JP2001229268A - Online shopping system and commodity proposal method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the intervention of man and to automatically generate a rule from accumulated information at the time of setting the rule applied when a commodity to be proposed to customer is decided in an online shopping system. SOLUTION: The rule information generation part 36 of a server 3 develops a sum-up table by individual attribute group types with the combination of plural customer attribute values as an entry on a memory based on the customer attribute of a customer attribute information DB 32, counts the customer frequency of the entry of the sum-up table, which is applied to the combination of the attribute values of the customer purchasing the commodity by referring to a purchase history information DB 33 and customer attribute information DB 32 on the specified commodity and extracts the combination of the attribute values, whose customer frequency is conspicuous. A rule information is generated from the combination of the extracted attribute values and the identifier of the commodity and it is stored in a rule information DB 39.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to electronic commerce, and more particularly to an online shopping system for proposing products to customers.

[0002]

2. Description of the Related Art With the widespread use of the Internet, online shopping for providing product information to customers using the WWW technology on the Internet, proposing products, and accepting product orders has been realized, and has attracted attention as a business. ing. For example, there is a technique described in JP-A-11-85846.

[0003]

In an online shopping system, when a product is proposed to a customer, a product to be proposed to the customer is determined by applying a preset rule to the attribute information of the customer. According to the prior art, when setting this rule, the customer attribute information, purchase history information, system access information, etc. are input to the analysis tool, and the rule information is manually created based on the analysis result obtained by human judgment. Was. Therefore, there was a risk that the subjectivity of the person in charge would enter.

An object of the present invention is to realize a shopping system and a product proposal method for automatically generating rule information based on the above information and proposing a product to a customer based on the rule information. is there.

[0005]

SUMMARY OF THE INVENTION According to the present invention, an aggregation table for each attribute group having a combination of a plurality of customer attribute values as an entry is developed on a storage device, and an attribute of a customer who has purchased a specific product in the past is provided. The customer frequency of the entry corresponding to the value combination is counted, a combination of attribute values having a remarkable customer frequency is extracted, and the rule information that concludes the product without the combination of the extracted attribute values as a formation condition, A merchandise proposal method and an online shopping system that match a combination of an attribute value of an accessed customer with a condition for establishing rule information and provide the accessed customer with the merchandise that satisfies the condition as a proposed merchandise. Features.

[0006]

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

FIG. 1 is a configuration diagram of an online shopping system for electronically providing product information. As shown in the figure, the system includes a client 1, which is a computer on the customer side, a server 3, which is a computer on the seller side for providing product information, and a network 2 connecting the client 1 and the server 3. The storage devices of the server 3 include a customer attribute information DB (database) 32, a purchase history information DB3
3. Store the product information DB 38 and the rule information DB 39. The memory of the server 3 stores the programs of the request receiving unit 35, the product information generating unit 40, and the rule information generating unit 36.

[0008] The request receiving unit 35 receives a request from the client 1 and authenticates the customer. The product information generation unit 40 determines a product expected to be suitable for the customer from the customer attribute information obtained from the customer attribute information DB 32 and the rule information stored in the rule information DB 39, and determines the product information D
The information about the product is extracted from B38, a product information providing page is generated, and transmitted to the client 1. When a product is purchased, the purchase history is stored in a purchase history information DB3.
Record in 3. The rule information generating unit 36 outputs the customer attribute information D
By referring to B32, the purchase history information DB 33 and the product information DB 38, rule information is generated from the purchase tendency based on the attribute of the customer, and stored in the rule information DB 39.

[0009] The electronic store targeted by the shopping system may be a single store, or a plurality of electronic stores may constitute a shopping mall.

FIG. 2 is a diagram showing the data structure of the customer attribute information DB 32. Each record of the customer attribute information DB 32 stores information such as member ID, name, region, gender, birthday, occupation, hobby, and number of purchases for each member.

FIG. 3 is a diagram showing the data structure of the purchase history information DB 33. The purchase history information DB 33 is a purchase history of a product, and each record includes a session ID, a member ID, an order date and time, a purchaser address, a purchase store name, a product ID, a purchase quantity, a purchase amount, and the like for each purchase of a product by a member. Record the information. The session ID is an identifier assigned for each session with the client 1.

FIG. 4 is a diagram showing the data structure of the product information DB 38. Each record of the product information DB 38 stores information such as a product ID, a store name, a category, a product name, a provided price, a special price, and a description for each product. The “special price” is a price for a member who has purchased a large number of times or the amount of purchase. The “description” is a description of a product to be placed on a page that provides product information.

