JP2014115951A - Apparatus, program and method for optimizing attribute information, apparatus, program and method for selecting recommendation object - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an apparatus for optimizing attribute information, etc. capable of optimizing attribute information on customers and attribute information on commodities, etc., respectively in selection of the commodities, etc. to be recommended to the customers, and to further provide an apparatus for selecting recommendation objects for selecting object commodities to be recommended to the customers, etc. by using the attribute information on the customers and the attribute information on the commodities, etc. which are optimized by the apparatus for optimizing the attribute information, etc.SOLUTION: An attribute item common to each of customers and commodities to be recommendation objects is set, and scores showing degrees of attributes are recorded in a customer attribute master and a commodity attribute master for every customer, and every commodity. The attribute information on the customers and the attribute information on the commodities are optimized, respectively through purchase behaviors by reflecting scores showing the degrees of attributes of the commodities on scores showing the degrees of attributes of the customers who purchase the commodities and also reflecting the degrees of attributes of the customers who purchase the commodities on the scores showing the degrees of attributes of the commodities for the common attribute item when the customers purchase the commodities.

Description

  The present invention relates to an attribute information optimizing device, an attribute information optimizing program, and attribute information for optimizing customer attribute information and product or service attribute information used for selecting a product or service recommended to a customer. Using the customer's attribute information and product / service attribute information optimized by these devices, programs or methods to select products or services to be recommended to the customer, or The present invention relates to a recommendation target selection device, a recommendation target selection program, and a recommendation target selection method for selecting a customer to be recommended for a product or service.

  In order to promote the sale of products and the use of services, so-called providing coupons at stores, displaying recommended products on e-commerce sites, etc. Recommendations are widely practiced. In such a recommendation for a customer, it is common to collect purchasing information such as products for each customer, analyze customer preferences from a purchase history, and select products to be recommended.

  As one of the recommended methods, based on the assumption that “customers with similar preferences also have similar purchase histories”, customers with similar purchase histories are extracted by correlation, and customers with similar purchase histories purchase. The customer recommends a product that has not yet been purchased. (In the example of FIG. 1 showing the purchase history for each customer, the customer D is the product D, and the customer V is the product E.) A method called collaborative filtering).

  As a problem that arises when adopting collaborative filtering, it is recommended to select new products that are not included in the purchase history, that is, the purchase results are not yet accumulated, because the products to be recommended are selected by comparison with the purchase history of others. It is mentioned that it cannot be the target of. In addition, in order to extract customers with similar purchase histories, there is a tendency that a large number of customer purchase histories need to be accumulated.

  As another typical recommendation method that can deal with these issues, we extract the characteristics of products that customers presumably prefer from the products purchased by customers in the past, and select products with similar characteristics from the product database. There can be mentioned a technique called a need estimation type that is selected and recommended (in the example of FIG. 2 showing a purchase history of a certain customer, a snack of P company is recommended to the customer). If it is a needs estimation type, even if it is a new product with no purchase record, it is possible to make a recommendation target by setting information on the characteristics of the product.

  With regard to how to extract products that meet customer preferences in the needs estimation type, for example, in order to recommend a Web page that matches the user's preference, the evaluation points of the attribute items in the content (Web page) In addition, the user profile is generated from the evaluation points of the attribute items of the browsed content, the content corresponding to the attribute preferred by the user is selected and recommended (see Patent Document 1), and the user's preference is met. Understanding content attributes and user-preferred attributes as vectors on the same coordinates so that content can be searched automatically, and selecting content that suits user preferences from the approximate calculation of the user position on the coordinates and the position of the content In addition, an invention in which the position of the user varies according to the content selection result (see Patent Document 2) As described above, the common attribute items are set for the customer and the product and scored, and the product that matches the taste of the customer is extracted by the approximate calculation of these scores, and the attribute of the customer is formed by the attribute of the selected product. A method is disclosed.

JP 2002-163508 A JP 2008-250957 A

  According to the method of setting the attribute items common to the customer and the product as illustrated in Patent Documents 1 and 2 and scoring, and extracting the product that matches the customer's preference by the approximate calculation of these scores, the attribute of the product By setting a score for each item, it is possible to add new products to the target of recommendation, and as customers purchase more products, customer preferences will be reflected more precisely. It would be possible to make recommendations. However, it is assumed that the following problems also occur when making a recommendation by this method.

  In the inventions of Patent Documents 1 and 2, the score (points and coordinate values) of items indicating user attributes varies depending on the selected content, while the initially set score is maintained for the attribute items of the content itself. It has become. Therefore, if the content attributes are not scored properly at the initial stage, the content will not be selected properly, and if the user selects such content, the user attributes will be inappropriately affected. End up. In other words, if this is used to recommend products to customers, if the product attributes are not properly scored at the initial stage, the product will not be selected properly and the customer will purchase such products. This will have an inappropriate effect on customer attributes.

  In order to deal with such problems, for example, for each product, the attributes of the customer who purchased the product are collected, and the product preferred by men and the product preferred by young people in their 20s. A method of accumulating product attributes based on actual results is conceivable. For example, there are customers who prefer products for women even for men, and those who prefer products for middle-aged people even in their twenties. There is a possibility that attributes of customers who like the product cannot be sufficiently reflected only by attributes classified in a formal manner such as a fee.

  The present invention has been made to cope with such problems, and in addition to generating customer attribute information from product or service attribute information in applications such as selection of a product or service recommended to a customer. An object of the present invention is to provide an attribute information optimization device, an attribute information optimization program, and an attribute information optimization method capable of optimizing customer attribute information and product or service attribute information. is there.

  Furthermore, the present invention recommends a customer using the attribute information of the customer and the attribute information of the product or service optimized by the attribute information optimization apparatus, the attribute information optimization program, or the attribute information optimization method. Another object of the present invention is to provide a recommendation target selection device, a recommendation target selection program, and a recommendation target selection method for selecting a target product or service, or for selecting a customer for which a product or service is recommended. To do.

  An invention related to an attribute information optimization device that solves the problem of the present application is an attribute information optimization device for optimizing customer attribute information and product or service attribute information, and the degree of each attribute for a customer attribute Customer attribute information storage means for storing customer attribute information indicated by score for each customer, and product attribute information indicating the degree of each attribute for the attribute of the product or service for each product or service Product attribute information storage means, and at least a part of the customer attribute in the customer attribute information and the product or service attribute in the product attribute information, a common attribute is set, and the product purchased by the customer Or accept purchase information that identifies the customer and the product or service for the service used or the product or service selected by the customer A purchase information receiving means, and a score indicating the degree of attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means, the score stored in the product attribute information storage means From customer attribute information updating means for updating a score indicating the degree of the attribute included in the product attribute information of the product or service specified from the purchase information, and the purchase information stored in the product attribute information storage means The degree of the attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage unit, a score indicating the degree of the attribute included in the product attribute information of the specified product or service Product attribute information updating means for updating the product attribute information to reflect the score indicating the attribute information optimization device.

  According to the present invention configured as described above, for a purchase or selection of a product by a customer, a score indicating the degree of the attribute of the product purchased by the customer is reflected in a score indicating the degree of the attribute of the customer In addition, the score indicating the degree of the attribute of the product, etc. is reflected in the score indicating the degree of the attribute of the product, etc. Since the score indicating the degree of satisfaction is adjusted through mutual influence through purchase, etc., it is possible to optimize customer attribute information and product or service attribute information, triggered by customer purchase, etc. Become.

