selection server 2
9 is a flowchart showing the flow of the process of FIG. Upon receiving the access from the client PC 14 of the member, the selected product transmission processing unit 38 of the mall server 3 (step 5).
1) The member number of the member is sent to the product selection server 2 (step 52).

Next, referring to FIG. 5B, the individual proposed product calculation processing section 37 of the product selection server 2 receives the transmitted member number (step 53),
2. Extract the data of the corresponding member from Step 2 (Step 5)
4). Next, the individual proposed product calculation processing unit 37 creates a copy of the proposed product list 23 for the member (step 55). Next, with respect to all the products listed in the plus product 44 of the extracted member information, if the copied proposed product list 23 includes the product, data on the product is deleted (step 56). By this processing, among the products that the user is conscious of, products with a positive value parameter value are deleted from the list of products proposed. In addition, the unresponsive product 46 of the retrieved member information
If the product is included in the copy of the same proposed product list 23 for all the products described in (1), the data on the product is deleted (step 57). Further, with respect to all the products described in the minus product 45 of the extracted member information, information on the corresponding product is taken out from the minus product group 24 (step 5).
8). Then, the same proposed product list 2 is applied to all the products listed in the negative product 45 and the related inappropriate product.
If the product is included in the copy of No. 3, the data relating to the product is deleted (step 59). In these steps 58 and 59, it is expected that the value parameter value of the user is negative and the value parameter value of the user is negative within the range that the user is not aware of. Products have been deleted.
The individual proposed product calculation processing unit 37, as a result of the above work,
From among the products remaining in the proposed product list 23, at least one product with a higher overall evaluation is selected as a product to be proposed, and the product number is entered in the proposed product history 47 of the personal data file 22. Further, the product number is sent to the mall server 3 (step 60). In the case of a member who has just joined, there is no value parameter for that member, so that a higher-rank product in the proposed product list 23 is selected as a product to be proposed.

Next, returning to FIG. 5A, the selected commodity transmission processing section 38 of the mall server 3 receives the transmitted commodity numbers (step 61), and transmits images, titles, texts, etc. corresponding to those commodity numbers. The information is extracted from the content 30, a file in a format for transmitting information to the member is created, and transmitted to the client PC 14 via the Internet 13 (step 62).

Steps 56 to 5 in the above description
The process of narrowing down proposed products to members corresponding to the processes up to 9 will be described using a specific example. FIG. 3 shows an example of a copy of the proposed product list 23, FIG. 2 shows an example of the personal data file 22, and FIG. 4 shows an example of the minus product group 24. A member who proposes a product has a membership number of ADBM0000.
Then, first, in the information of the member number 41 = ADBM0000 of the personal data file 22 in FIG.
Examine the products listed in 4. This plus product 44
Describes one product of 6493330002. When this product is searched for in the proposed product list 23 of FIG. 3, it is in the first line of the item of the product number, so this line is deleted. Subsequently, the member number 41 = A of the personal data file 22 in FIG.
In the information of DBM0000, a product described in the unreacted product 46 is checked. This unresponsive product 46 has 1
Since one product of 19229689 is described, FIG.
When this product is searched for a copy of the proposed product list 23, it is located on the fourth line of the item number, so this line is deleted. Finally, the member number 4 of the personal data file 22 in FIG.
1 = Minus goods 45 in the information of ADBM0000
Examine the products listed in. This minus product 45
Describes one product of 35242242100. In the minus merchandise group 24 in FIG. 4, the merchandise described in the related inappropriate merchandise for the corresponding merchandise is checked. Here, for minus product 353242100,
Related inappropriate products are 0467732601 and 0448771
41. When these three products are searched for a copy of the proposed product list 23 in FIG. 3, only the number 046732601 is on the fifth line of the item number, so this line is deleted.