FIG. 5 is a diagram showing a data structure of the rule information DB 39. The rule information DB 39 is composed of a plurality of pieces of rule information.
Consists of The rule establishment condition 42 is a condition for providing product information. Conclusion 43 indicates a product to be provided with information when the condition is satisfied.

FIGS. 6A and 6B are flowcharts showing the flow of the processing of the rule information generating unit 36 which is periodically executed. The rule information generation unit 36 inputs one product ID from the product information DB 38 (step 51), and develops an aggregation table for each attribute group on a storage device such as a memory (step 52).

FIG. 7 is a diagram showing an example of an aggregation table for each attribute group. Each record of the customer attribute information DB 32 includes a plurality of attributes related to the customer, and two or more attributes are selected from these attributes to create an attribute group. Customer attributes include natural attributes (age, gender, etc.) and social attributes (region, occupation, hobbies, etc.). The attribute group is based on the natural attribute and adds 1-2 social attributes to it. A combination of such attributes is appropriate. Each entry of the attribute group-based aggregation table is characterized by a combination of each attribute value of the attribute group. The number of entries in each attribute group-based aggregation table is the number of all combinations of each attribute value. “Count” is the frequency of the customer corresponding to the entry, and has an initial value of 0. n is the total number of counts in the attribute group-based aggregation table.

If the end of the search is not reached during the search of the purchase history information DB 33 (NO in step 53), the purchase history information DB 33 is searched by the input product ID, and one member ID included in the corresponding record is extracted. (Step 54). Next, the customer attribute information DB 32 is searched by the extracted member ID, and the corresponding record is read (step 55). Next, based on the attribute value of the read record, 1 is added to the "count" of the corresponding entry of the attribute-based aggregation table (step 56), and step 5
Return to 3.

Steps 53 to 56 are repeated as described above, and when the end of the purchase history information DB 33 is reached (step 53 YES), the process proceeds to FIG. Is calculated (step 57). If n is not equal to or greater than the predetermined number (NO in step 58), it is considered that the number of samples is insufficient, and the processing for the product ID is terminated. If n is equal to or greater than the predetermined number (step 58 YES), the entries of the attribute group-based aggregation table are sorted in descending order of the count (step 59). Next, for all the attribute group-by-attribute tables, entries of a predetermined number or more or a predetermined percentage of n or more, that is, salient attribute value groups (combinations of attribute values) are extracted in ascending order of count (step 60). ).
Next, the attribute values of the extracted attribute value group are logically combined to generate rule information (step 61). That is, one attribute value group is expressed as a combination of the attribute values of the element by AND. When a plurality of attribute value groups are extracted, a rule is generated by combining the attribute value groups by OR. Next, the generated rule information is stored in the rule information DB 39 (step 62), and the process for the product ID ends. In this way, the above processing is performed for each product on the product information DB 38, and the rule information DB 39 is constructed. Note that the determination of at least a predetermined number in step 58 may be bypassed, and if there is an attribute value group whose count is at least a predetermined number in step 60, the attribute value group may be extracted.

For a product having seasonality, an extraction period (current month, a fixed period from the current day, etc.) of purchase history information is input when a product ID is input (step 51). The input method is input via a keyboard or the product information DB 38
May be input as a data item set in the record of. Then, when searching the purchase history information DB 33, a record in which “order date and time” of last year corresponds to this extraction period and has a corresponding product ID is set as a search target. The condition of the extraction period is added to the attribute value group extracted at the time of generating the rule information by AND-linking.

FIG. 8 is a flowchart showing the processing flow of the request receiving unit 35 and the product information generating unit 40 of the client 1 and the server 3. The client 1 requests a login page from the server 3 via the network 2 and displays the login page when it is received (step 7).
1). The request receiving unit 35 receives the request from the client 1 and transmits a login page to the client 1 (Step 72). Client 1 receives the entered member ID
Is transmitted to the server 3 (step 7).
3). The request receiving unit 35 receives the customer information, searches the customer attribute information DB 32 based on the member ID, acquires the corresponding customer attribute information (step 74), and authenticates the customer (step 75). If the customer is not registered as a member, the client 1 performs a member registration procedure with the server 3, and the server 3 registers the received customer attribute information in the customer attribute information DB 32.