  In the present invention, the product attribute information update unit uses the score indicating the degree of the attribute included in the customer attribute information of the customer updated by the customer attribute information update unit, and the attribute included in the product attribute information. It is also possible to update the score indicating the degree of the above.

  In the present invention, the update of the customer attribute information and the update of the product attribute information may both be performed by updating the attribute information using the pre-update score, and after updating one of them, the other is updated. It is also possible to update the attribute information using the score, but with this configuration, the customer attribute information updated by the purchase information is reflected in the update of the product attribute information. It is suitable when it is considered that the conversion precedes customer attribute information optimization.

  In the present invention, the customer attribute information update unit is included in the customer attribute information using a score indicating a degree of attribute included in the product attribute information of the product or service updated by the product attribute information update unit. A score indicating the degree of the attribute may be updated.

  As described above, in the present invention, the update of the customer attribute information and the update of the product attribute information may both be performed by updating the attribute information using the score before the update, and after updating either one On the other hand, the attribute information may be updated using the updated score, but if configured in this way, the product attribute information updated by the purchase information is reflected in the update of the customer attribute information. This is suitable when it is considered that optimization of attribute information precedes optimization of product attribute information.

  The present invention can also be specified as an attribute information optimization program provided in the attribute information optimization apparatus according to the present invention.

  An attribute information optimization program according to the present invention is an attribute information optimization program for optimizing customer attribute information and product or service attribute information. Customer attribute information storage means for storing customer attribute information for each customer, and product attribute information storage for storing for each product or service the product attribute information indicating the degree of each attribute for the product or service attribute A product purchased or used by a customer in a computer system in which a common attribute is set in at least a part of a customer attribute in the customer attribute information and a product or service attribute in the product attribute information. Purchase information that identifies the customer and the product or service in relation to the service or product or service selected by the customer A purchase information receiving step for receiving, and a score indicating a degree of an attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage unit is stored in the product attribute information storage unit Customer attribute information update step for updating the product attribute information included in the product attribute information of the product or service identified from the purchase information, and the purchase information stored in the product attribute information storage means A score indicating the degree of an attribute included in the product attribute information of the product or service specified from the customer attribute information included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means Attribute information optimization characterized by executing a product attribute information update step that updates a score indicating a degree Is a program.

  In the attribute information optimization program according to the present invention, in the product attribute information update step, a score indicating a degree of an attribute included in the customer attribute information of the customer updated in the customer attribute information update step is stored in the computer system. And a process of updating a score indicating the degree of the attribute included in the product attribute information may be executed.

  In the customer attribute information update step, the attribute information optimization program according to the present invention indicates the degree of attributes included in the product attribute information of the product or service updated in the product attribute information update step in the computer system. A process of updating a score indicating the degree of the attribute included in the customer attribute information may be executed using the score.

  The present invention can also be specified as an attribute information optimization method executed by the attribute information optimization apparatus according to the present invention or the attribute information optimization program according to the present invention.

  An attribute information optimization method according to the present invention is an attribute information optimization method for optimizing customer attribute information and product or service attribute information, and scores the degree of each attribute for a customer attribute. Product attribute information for storing for each product or service the customer attribute information storage means for storing the customer attribute information indicated by the item for each customer and the product attribute information indicating the degree of each attribute for the product or service attribute as a score Information storage means, and a computer system in which a common attribute is set in at least a part of the customer attribute in the customer attribute information and the product attribute or service attribute in the product attribute information. Or accepting purchase information that identifies the customer and the product or service for the service used or the product or service selected by the customer A purchase information receiving step, and the computer system stores a score indicating a degree of an attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means. A customer attribute information update step that updates the score indicating the degree of the attribute included in the product attribute information of the product or service identified from the purchase information stored in the means, and the computer system includes the product attribute A customer identified from the purchase information stored in the customer attribute information storage means with a score indicating the degree of attribute included in the product attribute information of the product or service identified from the purchase information stored in the information storage means Product attribute information update step to be updated by reflecting the score indicating the degree of the attribute included in the customer attribute information of When a method of optimizing the attribute information, characterized in that it comprises a.

  In the attribute information optimization method according to the present invention, in the product attribute information update step, the computer system has a score indicating a degree of an attribute included in the customer attribute information of the customer updated in the customer attribute information update step. Is used to update the score indicating the degree of the attribute included in the product attribute information.

  In the attribute information optimization method according to the present invention, in the customer attribute information update step, the computer system determines the degree of attribute included in the product attribute information of the product or service updated in the product attribute information update step. The score indicating the degree of the attribute included in the customer attribute information may be updated using the score shown.

  Further, the present invention is a method for selecting a product or service to be recommended to a customer using the customer attribute information and the product or service attribute information optimized by the attribute information optimization device according to the present invention. It can also be specified as a recommendation target selection device.

  1st invention regarding the recommendation object selection apparatus which solves the subject of this application is for selecting the goods or service used as the object for recommendation to a customer using the attribute information of the customer and the attribute information of goods or services which were optimized. And a customer attribute information storage means for storing customer attribute information indicating the degree of each attribute as a score for each customer, and for each attribute of a product or service. Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree as a score, and at least one of a customer attribute in the customer attribute information and a product or service attribute in the product attribute information The common attributes are set in the department, and the product purchased by the customer or the service used or selected by the customer With respect to goods or services, purchase information receiving means for receiving purchase information specifying customers and goods or services, and attributes included in customer attribute information of customers specified from the purchase information stored in the customer attribute information storage means Customer attribute information that updates the score indicating the degree of the attribute included in the product attribute information of the product or service identified from the purchase information stored in the product attribute information storage means The update means and a score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage means, the score stored in the customer attribute information storage means Product attribute information to be updated by reflecting a score indicating the degree of the attribute included in the customer attribute information of the customer specified from the purchase information When receiving the designation of the update unit and the customer, at least the product attribute information of the product or service stored in the product attribute information storage unit is included in the customer attribute information of the customer stored in the customer attribute information storage unit A product or service corresponding to product attribute information for which a score indicating a degree of attribute and a score indicating the degree of attribute for a part of the attributes match a predetermined condition is selected as a product or service recommended to the customer. A recommendation target selection device comprising: a recommendation target selection unit.

  According to the first invention configured as described above, it is possible to select a product or service to be recommended to a customer using the optimized attribute information of the customer and the attribute information of the product or service. Become.

  A second invention related to a recommendation target selection device that solves the problem of the present application is to select a customer to be recommended for a product or service using the optimized attribute information of the customer and the attribute information of the product or service. And a customer attribute information storage means for storing customer attribute information indicating the degree of each attribute as a score for each customer, and for each attribute of a product or service. Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree as a score, and at least one of a customer attribute in the customer attribute information and a product or service attribute in the product attribute information The common attributes are set in the department, and the product purchased by the customer or the service used or selected by the customer With respect to goods or services, purchase information receiving means for receiving purchase information specifying customers and goods or services, and attributes included in customer attribute information of customers specified from the purchase information stored in the customer attribute information storage means Customer attribute information that updates the score indicating the degree of the attribute included in the product attribute information of the product or service identified from the purchase information stored in the product attribute information storage means The update means and a score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage means, the score stored in the customer attribute information storage means Product attribute information to be updated by reflecting a score indicating the degree of the attribute included in the customer attribute information of the customer specified from the purchase information Upon receipt of the designation of the product or service with the updating unit, the customer attribute information of the customer stored in the customer attribute information storage unit includes the product attribute information of the product or service stored in the product attribute information storage unit And selecting a customer corresponding to customer attribute information for which a score indicating the attribute level of at least some attributes and a score indicating the attribute level match a predetermined condition as a customer recommending the product or service A recommendation target selection device comprising: a recommendation target selection unit.