By the above processing, the six products in the proposed product list 23 of FIG. 3 become three products as shown in FIG. The merchandise number 046353362 of the merchandise having the highest overall evaluation among the merchandise selected in this way is sent to the mall server 3, and the merchandise is proposed to the member. FIG. 7 shows the mall server 3
Proposes products to members from the client PC
14 is an example of a screen displayed on the fourteen display devices, and a value parameter value can be directly obtained from a member by a button “liked” or “disliked” on the screen. "I like" is equivalent to a direct positive evaluation of a member, and "I dislike" is equivalent to a direct negative evaluation. The access log 28 acquired by the mall server 3 is:
A log that does not include a product number is recorded for a product proposal page or a page such as a list of multiple products in FIG. For a page relating to a single product, if “I like” in FIG. 7 is selected, the product number and the positive evaluation by the member are recorded as a log. "I do not like it"
In the case where is selected, the product number and the negative evaluation are similarly recorded. When the member moves to another action without selecting either the button “I like it” or “I don't like” in response to this product proposal, the access log 28 has no log with the product number, and this product is In the subsequent processing, it is stored in the unresponsive product 46 of the personal data file 22 of the member.

Up to this point, the description of the personal data file 22, the proposed merchandise list 23, and the minus merchandise group 24, and the selection of the merchandise proposed by the site to the member using these data and the process of making the proposal have been described. From here, the creation process of the personal data file 22, the suggested product list 23, and the minus product group 24 will be described. These files are updated at regular intervals.

FIG. 8 shows an example of data constituting the individual access data 25. Each record of the individual access data 25 has data items of member number, product number, number of purchases, positive evaluation, negative evaluation, access time, and number of accesses. For the number of purchases, the number of purchases is entered for each product number if there is a purchase record. In the present embodiment, the product evaluation directly performed by the member is performed by selecting the buttons of “I like” and “I do not like”. The product for which "I liked" has a positive value parameter value for the member, and when it is selected, detailed information of the product can be obtained.
A product for which "I do not like" is selected as a member whose value parameter value is negative. "Positive evaluation"
Sets a flag when the member evaluates that the value parameter value of the product is positive. Similarly, “minus evaluation” is to set a flag when the member evaluates that the value parameter value of the product is negative. The “access time” is used to enter the time of accessing the detailed information on the product, and the “access count” is entered the number of times the detailed information on the product is accessed.

Each personal data in the personal data file 22 is created at the time of member registration. At that time, only the member number 41, age 42, and gender 43 are input, and the plus product 44 and the minus product 45 are entered. , The unresponsive product 46 and the proposed product history 47 are blank.

FIG. 9 is a flowchart showing the flow of processing of the personal data creation processing section 33 of the product selection server 2. The personal data creation processing unit 33 first retrieves the next personal data from the personal data file 22 (step 7).
1). Next, referring to the proposed product history 47 of the personal data, the proposed product is a plus product 44 and a minus product 45.
Alternatively, it is checked whether the product is included in the unreacted product 46.
If it is not included, it is checked whether or not the product number of each proposed product exists in the individual access data 25, and a proposed product that does not exist is extracted (step 72). If there is no proposed product in the individual access data, the member has not taken any action to determine the value parameter value in response to the proposal.
The item number is registered in the item of the unresponsive item 46 of the corresponding member (step 73).