Next, the merchandise information generating unit 40 stores a rule information DB
The rule information is read from 39 (step 76), the acquired customer attribute information is checked against the rule establishment condition 42 of each rule information, and if both match, the proposed product is determined based on the conclusion 43 (step 77). . Next, the product information DB 38 is searched according to the determined product ID, and the corresponding product information is read (step 78). Next, the product information generation unit 40 creates a product information providing page from the read product information and transmits it to the client 1 (step 79). The client 1 receives the product information providing page and displays it on the display device (Step 80). FIG.
Shows an example of the product information providing page. When the order information is input and the order is instructed, the client 1 transmits the order information to the server 3 (step 81). Server 3
Receives the order information, performs necessary order processing, and records the purchase history information in the purchase history information DB 33 (step 82).

[0021]

As described above, according to the present invention, since the rule information is automatically created based on the customer attribute information and the purchase history information, the trouble of the person in charge is reduced, and the rule by the person in charge is subjective. Creation can be eliminated. In addition, it is possible to provide product information that is well suited to customers.

[Brief description of the drawings]

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

FIG. 2 is a diagram illustrating a data configuration of a customer attribute information DB 32 according to the embodiment.

FIG. 3 is a diagram illustrating a data configuration of a purchase history information DB 33 according to the embodiment.

FIG. 4 is a diagram illustrating a data configuration of a product information DB according to the embodiment.

FIG. 5 is a diagram illustrating a data configuration of a rule information DB 39 according to the embodiment.

FIG. 6A is a flowchart illustrating a flow of processing of a rule information generation unit 36 according to the embodiment.

FIG. 6B is a flowchart (continued) showing the flow of the process of the rule information generator 36 of the embodiment.

FIG. 7 is a diagram illustrating an example of an attribute group-based aggregation table.

FIG. 8 is a flowchart showing a processing flow of a client 1, a request receiving unit 35, and a product information generating unit 40 of the embodiment.

FIG. 9 is a diagram showing a display example of a product information providing page.

[Explanation of symbols]

1: client, 3: server, 32: customer attribute information D
B, 33: purchase history information DB, 35: request receiving unit, 3
6: Rule information generation unit, 38: Product information DB, 39: Rule information DB, 40: Product information generation unit

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference)

Claims (4)

[Claims] In a product proposal method of an online shopping system, an aggregation table for each attribute group having a combination of a plurality of customer attribute values as an entry is developed on a storage device, and a customer who has purchased a specific product in the past is stored. The customer frequency of the entry corresponding to the combination of the attribute values is counted, the combination of the attribute values having the remarkable customer frequency is extracted, and the rule information that concludes the product without the extracted attribute value combination as the establishment condition is defined. It is characterized in that the generated information is compared with a combination of the attribute values of the accessed customers and the conditions for establishing the rule information, and the products satisfying the conditions are provided as proposed products to the accessed customers with the product information. Product proposal method. 2. An on-line shopping system in which a customer computer and a seller computer are connected via a network, wherein the seller computer includes an aggregation table for each attribute group having a combination of a plurality of customer attribute values as an entry. On the storage device, and counts the customer frequency of the entry corresponding to the combination of the attribute values of the customers who have purchased the specific product in the past, and extracts the combination of the attribute values with the remarkable customer frequency. Means for generating a rule information that concludes the product with no condition for establishing a combination of the extracted attribute values, a combination of the attribute value of the customer and the rule in response to access from the customer-side computer. Means for comparing the information satisfying the condition with the information, and transmitting the product information to the customer-side computer as a product for which the condition is satisfied as a proposed product. Online shopping system, characterized in that. 3. A customer attribute information database for storing a plurality of attribute values for each customer, a purchase history information database for storing correspondence between customer identifiers and product identifiers for customers who have purchased products in the past, Means for developing, on a storage device, an aggregation table for each attribute group having a combination of a plurality of customer attribute values as an entry, and acquiring the identifier of a customer who has purchased the product by searching the purchase history information database for a specific product Means, means for searching the customer attribute information database by the obtained customer identifier to obtain the corresponding customer attribute value, and the customer of the entry of the attribute group-based totaling table corresponding to the combination of the obtained customer attribute value. Means for counting the frequency, and if the sum of the customer frequencies in the attribute group summary table is a predetermined number or more, the customer frequency Means for extracting a combination of attribute values, and means for logically combining the attribute values included in the extracted combination of attribute values to generate rule information that satisfies the condition for establishment and the conclusion of the product identifier. And the seller's computer. 4. A means for searching the customer attribute information database for a specific customer to obtain an attribute value of the customer, and comparing a combination of the obtained attribute values with a condition for establishing the rule information to establish a condition. 4. The seller-side computer according to claim 3, further comprising: means for acquiring a product identifier; means for creating product information on the acquired product; and means for providing the created product information to the customer.