  According to the second invention configured as described above, it is possible to select a customer to be recommended for a product or service using the optimized customer attribute information and the product or service attribute information. Become.

  In the invention relating to the recommendation target selection device, the configuration for optimizing the attribute information can be the same as the configuration of the attribute information optimization device according to the present invention.

  Moreover, this invention can also be specified as the recommendation object selection program with which the recommendation object selection apparatus concerning this invention is equipped.

  1st invention regarding the recommendation object selection program which solves the subject of this application is for selecting the goods or service used as the object for recommendation to a customer using the attribute information of the customer and the attribute information of goods or services which were optimized. And a customer attribute information storage means for storing customer attribute information indicating the degree of each attribute as a score for each customer, and for each attribute of a product or service. Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree as a score, and at least one of a customer attribute in the customer attribute information and a product or service attribute in the product attribute information In the department, products purchased by customers or services used on computer systems with common attributes set Alternatively, with respect to the product or service selected by the customer, a purchase information receiving step for receiving purchase information specifying the customer and the product or service, and the customer attribute of the customer specified from the purchase information stored in the customer attribute information storage means A score indicating the degree of the attribute included in the information is reflected on the score indicating the degree of the attribute included in the product attribute information of the product or service identified from the purchase information stored in the product attribute information storage unit. A customer attribute information storage unit that updates a customer attribute information update step and a score indicating a degree of an attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit; Indicating the degree of the attribute included in the customer attribute information of the customer specified from the purchase information stored in The product attribute information update step for updating the product and the designation of the customer is accepted, and the product attribute information of the product or service stored in the product attribute information storage means is stored in the customer attribute information storage means. A score indicating the degree of attribute for at least some of the attributes included in the customer attribute information of the customer, and a product or service corresponding to the product attribute information in which the score indicating the degree of attribute matches a predetermined condition, It is a recommendation object selection program characterized by causing a customer to execute a recommendation object selection step of selecting as a recommended product or service.

  A second invention related to a recommendation target selection program that solves the problem of the present application is to select a customer to be recommended for a product or service using the optimized attribute information of the customer and the attribute information of the product or service. A customer attribute information storage means for storing, for each customer, customer attribute information indicating a degree of each attribute for a customer attribute, and a product or service attribute. Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree as a score, and at least one of a customer attribute in the customer attribute information and a product or service attribute in the product attribute information In some departments, products purchased or used by customers are added to computer systems with common attributes. A purchase information reception step for receiving purchase information in which the customer and the product or service are specified, and the customer information specified from the purchase information stored in the customer attribute information storage means. The score indicating the degree of attribute included in the customer attribute information reflects the score indicating the degree of attribute included in the product attribute information of the product or service identified from the purchase information stored in the product attribute information storage unit The customer attribute information update step, and a score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage means, the customer attribute information The degree of the attribute included in the customer attribute information of the customer specified from the purchase information stored in the storage means When the product attribute information update step for updating the score and the specification of the product or service is received, the customer attribute information stored in the customer attribute information storage means is stored in the product attribute information storage means. Further, a score indicating the degree of attribute for at least some attributes included in the product attribute information of the product or service, and a customer corresponding to the customer attribute information whose score indicating the degree of attribute matches a predetermined condition, A recommendation target selection program for executing a recommendation target selection step of selecting the product or service as a customer to be recommended.

  In the invention relating to the recommendation target selection program, the procedure for optimizing the attribute information can be the same as the procedure for the attribute information optimization program according to the present invention.

  Moreover, this invention can also be specified as the recommendation object selection method performed by the recommendation object selection apparatus concerning this invention, or the recommendation object selection program concerning this invention.

  1st invention regarding the selection method of the recommendation object which solves the subject of this application selects the goods or service used as the object for recommendation to a customer using the attribute information of the customer and the attribute information of goods or services which were optimized. A customer attribute information storing means for storing, for each customer, customer attribute information indicating a degree of each attribute with respect to a customer attribute, and a product or service attribute. Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree of attribute for each product or service, the customer attribute in the customer attribute information and the attribute of the product or service in the product attribute information At least in part, a computer system with common attributes set may be used by a customer to purchase goods or services Or a purchase information receiving step for receiving purchase information in which the customer and the product or service are specified with respect to the product or service selected by the customer, and the computer system is specified from the purchase information stored in the customer attribute information storage means The degree of the attribute included in the product attribute information of the product or service identified from the purchase information stored in the product attribute information storage means is a score indicating the degree of attribute included in the customer attribute information of the customer A customer attribute information update step that updates the score to reflect, and a degree of the attribute included in the product attribute information of the product or service specified by the purchase information stored in the product attribute information storage means by the computer system Is determined from the purchase information stored in the customer attribute information storage means. A product attribute information update step that updates the score indicating the degree of the attribute included in the customer attribute information of the customer, and when the computer system accepts a customer designation, the product attribute information stored in the product attribute information storage means In the product attribute information of the product or service, a score indicating the degree of attribute for at least some attributes included in the customer attribute information of the customer stored in the customer attribute information storage unit, and a score indicating the degree of the attribute A recommendation target selection step of selecting a product or service corresponding to product attribute information that meets a predetermined condition as a product or service to be recommended to the customer.

  A second invention related to a method for selecting a recommendation target that solves the problem of the present application selects a product or service to be recommended to a customer using the optimized attribute information of the customer and the attribute information of the product or service. A customer attribute information storing means for storing, for each customer, customer attribute information indicating a degree of each attribute with respect to a customer attribute, and a product or service attribute. Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree of attribute for each product or service, the customer attribute in the customer attribute information and the attribute of the product or service in the product attribute information At least in part, a computer system with common attributes set may be used by a customer to purchase goods or services Or a purchase information receiving step for receiving purchase information in which the customer and the product or service are specified with respect to the product or service selected by the customer, and the computer system is specified from the purchase information stored in the customer attribute information storage means The degree of the attribute included in the product attribute information of the product or service identified from the purchase information stored in the product attribute information storage means is a score indicating the degree of attribute included in the customer attribute information of the customer A customer attribute information update step that updates the score to reflect, and a degree of the attribute included in the product attribute information of the product or service specified by the purchase information stored in the product attribute information storage means by the computer system Is determined from the purchase information stored in the customer attribute information storage means. Product attribute information update step that updates the score indicating the degree of the attribute included in the customer attribute information of the customer, and when the computer system accepts the specification of the product or service, stores it in the customer attribute information storage means In the customer attribute information of the customer, the score indicating the degree of attribute for at least some attributes included in the product attribute information of the product or service stored in the product attribute information storage unit, and the degree of the attribute A recommendation target selection method comprising: selecting a customer corresponding to customer attribute information whose score matches a predetermined condition as a customer who recommends the product or service.

  In the invention relating to the selection method of the recommendation target, the operation for optimizing the attribute information can be the same as the operation for optimizing the attribute information according to the present invention.