Next, the access data of the next product is referred to from the top of the member information in the individual access data 25 (step 74). 1 in the item of “number of purchases” of the product
If the above numerical values are entered, it means that there is a purchase record (step 75 YES). Since it can be determined that the merchandise has a purchase result, the merchandise number is registered in the item of the plus merchandise 44 of the member in the personal data file 22 (step 80). If there is no purchase record (step 75 NO), it is checked whether a flag is set in the item of “plus evaluation” of the product (step 76). If the flag is set, the member has evaluated that the value parameter value is positive,
The product number is registered in the item of the plus product 44 of the member in the personal data file 22 (step 80). If there is no evaluation of “plus evaluation” by the member (step 76
NO), it is checked whether a flag is set in the item of "minus evaluation" of the product (step 77). If the flag is set, the member himself has evaluated that the value parameter value is negative, and the item number is registered in the item of the negative item 45 of the member in the personal data file 22 (step 79). If there is no “plus evaluation” or “minus evaluation” by the member, the “access time” and “access count” of the product are referred to. By calculating the access time / access count, the amount of information that the member has acquired in one access is estimated (step 7).
8). If the average access time is equal to or longer than a predetermined value sufficient to acquire the product information, it indicates that the interest of the member, that is, the necessity is high, the value parameter value is considered to be positive, and the member data in the personal data file 22 is considered to be positive. Plus product 4
The product number is registered in the item No. 4 (step 80). For example, if the access time / access count is 10 seconds or more, it is determined that a certain amount of product information can be obtained, and a positive evaluation is made. Conversely, if the average access time is a short time that is less than the predetermined value at which it is considered impossible to acquire the product information, it indicates that the interest of the member, that is, the necessity is low, the value parameter is considered negative, and the personal data file 22 The item number is registered in the item of the negative item item 45 of the member. For example, if the access time / access count is within 10 seconds, it is determined that almost no information can be obtained, and a negative evaluation is made (step 79). The threshold of the access time / access count for the plus product and the threshold of the access time / access count for the minus product may be set separately.

If the process has not been completed for all products for the member (NO in step 81), the process returns to step 74, and the above process is repeated for the next product.
When the above processing is completed for all products (step 81 YES), the personal data file 22 is updated with the personal data of the member (step 82). If the processing has not been completed for all members (step 83
NO), returning to step 71 and repeating the above processing for the next member. When the above processing has been completed for all members (step 83 YES), the processing of the personal data creation processing unit 33 ends.

FIG. 10 shows an example of data constituting the access log 28 stored in the mall server 3. The access log 28 records the “member number” of the member who accessed in chronological order and the place where the member accessed, based on the “access time”. When accepting product purchases from members, the buttons shown in FIG.
For example, when a member directly receives an evaluation of a product, such as selecting the item, or receives access to detailed information of a product, the product number is recorded in a log. For other access actions, the action history is recorded without specifying the product number.

The order of the access log 28 is rearranged in the order of “access time” to “member number” to obtain the access log 26 in the ID order. FIG. 11 shows an example of data constituting the access log 26 in the ID order. . ID order is the order of member numbers. By rearranging the data based on the “member number”, the “access time” and the “access place” arranged in chronological order for each member are stored in the ID-ordered access log 26. By this rearrangement, the individual member's action history can be understood, and by taking the difference between the previous and next “access time”, the “access time” for accessing the previous access location becomes clear. In the example of FIG.
Only the access time relating to individual product information requiring access time information is calculated and stored. The access log processing unit 35 creates the access log 26 in ID order by performing the above processing.

FIG. 12 is a flowchart showing the processing flow of the individual access data calculation processing section 36. ID
The access data of the next member number is retrieved from the sequential access log 26 (step 91), and the member number is entered in a work area on the memory for storing individual access data. Next, the access data of the next line is referred to for the member (step 92). If the access data in the row does not relate to an individual product such as product purchase, access to detailed product information, and direct evaluation of the product by the member, the process goes to step 104 because there is no product number (YES in step 93). If there is a new product number in the work area when there is a product number (step 94YE
S) Write a new product number in the work area (step 95). If the content of the access data indicates purchase of a product (YES in step 96), 1 is added to the item of the number of purchases corresponding to the corresponding product number in the work area (step 97). Here, it is assumed that a log is generated individually when a plurality of products are purchased, so the purchase number is always added by one. If the content of the access data indicates access to the detailed information of the product (step 98)
(YES), the numerical value of the access time recorded in the ID-order access log 26 is added to the numerical value of the access time in the work area, and 1 is added to the number of accesses (step 9).
9). If the content of the access data indicates a positive evaluation of the product by the user (step 1
(YES), 1 is written as a flag in the item of plus evaluation on the work area (step 101). If the content of the access data indicates a negative evaluation of the product by the user (step 102YE
S), 1 is written as a flag in the item of negative evaluation on the work area (step 103). Next, if it is not the end of the access data (NO in step 104), the process returns to step 92 and repeats the above processing. If it is the end of the access data (YES at step 104) and the processing has not been completed for all members (NO at step 105), the process returns to step 91 and repeats the above processing. When the processing is completed for all members (step 105 YES), the access data in the work area is stored in the storage device 5 as the individual access data 25 (step 106).