  According to the present invention, when performing a recommendation using a method of setting a common attribute item for a customer and a product and making a score, and extracting a product or the like that matches the taste of the customer by an approximate calculation of these scores, In addition to generating customer attribute information reflecting product attribute information, customer attribute information and product attribute information are mutually optimized by reflecting customer attribute information in product attribute information as well. It becomes possible to grasp each attribute more accurately.

  In addition, by using optimized attribute information related to recommendations of products, etc., it is possible to more effectively select products that are recommended for customers, and to make customers more effective for recommending products. Can be selected automatically. In addition, the use of optimized products and customer attribute information is not limited to recommendations such as products. Appropriately grasp the attributes of products and use them in advertising production (for example, we recognize products for men) If the product ’s male attribute score is unexpectedly low, you may want to add a little female-facing elements to your ads), analyze product attributes and customer attributes, It can be used widely for product development.

It is a figure which shows an example of the recommendation by collaborative filtering. It is a figure which shows an example of the recommendation by a needs guess type | mold. It is a figure which shows the outline | summary of the method of optimizing customer attribute information and goods attribute information by this invention. It is a figure which shows an example of the outline | summary of embodiment of this invention. It is a figure which shows an example of a structure of the recommendation object selection apparatus provided with the attribute information optimization apparatus concerning this invention. It is a flowchart which shows the flow which optimizes customer attribute information and goods attribute information by this invention. It is a 1st figure which shows an example by which customer attribute information and goods attribute information are optimized by this invention. It is a 2nd figure which shows an example by which customer attribute information and goods attribute information are optimized by this invention. It is a 3rd figure which shows an example by which customer attribute information and goods attribute information are optimized by this invention. It is a 4th figure which shows an example by which customer attribute information and goods attribute information are optimized by this invention. It is a 5th figure which shows an example by which customer attribute information and goods attribute information are optimized by this invention. It is a 6th figure which shows an example by which customer attribute information and goods attribute information are optimized by this invention. It is a 7th figure which shows an example by which customer attribute information and goods attribute information are optimized by this invention. It is a flowchart which shows the flow which selects the goods used as the object recommended to a customer by this invention. It is a flowchart which shows the flow which selects the customer used as the object which recommends goods by this invention.

  Embodiments for carrying out the present invention will be described below in detail with reference to the drawings. In the following description, an example of an embodiment of the present invention will be described in which the customer and product attribute information is updated and optimized from the purchase information of the product purchased by the customer, and these are used for recommending the product to the customer. However, the present invention is not limited to such an embodiment. For example, the present invention can target not only the purchase of products but also the use of services, and the attribute information can be updated for the browsing information of product pages instead of the purchase information of products.

  FIG. 3 shows an overview of a method for optimizing customer attribute information and product attribute information according to the present invention. In the present invention, a common attribute item is set for each customer and product to be recommended, and a score indicating the degree of each attribute is recorded in the customer attribute master and the product attribute master for each customer and each product.

  The contents of the attribute items set here are not particularly limited, but attributes such as customer profile (gender, age group, family composition, etc.) and insight (preference, lifestyle, values, events, etc.) It may be set as an item. Since the degree of these attributes is indicated as a score, for example, the degree of the male degree of the customer (whether or not a product that is generally preferred by men is often purchased) is determined by the score for the attribute item “male” in the customer attribute master. It will be able to grasp.

  On the other hand, the same attribute item as the customer attribute master is provided in the product attribute master, and the degree of the attribute is indicated as a score. It will be possible to grasp (whether the product is generally easily purchased by men).

  Assuming the presence of the customer attribute master and the product attribute master as described above, in the present invention, when purchase information for purchasing a certain product is received, the attribute score of the purchased product is obtained for the attributes common to both. The customer attribute score of the purchased product is reflected in the score of the attribute of the purchased product, and the customer attribute information and the product attribute information are updated. In this way, due to the purchase behavior of the customer, both the attribute information of the customer and the attribute information of the product interact with each other. It will approach the optimal state.

  For example, when a customer purchases a product with a high masculine score, the score is reflected in the masculine score of the customer, and the masculine score of the customer is also reflected in the masculine score of the product. If the customer's masculine score is relatively low, purchasing a product that is favored by men increases the masculine score of the customer, but a customer who does not like men's products purchases the product. Therefore, conversely, the masculine score of the product will decrease, and both will influence each other toward the optimal state.

  FIG. 4 shows an example of the outline of the embodiment of the present invention. The customer database that stores the customer attribute master and the product database that stores the product attribute master described with reference to FIG. 3 are selected, the customer attribute information and the product attribute information are optimized, and the product recommended to the customer is selected. The server having the function to correspond to the recommendation server in FIG.

  The recommendation server accepts purchase information specifying the customer (customer code) and the product (product code) for the product purchased by the customer, and updates the customer attribute master and the product attribute master. Whether the route is accepted is not particularly limited. For example, the customer's point card membership number (corresponding to the customer code) received at the cash register terminal and the product code corresponding to the sold product may be received as purchase information via the sales management server, Connect to the Internet using a mobile phone, enter the user ID (corresponding to the customer code), and accept the product code corresponding to the product purchased on the web page of the online shop as purchase information via the e-commerce server It is good.

  In the present invention, the information specifying the customer (customer code) and the product (product code) used for updating the customer attribute master and the product attribute master is not limited to information based on the purchase of the product. For example, the membership number (corresponding to the customer code) of the customer's point card received at the in-store terminal installed in the store and the product code corresponding to the product selected by the customer to confirm the product information via the sales management server Goods on the product page that the customer connected to the Internet using a PC or mobile phone, entered the user ID (corresponding to the customer code), searched on the web page of the logged-in net shop, and selected It is good also as receiving the goods code corresponding to to via an electronic commerce server.

  The recommendation server searches the customer attribute master and the product attribute master in response to a product recommendation request from a sales management server, electronic commerce server, etc., selects the product to be recommended, and returns it to the server that sent the request . For example, when a customer who visits a store operates a terminal in the store and enters a membership card number (corresponding to a customer code) on a point card and searches for “recommended products”, the recommendation server of the products is recommended to the recommendation server via the sales management server. A request is transmitted, information about the product selected as a recommendation target is returned via the sales management server, and is output to the in-store terminal. When a customer connects to the Internet using a PC or mobile phone, enters a user ID (corresponding to a customer code) and accesses a personal page of the net shop, a request for a product recommendation is sent to the recommendation server via the e-commerce server. Information relating to the product sent and selected for recommendation is sent back via the electronic commerce server and output to the PC or mobile phone.

  Although not shown in FIG. 2, the recommendation server searches the customer attribute master and the product attribute master in response to a request from a Web server, etc., selects a customer to be recommended for the product, and transmits the request. It is also possible to reply to the server that made it. For example, when a person in charge of sales of a product or the like operates a PC or the like to access a recommendation management Web page and requests a list of customers to whom a product to which a product code is input is recommended, via a Web server Then, a request for transmitting the customer list is transmitted to the recommendation server, and information related to the customer selected as the recommendation target is returned via the Web server and output to the PC or the like.

  Here, the method for the recommendation server to search the customer attribute master and the product attribute master and select the product or customer to be recommended is not particularly limited. Since common attribute items are provided and a score indicating the degree of each attribute is recorded, the approximate calculation of the scores for the common attribute items can be used to identify customers who prefer products and products that match the customer's preferences. You can search. Also, instead of selecting products and customers only by such approximate calculations, select products and customers by using in combination with other methods, for example, narrowing down by general attributes (age, residence, etc.) It is good as well.

  In addition, in the present invention, the target of the recommendation is not limited to the product, but may be intended for use of the service. In this case, the product may be replaced with the service, and the purchase information of the product may be replaced with service usage information.

  FIG. 5 shows an example of the configuration of a recommendation target selection device provided with the attribute information optimization device according to the present invention. In FIG. 5, the recommendation server 10 corresponds to the recommendation target selection device provided with the attribute information optimization device according to the present invention, and has the function of the recommendation server described with reference to FIG.

  The sales management server 20, the electronic commerce server 30, and the Web server 40 are all server computers that are connected to the recommendation server 10 via a network and can communicate with the recommendation server 10. In any case, the physical configuration is not particularly limited, but each corresponds to the sales management server, the electronic commerce server, and the Web server described above with reference to FIG. 4. It is configured to receive information and requests and return information such as recommended products and customer lists.

  Although the physical configuration of the recommendation server 10 is not particularly limited, a general computer having an auxiliary storage device such as a CPU, a main memory, and an HDD and having a data communication function can be used. A predetermined function in the recommendation server 10 is realized by calling an application program stored in an auxiliary storage device such as an HDD into the main memory and performing arithmetic processing by the CPU.

  The purchase information receiving unit 14, the attribute information updating unit 15, and the recommendation target selecting unit 16 provided in the recommendation server 10 are all functionally specified, and an application program corresponding to the function of each unit is an HDD or the like. A predetermined function is realized by being called from the auxiliary storage device to the main memory and performing arithmetic processing by the CPU.

  The customer attribute information storage unit 11 (customer attribute master), the product attribute information storage unit 12 (product attribute master), and the product information storage unit 13 provided in the recommendation server 10 each have a predetermined storage area of an auxiliary storage device such as an HDD. Are not required to be physically installed in one computer. Any part or all of them may be configured as a whole as one computer system, such as being part of another computer such as a database server that is different from the server computer that executes the application program described above.

  Based on the above configuration, a flow for optimizing customer attribute information and product attribute information according to the present invention will be described using the flowchart of FIG.

  When purchase information on the purchase of the product by the customer is transmitted from the sales management server 20 or the electronic commerce server 30 with the customer code and the product code specified, the purchase information receiving unit 14 is activated in the recommendation server 10 and received. A predetermined process for receiving the purchase information, such as writing the purchase information in the database as a purchase history, is executed (S01).

  Subsequently, the attribute information update unit 15 is activated, and processing for updating the customer attribute information and the product attribute information is performed. Specifically, the customer code included in the purchase information is specified (S02), and the customer attribute information storage unit 11 (customer attribute master) is searched (S03, S04). If customer attribute information corresponding to the customer code is stored, this customer attribute information is read (S05). If customer attribute information corresponding to the customer code is not stored, the customer attribute information storage unit 11 (customer Customer attribute information to be newly written in (attribute master) is generated (S06).

  Further, the product code included in the purchase information is specified (S07), and the product attribute information storage unit 12 (shipment attribute master) is searched (S08, S09). If the product attribute information corresponding to the product code is stored, the product attribute information is read (S10). If the product attribute information corresponding to the product code is not stored, the product attribute information storage unit 12 (product Product attribute information to be newly written in (attribute master) is generated (S11).

  Subsequently, for the customer attribute information read in S05 or the customer attribute information generated in S06, the product attribute information read in S10 is a score indicating the degree of attribute recorded in each attribute item included in the customer attribute information. Or updated to reflect the score indicating the degree of attribute recorded in the common attribute item included in the product attribute information generated in S11, and store customer attribute information as updated customer attribute information including a new score It writes in part 11 (customer attribute master) (S12).

  Further, for the product attribute information read out in S10 or the product attribute information generated in S11, the customer attribute information read out in S05, the score indicating the degree of attribute recorded in each attribute item included in the product attribute information, Alternatively, the product attribute information storage unit is updated as the updated product attribute information including the new score by reflecting the score indicating the degree of the attribute recorded in the common attribute item included in the customer attribute information generated in S06. 12 (product attribute master) is written (S13), and the process of updating the customer attribute information and the product attribute information is terminated.

  FIGS. 7 to 13 show specific examples of the processing for updating the above customer attribute information and product attribute information. As a premise, the customer attribute information and the product attribute information indicate, as common attribute items, the degree of attributes such as attribute 1 = masculine, attribute 2 = youthful, attribute 3 = smooth nature, and attribute 4 = outdoor orientation. Assume that a score is recorded.

  In FIG. 7, customer X, customer Y, customer Z, and customer V are all new customers, and as shown in the left column, the scores indicating the degree of attributes are all zero (no score exists). However, what value is set as the initial value of the score indicating the degree of the attribute of the new customer is not particularly limited. In this figure, the number “0” recorded on the right side of customer X, customer Y, customer Z, and customer V represents the cumulative value of the number of items purchased by each customer.

  In this state, customer X receives product A (hair growth agent), customer Y receives product B (mascara), customer Z receives product C (adhesive bandage), and customer V purchases product D (sunscreen lotion). To do. If the product attribute information of these products is as shown in the center column of FIG. 7, the customer attributes of customer X, customer Y, customer Z, and customer V are reflected by reflecting the scores recorded in the common attribute items. The information is updated as shown in the right column of FIG. In this example, since customer X, customer Y, customer Z, and customer V are all new customers and do not have scores for any attribute items, the scores of the purchased products are reflected as they are. In this figure, the numbers recorded on the right side of the products A, B, C, and D indicate the cumulative value of the purchase quantity of each product.

  As shown in this example, the “hair restorer” of the product A is a product that has a high male attribute score of 90.0, that is, a product that is easily purchased by men, and purchased the product A. The high score of 90.0 is also reflected in the score indicating the male attribute of customer X. However, it is a feature of the present invention that instead of recognizing the customer X as “male”, the score indicating the male attribute is recognized as 90.0. On the other hand, the score indicating the male attribute of the “mascara” of the product B is 0.0 indicates that the product has never been purchased by a customer who has any male attribute. The score indicating the male attribute of the customer Y who purchased the product B is also 0.0.

  At the time of purchase of the first product above, since the customer attribute information before the update has no score for any attribute item, the score included in the customer attribute information is reflected in the product attribute information. Absent.

  Next, FIGS. 8 and 9 show processing for updating customer attribute information and product attribute information when customer X, customer Y, customer Z, and customer V purchase the second product. Here, it is assumed that customer X has purchased product E (rice pack), customer Y has purchased product F (shampoo), customer Z has purchased product G (contact cleaning liquid), and customer V has purchased product H (detergent).

  The left column of FIG. 8 shows customer attribute information of customer X, customer Y, customer Z, and customer V before purchasing the second product. Here, if the purchase information for purchasing the above-mentioned products is received and the product attribute information of those products is as shown in the center column of FIG. 8, the scores recorded in the common attribute items are reflected. The customer attribute information of customer X, customer Y, customer Z, and customer V is updated as shown in the right column of FIG. In this example, for example, the score for attribute 1 of customer X is 65.0 by calculating the average value of 90.0 before update and 40.0 of the purchased product. The formula only needs to reflect the score included in the product attribute information of the newly purchased product, and the specific calculation formula is not particularly limited.