FIG. 13 is an example showing the data structure of the product master 27. The product master 27 stores the attributes of all the products handled, including both Internet shopping sites and physical stores, and information on sales and access. When registering a new product,
A product number, a product attribute such as a size determined for each product, and a registration date indicating the date registered in the product master are entered, and 0 is entered in other items. The registration of a new product is performed from the input device 10 as needed, and the update of other data is updated in accordance with the update of the proposed product list 23. The items related to Internet shopping are updated based on the data of the Internet product-specific results 29. The “Internet access time”, “Internet access count”, and “Internet sales” in the product master 27 are the contents of the Internet product-specific results 29 sent to the product selection server 2.

FIG. 14 shows the results 29 by Internet product.
1 is an example showing the structure of the data. The Internet product-specific results 29 are obtained by totaling and recording the access logs 28 collected by the mall server 3 for each product at regular intervals. "Detailed information access time", "Internet sales", "Access frequency" for each "product number"
Are the total data after the product is registered in the product master 27. "Store sales" in the product master 27
Is the data calculated by the merchandise sales information calculation processing unit 31 by referring to the merchandise sales information file 21 stored in the storage device 4 of the store server 1 through the POS 7 of the actual store, and transmitted to the merchandise selection server 2. Data.

FIG. 15 shows a suggested product list creation processing section 32.
3 is a flowchart showing the flow of the processing of FIG. The proposed product list creation processing unit 32 extracts the data of the next product from the product master 27 (step 111),
Copy the data of "Internet access time", "Internet access count", "Internet sales" and "store sales" onto the work area (step 11).
2). Next, the number of days from the registration date to the processing date described in the product master 27 is calculated, and the result is entered in the item of the number of registration days of the corresponding product on the work area (step 113). Next, the number of completed “access times” “access time” “internet sales” “real store sales”
Then, the total points are calculated by weighting the values of the “registration days” and the “registration days” respectively, and are entered in the “comprehensive evaluation” (step 114).

As an example of selecting a product considered to be effective from the range of information without the user's consciousness, in order to take out a new product, the number of registration days * (− 30) is used.
It is conceivable that the longer the number of registration days, the lower the overall evaluation. In order to retrieve a product with a long average access time, "access time" / "access count" *
An average access time, such as 10, can be determined. Here, it was used as a weight (-30),
A number such as 10 is arbitrarily specified. Since the values of the indices are different from each other, the weight is increased when the newness of the information is given the highest priority. It can be determined that the higher the total evaluation point calculated here is, the higher the general evaluation is.

If the processing has not been completed for all the products (NO in step 115), the process returns to step 111 and the above processing is repeated. When the above processing is completed for all products (YES in step 115), the product information is rearranged in descending order of the comprehensive evaluation (step 116). Through the above processing, a suggested product list in priority order to be selected for proposal to the user among all products handled is created. A certain number of product data is left from the top of this list, and product data below that is deleted (step 117). Since a proposal list is created from a huge number of products, it is easier to process if a certain number of products are limited. For example, the top 50 products are left in the list. The new list created in this way is stored on the storage device 5 as the suggested product list 23 (step 118).