  On the other hand, the left column of FIG. 9 shows the product attribute information of the product E, the product F, the product G, and the product H before being purchased by the customer X, the customer Y, the customer Z, and the customer V. Here, when each purchase information purchased by the customer is received, since the customer attribute information of the customer is as shown in the center column of FIG. 9, the score recorded in each common attribute item is reflected. Thus, the product attribute information of the product E, the product F, the product G, and the product H is updated as shown in the right column of FIG. In this example, for example, the score for the attribute 1 of the product E is 90.0 of the customer X newly purchased in consideration of the purchase quantity with respect to 40.0 before the update ((40 0.0 × 100 + 90.0 × 1) ÷ (100 + 1)), which is updated to 40.5, the calculation formula here is the score included in the customer attribute information of the customer who newly purchased the product It is only necessary to be reflected, and the specific arithmetic expression is not particularly limited.

  In the examples of FIGS. 8 and 9, when updating the customer attribute information and the product attribute information, the scores relating to the customer and product attributes specified from the new purchase information are obtained based on the scores before the update. The score is to be updated and the score included in either the customer attribute information or the product attribute information is reflected in the score related to the customer or product attribute specified from the new purchase information. The score of the other attribute information may be updated using the updated score.

  In the case of the purchase of the second product, when the product attribute information is updated after the customer attribute information is updated as shown in FIG. 8, the customer attribute information in the center column of FIG. Can be replaced with the customer attribute information after the purchase of the second product in the right column of FIG. 8, and if the update process is performed in such a procedure, the customer after being updated with the purchase information Since attribute information is reflected in the update of product attribute information, it is suitable when optimization of product attribute information is considered to precede optimization of customer attribute information (when customer attribute information is not sufficiently formed) It is.

  Conversely, when updating the customer attribute information after updating the product attribute information as shown in FIG. 9, the product attribute information in the center column of FIG. 8 reflected at the time of updating is in the right column of FIG. It is also possible to replace the product attribute information after the purchase of the second product by the customer. When the update process is performed in such a procedure, the product attribute information updated by the purchase information becomes the customer attribute information. Since it is reflected in the update, it is suitable when it is considered that the optimization of the customer attribute information precedes the optimization of the product attribute information (when the product attribute information is not sufficiently formed).

  Next, FIGS. 10 and 11 show processing for updating customer attribute information and product attribute information when customer X, customer Y, customer Z, and customer V purchase a third product. Here, it is assumed that customer X purchases product I (natto), customer Y purchases product J (glass polishing), customer Z purchases product K (shaving blade), and customer V purchases product L (cold medicine).

  The left column of FIG. 10 shows customer attribute information of customer X, customer Y, customer Z, and customer V before purchasing the third product. Here, if the purchase information for purchasing the above-mentioned products is received and the product attribute information of those products is as shown in the center column of FIG. 10, the score recorded in the common attribute item is reflected. The customer attribute information of customer X, customer Y, customer Z, and customer V is updated as shown in the right column of FIG. In this example, for example, the score for the attribute 1 of the customer X is a weighted average ((65 0.0 × 2 + 50.0 × 1) ÷ (2 + 1)), the score is updated to 60.0, but the calculation formula here reflects the score included in the product attribute information of the newly purchased product Any specific arithmetic expression is not particularly limited.

  On the other hand, the left column of FIG. 11 shows the product attribute information of the product I, the product J, the product K, and the product L before being purchased by the customer X, the customer Y, the customer Z, and the customer V. Here, when each purchase information purchased by the customer is received, since the customer attribute information of the customer is as shown in the center column of FIG. 11, the score recorded in each common attribute item is reflected. The product attribute information of the product I, the product J, the product K, and the product L is updated as shown in the right column of FIG. In this example, for example, the score for the attribute 1 of the product I is 65.0 of the customer X newly purchased in consideration of the purchase quantity with respect to 50.0 before the update ((50 0.0 × 50 + 65.0 × 1) ÷ (50 + 1)), which is updated to 50.3, the calculation formula here is the score included in the customer attribute information of the customer who newly purchased the product. It is only necessary to be reflected, and the specific arithmetic expression is not particularly limited.

  In addition, when updating customer attribute information and product attribute information, the score included in the attribute information of either customer attribute information or product attribute information is the score related to the customer or product attribute specified from the new purchase information. As described in the examples of FIGS. 8 and 9, the score of the other attribute information may be updated using the updated score.

  Next, FIGS. 12 and 13 show processing for updating customer attribute information and product attribute information when customer X, customer Y, customer Z, and customer V purchase the fourth product. Here, it is assumed that customer X purchases product L (cold medicine), customer Y purchases product M (hairdressing agent), customer Z purchases product E (rice pack), and customer V purchases product N (lipstick).

  The left column of FIG. 12 shows customer attribute information of customer X, customer Y, customer Z, and customer V before purchasing the fourth product. Here, if the purchase information for purchasing the above-mentioned products is received and the product attribute information of those products is as shown in the center column of FIG. 12, the score recorded in the common attribute item is reflected. The customer attribute information of customer X, customer Y, customer Z, and customer V is updated as shown in the right column of FIG. In this example, for example, the score for the attribute 1 of the customer X is the weighted average ((60) of the newly purchased product L in consideration of the number of purchases with respect to 60.0 before the update. 0.0 × 3 + 48.3 × 1) ÷ (3 + 1)), the score is updated to 57.1, but the calculation formula here reflects the score included in the product attribute information of the newly purchased product Any specific arithmetic expression is not particularly limited.

  On the other hand, the left column of FIG. 13 shows the product attribute information of the product L, the product M, the product E, and the product N before being purchased by the customer X, the customer Y, the customer Z, and the customer V. Here, when each purchase information purchased by the customer is received, since the customer attribute information of the customer is as shown in the center column of FIG. 13, the score recorded in each common attribute item is reflected. Thus, the product attribute information of the product L, the product M, the product E, and the product N is updated as shown in the right column of FIG. In this example, for example, the score for the attribute 1 of the product L is a weighted average ((48) of 60.0 of the customer X newly purchased in consideration of the purchase quantity with respect to 48.3 before the update. .3 × 21 + 60.0 × 1) ÷ (21 + 1)), which is updated to 48.8, the calculation formula here is the score included in the customer attribute information of the customer who newly purchased the product. It is only necessary to be reflected, and the specific arithmetic expression is not particularly limited.

  In addition, when updating customer attribute information and product attribute information, the score included in the attribute information of either customer attribute information or product attribute information is the score related to the customer or product attribute specified from the new purchase information. As described in the examples of FIGS. 8 and 9, the score of the other attribute information may be updated using the updated score.

  As described above, by repeatedly updating the customer attribute information and the score of the common attribute item included in the product attribute information from the purchase information of the customer's product, the customer attribute information and the product As attribute information influences each other and more purchase information is accumulated, the attribute information is optimized as it is accumulated.

  As a result, for example, regarding the attributes of customer X, there is a tendency to prefer products for men rather than just “non-young men” (male attribute score 57.1), and purchase products for young people as well. Is not so strong (youth attribute score of 40.2), but tends to do relatively frequent bulk purchases without buying products (bean attribute score of 32.5). It can be grasped as detailed attribute information in the form of not much interest (outdoor attribute score 32.5).

  Next, on the premise of the configuration shown in FIG. 5, a flow for selecting a product to be recommended to a customer according to the present invention will be described using the flowchart of FIG.

  When a recommendation request for a product for which a customer code is specified is transmitted from the sales management server 20 or the electronic commerce server 30, the recommendation server 10 activates the recommendation target selection unit 15 and includes the customer code included in the recommendation request. Is accepted (S21).

  The customer attribute information storage unit 11 (customer attribute master) is searched using the received customer code as a key (S22, S23). If customer attribute information corresponding to the customer code is stored, this customer attribute information is read (S24). ), Whether there is product attribute information that meets a predetermined condition, such as a score indicating the degree of attribute for at least some of the attributes included in the customer attribute information is similar to a score indicating the degree of common attribute, Search in the product attribute information storage unit 12 (product attribute master) (S25, S26).

  The predetermined condition for searching for the product attribute information is not particularly limited, and may be a condition that the scores for the common attribute items included in the customer attribute information and the product attribute information are all approximate. , A part of which may be approximated. In addition, the predetermined condition for searching for product attribute information is not only that the scores for the common attribute items are approximate, but also other attribute items, for example, that the unit price is 500 yen or more, and products within one month of sale. It is also possible to use a value of an attribute item such as “in combination” as a narrowing-down condition.

  In S23, if there is no customer attribute information corresponding to the customer code, or if there is no product attribute information that matches the predetermined condition in S26, error processing is performed (S28), but product attribute information that matches the predetermined condition. If there is a product, the product corresponding to the product attribute information that meets the conditions is set as the product to be recommended, and the product information such as the price, product description, and image of the target product is read from the product information storage unit 13 for the customer. Is generated and returned to the server that sent the request (S27).

  Next, on the premise of the configuration shown in FIG. 5, a flow for selecting a customer to be recommended for a product according to the present invention will be described using the flowchart of FIG. 15.

  When a request for a customer list of a customer to recommend a product for which a product code is specified is transmitted from the Web server 40 or the like, the recommendation server 10 activates the recommendation target selection unit 15 to include the product included in the customer list request. The code is accepted (S31).

  The product attribute information storage unit 12 (product attribute master) is searched using the received product code as a key (S32, S33). If product attribute information corresponding to the product code is stored, the product attribute information is read (S34). ), Whether there is customer attribute information that matches a predetermined condition, such as a score indicating the degree of attribute for at least some of the attributes included in the product attribute information is similar to a score indicating the degree of common attribute, Search in the customer attribute information storage unit 11 (product attribute master) (S35, S36).

  Note that the predetermined condition for searching the customer attribute information is not particularly limited, and may be that the scores for the common attribute items included in the customer attribute information and the product attribute information are all approximate. , A part of which may be approximated. Further, the predetermined condition for searching for customer attribute information is not only that the scores for the common attribute items are approximate, but also other attribute items such as being resident in Tokyo, 1 after joining a point card member The value of an attribute item such as being a customer within a month may be used together as a narrowing condition.

  In S33, if there is no customer attribute information corresponding to the customer code, or if there is no customer attribute information that matches the predetermined condition in S36, error processing is performed (S38), but customer attribute information that matches the predetermined condition. Is present, the customers corresponding to the customer attribute information that matches the conditions are extracted as the customers to be recommended for the product, the customer list of these customers is generated, and returned to the server that sent the request ( S37).

DESCRIPTION OF SYMBOLS 10 Recommendation server 11 Customer attribute information storage part 12 Product attribute information storage part 13 Product information storage part 14 Purchase information receiving part 15 Attribute information update part 16 Recommendation object selection part 20 Sales management server 30 Electronic commerce server 40 Web server

Claims (11)