FIG. 16 shows an example of data constituting the related inappropriate product calculation file 140. The related inappropriate product calculation file 140 indicates, for each product indicated by the product number 141, the number of members 142 whose value parameter value for the product is negative, and the other products 143 among the members. This is a file in which the number of people 144 having a negative value parameter value is totaled and recorded.

FIG. 17 is a flowchart showing the flow of the processing of the minus merchandise group creation processing section 34. First, the minus product group creation processing unit 34 sets the product master 2
Then, the product numbers of all the products handled from 7 are copied to the minus product group 24 in the memory, and are set as the product number 141 (step 121). Next, the minus member 45 is referred to for the next member of the personal data file 22 (step 122). If there is no remaining unselected commodity in the minus commodity 45 (step 123 NO), step 126
Go to If there are remaining products (step 123YE
S), one remaining product is selected, a corresponding product number 141 is found in the related inappropriate product calculation file 140, and 1 is added to the number 142 (step 124). Next, for all products other than the product included in the item of the negative product 45 of the member, the related inappropriate product calculation file 14
If there is a product number corresponding to 0 other products 143, 1 is added to the item of the number of people 144. If there is no corresponding product number, a new line is added to the item of the other product 143, the product number is entered, 1 is entered in the number of persons 144 (step 125), and the process returns to step 123.

If the above processing is not completed for all members (step 126 NO), the process returns to step 122 and repeats the above processing. When the processing is completed for all members (step 126 YES), data of a minus commodity group is extracted for the next commodity number 141 in the related inappropriate commodity calculation file 140 (step 12).
7). In order to obtain the ratio of the number 144 of members having a negative value parameter value of the product indicated by the other product 143 to the number 142 of members having a negative value parameter value of the product indicated by the product number 141, the number of members 144 /
The number of persons 142) is calculated (step 128). It is considered that the closer this value is to 1, the higher the relevance between the products. For example, a case where (number of persons 144 / number of persons 142) ≧ 0.5 has high relevance. If the product indicated by the product number 141 is highly relevant (YES in step 129), the product number group of the other products 143 is entered in the “related inappropriate product” item of the corresponding “product number” of the minus product group 24. (Step 130). If the relevancy is not high (step 129 NO), the procedure goes to step 131. Next, if the processing has not been completed for the minus commodity group relating to all the commodity numbers 141 in the related inappropriate commodity calculation file 140 (131 NO), the processing returns to step 127 and repeats the above processing. If the process is completed for all products (131Y
ES), the process of the minus merchandise group creation processing unit 34 ends.

The above embodiment is an example, and other modified configurations are possible. For example, regarding the entire configuration of the system shown in FIG. 1, an online network other than the Internet may be used, or the product proposal server 2
And the mall server 3 may be integrated into the same server. Further, when the actual store is not sold, only the data obtained from the Internet shopping may be used to select a suggested product. Further, there may be a plurality of store servers depending on the type of the actual store.

Also, for example, when the processing for narrowing down the products to be proposed to the user from the proposed product list 23 is completed, all the products to be proposed may be deleted and not left. In response to this, a part of the history may be cleared and the products to be proposed again may be narrowed down, such as a method of not making a product proposal or a non-reactive product 46 of the personal data file 22.

In the above-described embodiment, in the Internet shopping site of a company that also manages a physical store, the value of the merchandise information that the site intends to provide by using the method of selecting unread information is generally valued. Prioritize products that are likely to be present, and select products from which a member has not established evaluation parameter values directly or by guessing from action history from a list of products with higher rankings, and further evaluate the members. He explained how to anticipate and remove information that would result in a negative, and propose a higher-ranking product.

Although the above embodiment has been described with respect to a system for providing product information to a member user, the product here is not limited to articles but includes a wide range of pay information and various services. For example, the present invention can be applied to a system for distributing news to a contracted user for a fee. In a news distribution system, news articles are products. In this system, there is no actual store, and the store server 1 is unnecessary. Users are offered a summary of the news article for free,
A news article for which detailed information is requested by a user may be charged, and may be counted as “Internet sales”. In addition, the present invention may be applied to a system that provides books, music contents, video contents, and the like with a membership system. In such a system, as a piece of information about a proposed product, an outline of contents such as books, music contents, and video contents is suggested to a user via a display screen as shown in FIG. It is sufficient to obtain book / dislike / purchase value parameter values and provide books, music contents, video contents, and the like when the user instructs purchase. The present invention can also be applied to a service for providing information on a travel guide system that provides a member with a package tour as a product.

[0056]

As described above, according to the present invention, products that have already been proposed to the user are excluded from the list of products to be proposed, and products that are highly likely to be evaluated as negative are excluded. To be able to select and provide the user with product information that is presumed to be valuable because it is product information that cannot be obtained on his own without being conscious of it, thereby discovering potential needs of the user. Can be. In addition, the time required for the user to browse the product information for the proposed product information is measured as the access time, and the evaluation of the product whose access time is extremely short is determined as the negative evaluation. As a result, the products that the user unconsciously evaluated negatively and the related inappropriate products are deleted from the next proposed product list, so that the user's unconscious behavior can be reflected in the evaluation and the user's workload is reduced. Can be reduced.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a system related to Internet shopping according to an embodiment.

FIG. 2 is a diagram showing an example of data constituting a personal data file 22.

FIG. 3 is a diagram showing an example of data constituting a suggested product list 23;

FIG. 4 is a diagram showing an example of data constituting a minus product group 24;

FIG. 5A is a flowchart illustrating a processing flow of a selected product transmission processing unit 38 according to the embodiment.

FIG. 5B is a flowchart illustrating the flow of a process performed by an individual proposed product calculation processing unit 37 according to the embodiment.

FIG. 6 is a diagram illustrating an example of data of a suggested product list for individuals.

FIG. 7 is a diagram showing a display example of proposed product information.

FIG. 8 is a diagram showing an example of data constituting individual access data 25;

FIG. 9 is a flowchart illustrating a processing flow of a personal data creation processing unit 33 according to the embodiment.

FIG. 10 is a diagram showing an example of data constituting an access log 28;

FIG. 11 is a diagram showing an example of data constituting an access log 26 in ID order.

FIG. 12 is a flowchart illustrating a flow of processing of an individual access data calculation processing unit according to the embodiment.

FIG. 13 is a diagram showing an example of data constituting a product master 27;

FIG. 14 is a diagram showing an example of data constituting an Internet product-specific result 29;

FIG. 15 is a flowchart illustrating a flow of a process performed by a suggested product list creation processing unit 32 according to the embodiment.

FIG. 16 is a diagram showing an example of data constituting a related inappropriate product calculation file 140.

FIG. 17 is a flowchart illustrating a processing flow of a minus product group creation processing unit according to the embodiment.

[Explanation of symbols]

1 ... store server, 2 ... product selection server, 3
... Mall server, 22 ... Personal data file, 23 ... Proposed product list, 24 ... Minus product group, 33 ... Personal data creation processing unit, 37
..Individual proposed product calculation processing section, 44: plus product, 45: minus product

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Megumi Yonezawa 890 Kashimada, Saiwai-ku, Kawasaki-shi, Kanagawa F-term in the System Development Division, Hitachi, Ltd. 5B049 BB11 BB26 BB61 CC02 CC08 EE01 EE05 EE59 GG04 GG07 5B075 KK03 KK07 KK13 KK33 KK38 ND03 ND08 ND14 ND23 ND36 PP03 PP13 PP30 PQ00 PR03 QP10 UU40 5B089 GA02 GA11 GA21 JA04 JA12

Claims (7)

[Claims] 1. A system for providing product information having a product selection server and an information transmission server, wherein the product selection server corresponds to an identifier of each user and displays a product that has been proposed to that user. First storage means for storing an identifier;
A second storage unit for storing a list of product candidates providing preset product information; and a user product identifier received from the information transmission server and referred to the first and second storage units. Processing means for transmitting the identifiers of the remaining proposed products to the information transmission server except for the products already proposed to the user from the list, the information transmission server comprising: storage means for storing product information of the proposed products; Processing means for receiving an access from the client device of the user and transmitting the identifier of the user to the product selection server; receiving the identifier of the proposed product from the product selection server; Processing means for extracting information and transmitting the information to the client device of the user. 2. A method according to claim 1, further comprising: storing, in correspondence with the identifier of each user, an identifier of a product proposed to the user and an identifier of a product evaluated negatively by the user among the proposed products. Storage means, and second storage means for storing a list of product candidates for providing preset product information;
Third storage means for arranging identifiers of other related products for which a predetermined ratio or more of users have a negative rating corresponding to the identifiers of the products that the user has evaluated negatively, and product proposals for specific users In providing the product information for the specific user, a product identifier that has been proposed to the specific user is extracted with reference to the first storage unit, and the proposed product identifier is extracted from the proposed product list in the second storage unit. Deleting the identifier of the related product of the product evaluated by the specific user as negative by referring to the third storage means from the suggested product list, and replacing the identifier of the remaining proposed product with the specific user. And a processing unit for setting the product information as a target product for providing product information. 3. A program stored on a computer-readable storage medium, the program including the following functions: (a) The user has given a positive evaluation in accordance with the identifier of each user. A function of storing the identifier of the product, the identifier of the product evaluated negatively by the user, and the identifier of the other proposed product in the first storage means; (b) the function of storing the product information that provides preset product information; A function of storing the list of candidates in the second storage means; (c) identifiers of other related products for which a predetermined ratio or more of users have received a negative rating corresponding to the identifiers of the products that the user has given a negative rating (D) when providing product information to a specific user for product proposal to a specific user, referring to the first storage device, a product already proposed to the specific user. Take an identifier (E) a function to delete a proposed product identifier from the proposed product list on the second storage means; (f)
A function of referring to the third storage means to delete an identifier of a related product of a product evaluated negatively by the specific user from the proposed product list, and (g) replacing the identifier of the remaining proposed product with the specific user. Function to be a target product for providing product information to customers. 4. The product evaluated positively by the user includes a product whose average access time is equal to or longer than a predetermined value, in addition to a product directly specified by the user for positive evaluation. A storage medium for storing the program according to Item 3. 5. The product evaluated negatively by the user includes a product whose average access time is less than a predetermined value, in addition to a product directly specified by the user to perform negative evaluation. A storage medium for storing the program according to 3. 6. Select a product to be proposed for a specific user,
A method for providing merchandise information related to a system having a server for transmitting merchandise information of a selected merchandise and a client for receiving proposed merchandise information, comprising: (a) at least one of the proposed merchandise set by the server; Transmitting the product information to the client; (b) displaying the product information received by the client on a display device;
(C) when the user instructs one of a positive evaluation and a negative evaluation on the displayed product information, the client transmits a value parameter corresponding to the product identifier to the server, d) In accordance with the identifier of each user by the server, the identifier of the product that the user gave a positive evaluation, the identifier of the product that the user gave a negative evaluation, and the identifiers of other proposed products are described. (E) When providing product information to a specific user for product proposal, the product identifier that has been proposed to the specific user is retrieved by referring to the first storage device. (F) deleting a proposed product identifier from a proposed product list in a second storage means for storing a list of product candidates providing preset product information; The specific user is referred to the third storage means for storing the identifiers of other related products evaluated by the users of a predetermined ratio or more corresponding to the identifiers of the products evaluated negative, and Remove the identifier of the related product of the evaluated product from the proposed product list, and (h)
The merchandise information about the remaining proposed merchandise is transmitted to the client. 7. The server measures the time during which the client has access to the product information of the proposed product, and regards a product whose access time is less than a predetermined value as a product evaluated negatively by the user. The method for providing product information according to claim 6, wherein