An attribute information optimization device for optimizing customer attribute information and product or service attribute information,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute as a score for each customer, for each customer;
Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree of each attribute for the product or service attribute.
At least a part of the customer attribute in the customer attribute information and the attribute of the product or service in the product attribute information, a common attribute is set,
Purchasing information receiving means for receiving purchasing information in which the customer and the product or service are specified with respect to the product purchased or used by the customer or the product or service selected by the customer;
A score indicating the degree of attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means is specified from the purchase information stored in the product attribute information storage means. Customer attribute information update means for reflecting the score indicating the degree of the attribute included in the product attribute information of the product or service;
A score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit is specified from the purchase information stored in the customer attribute information storage unit Merchandise attribute information updating means for reflecting the score indicating the degree of the attribute included in the customer attribute information of the customer to be
A device for optimizing attribute information. The product attribute information update means indicates the degree of the attribute included in the product attribute information using a score indicating the degree of attribute included in the customer attribute information of the customer updated by the customer attribute information update means. The attribute information optimizing apparatus according to claim 1, wherein the score is updated. The customer attribute information update unit uses the score indicating the degree of attribute included in the product attribute information of the product or service updated by the product attribute information update unit, and the degree of the attribute included in the customer attribute information The attribute information optimizing apparatus according to claim 1, wherein a score indicating the update is updated. An attribute information optimization program for optimizing customer attribute information and product or service attribute information,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute with respect to customer attributes for each customer, and product attribute information indicating the degree of each attribute with respect to attributes of products or services Product attribute information storage means for storing each product or service, and a common attribute is set in at least a part of the customer attribute in the customer attribute information and the product attribute or service attribute in the product attribute information Computer system
A purchase information receiving step for receiving purchase information in which the customer and the product or service are identified with respect to the product purchased or used by the customer or the product or service selected by the customer;
A score indicating the degree of attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means is specified from the purchase information stored in the product attribute information storage means. A customer attribute information update step that updates the score indicating the degree of the attribute included in the product attribute information of the product or service;
A score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit is specified from the purchase information stored in the customer attribute information storage unit A product attribute information update step for updating to reflect a score indicating the degree of the attribute included in the customer attribute information of the customer to be
An attribute information optimization program characterized by causing An attribute information optimization method for optimizing customer attribute information and product or service attribute information,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute with respect to customer attributes for each customer, and product attribute information indicating the degree of each attribute with respect to attributes of products or services Product attribute information storage means for storing each product or service, and a common attribute is set in at least a part of the customer attribute in the customer attribute information and the product attribute or service attribute in the product attribute information A purchase information reception step for receiving purchase information in which the customer and the product or service are identified with respect to the product purchased or used by the customer or the product or service selected by the customer;
The computer system stores a score indicating a degree of an attribute included in customer attribute information of a customer specified from the purchase information stored in the customer attribute information storage unit, and the purchase stored in the product attribute information storage unit A customer attribute information update step that reflects and updates a score indicating the degree of the attribute included in the product attribute information of the product or service specified from the information;
The computer system stores, in the customer attribute information storage unit, a score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit A product attribute information update step that updates the score indicating the degree of the attribute included in the customer attribute information of the customer specified from the purchase information;
A method for optimizing attribute information characterized by comprising: A recommendation target selection device for selecting a product or service to be recommended to a customer using optimized customer attribute information and product or service attribute information,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute as a score for each customer, for each customer;
Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree of each attribute for the product or service attribute.
At least a part of the customer attribute in the customer attribute information and the attribute of the product or service in the product attribute information, a common attribute is set,
Purchasing information receiving means for receiving purchasing information in which the customer and the product or service are specified with respect to the product purchased or used by the customer or the product or service selected by the customer;
A score indicating the degree of attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means is specified from the purchase information stored in the product attribute information storage means. Customer attribute information update means for reflecting the score indicating the degree of the attribute included in the product attribute information of the product or service;
A score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit is specified from the purchase information stored in the customer attribute information storage unit Merchandise attribute information updating means for reflecting the score indicating the degree of the attribute included in the customer attribute information of the customer to be
When receiving the designation of the customer, in the product attribute information of the product or service stored in the product attribute information storage means, at least some of the attributes included in the customer attribute information of the customer stored in the customer attribute information storage means A recommendation target selecting means for selecting, as a product or service to be recommended to the customer, a product or service corresponding to product attribute information for which the score indicating the attribute level and the score indicating the attribute level match a predetermined condition When,
A recommendation target selection device comprising: A recommendation target selection device for selecting a customer to be recommended for a product or service using the optimized attribute information of the customer and the attribute information of the product or service,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute as a score for each customer, for each customer;
Product attribute information storage means for storing, for each product or service, product attribute information indicating the degree of each attribute for the product or service attribute.
At least a part of the customer attribute in the customer attribute information and the attribute of the product or service in the product attribute information, a common attribute is set,
Purchasing information receiving means for receiving purchasing information in which the customer and the product or service are specified with respect to the product purchased or used by the customer or the product or service selected by the customer;
A score indicating the degree of attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means is specified from the purchase information stored in the product attribute information storage means. Customer attribute information update means for reflecting the score indicating the degree of the attribute included in the product attribute information of the product or service;
A score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit is specified from the purchase information stored in the customer attribute information storage unit Merchandise attribute information updating means for reflecting the score indicating the degree of the attribute included in the customer attribute information of the customer to be
When designation of a product or service is accepted, at least a part of the customer attribute information of the product or service stored in the product attribute information storage unit in the customer attribute information of the customer stored in the customer attribute information storage unit A recommendation target selecting means for selecting, as a customer who recommends the product or service, a score indicating the degree of the attribute for the attribute and a customer corresponding to customer attribute information whose score indicating the degree of attribute matches a predetermined condition When,
A recommendation target selection device comprising: A recommendation target selection device for selecting a product or service to be recommended to a customer using optimized customer attribute information and product or service attribute information,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute with respect to customer attributes for each customer, and product attribute information indicating the degree of each attribute with respect to attributes of products or services Product attribute information storage means for storing each product or service, and a common attribute is set in at least a part of the customer attribute in the customer attribute information and the product attribute or service attribute in the product attribute information Computer system
A purchase information receiving step for receiving purchase information in which the customer and the product or service are identified with respect to the product purchased or used by the customer or the product or service selected by the customer;
A score indicating the degree of attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means is specified from the purchase information stored in the product attribute information storage means. A customer attribute information update step that updates the score indicating the degree of the attribute included in the product attribute information of the product or service;
A score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit is specified from the purchase information stored in the customer attribute information storage unit A product attribute information update step for updating to reflect a score indicating the degree of the attribute included in the customer attribute information of the customer to be
When receiving the designation of the customer, in the product attribute information of the product or service stored in the product attribute information storage means, at least some of the attributes included in the customer attribute information of the customer stored in the customer attribute information storage means A recommendation target selection step of selecting, as a product or service to be recommended to the customer, a product or service corresponding to product attribute information for which the score indicating the attribute level and the score indicating the attribute level match a predetermined condition When,
The recommendation object selection program characterized by performing this. A recommendation target selection program for selecting a customer to recommend a product or service using the optimized customer attribute information and product or service attribute information,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute with respect to customer attributes for each customer, and product attribute information indicating the degree of each attribute with respect to attributes of products or services Product attribute information storage means for storing each product or service, and a common attribute is set in at least a part of the customer attribute in the customer attribute information and the product attribute or service attribute in the product attribute information Computer system
A purchase information receiving step for receiving purchase information in which the customer and the product or service are identified with respect to the product purchased or used by the customer or the product or service selected by the customer;
A score indicating the degree of attribute included in the customer attribute information of the customer specified from the purchase information stored in the customer attribute information storage means is specified from the purchase information stored in the product attribute information storage means. A customer attribute information update step that updates the score indicating the degree of the attribute included in the product attribute information of the product or service;
A score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit is specified from the purchase information stored in the customer attribute information storage unit A product attribute information update step for updating to reflect a score indicating the degree of the attribute included in the customer attribute information of the customer to be
When designation of a product or service is accepted, at least a part of the customer attribute information of the product or service stored in the product attribute information storage unit in the customer attribute information of the customer stored in the customer attribute information storage unit A recommendation target selection step of selecting a customer corresponding to customer attribute information in which a score indicating an attribute level of the attribute and a score indicating the attribute level match a predetermined condition as a customer recommending the product or service When,
The recommendation object selection program characterized by performing this. A method of selecting a recommendation target for selecting a product or service to be recommended to a customer using the optimized attribute information of the customer and the attribute information of the product or service,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute with respect to customer attributes for each customer, and product attribute information indicating the degree of each attribute with respect to attributes of products or services Product attribute information storage means for storing each product or service, and a common attribute is set in at least a part of the customer attribute in the customer attribute information and the product attribute or service attribute in the product attribute information A purchase information reception step for receiving purchase information in which the customer and the product or service are identified with respect to the product purchased or used by the customer or the product or service selected by the customer;
The computer system stores a score indicating a degree of an attribute included in customer attribute information of a customer specified from the purchase information stored in the customer attribute information storage unit, and the purchase stored in the product attribute information storage unit A customer attribute information update step that reflects and updates a score indicating the degree of the attribute included in the product attribute information of the product or service specified from the information;
The computer system stores, in the customer attribute information storage unit, a score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit A product attribute information update step that updates the score indicating the degree of the attribute included in the customer attribute information of the customer specified from the purchase information;
When the computer system accepts a customer designation, the product attribute information of the product or service stored in the product attribute information storage means is included in the customer attribute information of the customer stored in the customer attribute information storage means Select a product or service corresponding to product attribute information for which a score indicating the degree of attribute for at least some attributes and a score indicating the degree of attribute match a predetermined condition as a product or service to recommend to the customer A recommendation target selection step,
A method for selecting a recommendation target, characterized by comprising: A method of selecting a recommendation target for selecting a product or service to be recommended to a customer using the optimized attribute information of the customer and the attribute information of the product or service,
Customer attribute information storage means for storing customer attribute information indicating the degree of each attribute with respect to customer attributes for each customer, and product attribute information indicating the degree of each attribute with respect to attributes of products or services Product attribute information storage means for storing each product or service, and a common attribute is set in at least a part of the customer attribute in the customer attribute information and the product attribute or service attribute in the product attribute information A purchase information reception step for receiving purchase information in which the customer and the product or service are identified with respect to the product purchased or used by the customer or the product or service selected by the customer;
The computer system stores a score indicating a degree of an attribute included in customer attribute information of a customer specified from the purchase information stored in the customer attribute information storage unit, and the purchase stored in the product attribute information storage unit A customer attribute information update step that reflects and updates a score indicating the degree of the attribute included in the product attribute information of the product or service specified from the information;
The computer system stores, in the customer attribute information storage unit, a score indicating the degree of attribute included in the product attribute information of the product or service specified from the purchase information stored in the product attribute information storage unit A product attribute information update step that updates the score indicating the degree of the attribute included in the customer attribute information of the customer specified from the purchase information;
When the computer system accepts designation of a product or service, in the customer attribute information of the customer stored in the customer attribute information storage unit, the product attribute information of the product or service stored in the product attribute information storage unit Select a customer corresponding to customer attribute information for which a score indicating the attribute level of at least some of the included attributes and a score indicating the attribute level match a predetermined condition as a customer recommending the product or service A recommendation target selection step,
A method for selecting a recommendation target, characterized by comprising: