JP2010073170A - Recommended commodity selection device, recommended commodity selection program and commodity retrieval device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow even a user using online shopping for the first time to select a proper recommended commodity, and to easily set a condition for selecting the recommended commodity. <P>SOLUTION: This recommended commodity selection device 13 has: a totaling part 13a collating a database storing a plurality of conditions related to attribute information of a user in each commodity as recommendation conditions for deciding whether the commodity is set as the recommended commodity or not with the attribute information of a user having purchased the commodity, and totaling a frequency wherein the recommendation condition matches the attribute information of the user having purchased the commodity, for each recommendation condition; a weight evaluation part 13b determining weight showing an importance level of each recommendation condition based on a totaling result of the totaling part 13a; and a recommended commodity selection part 13c evaluating the importance level of the recommendation condition matching the attribute information of a designated customer by the weight to select a recommended commodity. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a recommended product selection device, a recommended product selection program, and a product search device used in an online shopping system, and in particular, can select a recommended product suitable for a user who uses the online shopping system for the first time, The present invention relates to a recommended product selection device, a recommended product selection program, and a product search device that can easily set conditions for selection.

  Many online shopping systems are configured to display recommended products on a shopping screen that a user refers to in order to purchase products, and to encourage purchase of recommended products. In order for the user to purchase the recommended product, it is important that the recommended product is a product that the user is interested in. For this reason, the recommended product is generally selected from products related to the product purchased by the user in the past.

  As a method for selecting a recommended product, a method of checking the current location of the user, the weather at the current location, the current time, and the like against a predetermined condition may be considered. By using this method, it is possible to select a recommended product according to the situation, such as presenting a product that feels cool as a recommended product when it is hot. A technique for providing information to a user based on the situation is disclosed in Patent Document 1, for example.

JP 2006-65650 A

  However, when a user selects a recommended product from products related to a product purchased in the past, the user uses the online shopping system for the first time, or purchases the product only several times using the online shopping system. If there is not, there is a problem in that it is impossible to select an appropriate recommended product because there is no purchase record or there is a shortage.

  In addition, when selecting products to recommend to users based on the current location, weather, and other conditions, it is necessary to set complicated conditions for which products to recommend in which circumstances. There was a problem that the load on the administrator to set was high.

  The present invention has been made in view of the above, and it is possible to select an appropriate recommended product for a user who uses the online shopping system for the first time, and to easily set conditions for selecting the recommended product. An object of the present invention is to provide a recommended product selection device, a recommended product selection program, and a product search device.

  In order to solve the above-described problems and achieve the object, according to one aspect of the present invention, as a recommendation condition for determining whether or not a product is a recommended product, a condition regarding user attribute information is set as a product. A totaling unit that collates a database that stores a plurality of data every time and the attribute information of the user who purchased the product, and totals the number of times the recommended condition matches the attribute information of the user who purchased the product for each recommended condition; , Weight evaluation means for determining the weight indicating the importance level of each recommendation condition based on the counting result of the counting means, and recommending by evaluating the importance level of the recommendation condition that matches the attribute information of the specified customer with the weight A recommended product selection means for selecting a product is provided.

  According to this aspect, since the condition regarding the attribute information of the user is set as the recommendation condition, it is possible to select an appropriate recommended product even for a user who uses the online shopping system for the first time. In addition, since the weight of the recommendation condition is automatically optimized based on the purchase record, the burden on the administrator who sets the recommendation condition can be reduced.

  ADVANTAGE OF THE INVENTION According to this invention, there exists an effect that the user who uses an online shopping system for the first time can select an appropriate recommended goods, and can reduce the burden of the administrator who sets recommendation conditions.

  Embodiments of a recommended product selection device, a recommended product selection program, and a product search device according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

  First, the configuration of the online shopping system 1 according to the present embodiment will be described. The online shopping system 1 according to the present embodiment recommends an appropriate product to each user by dynamically adjusting the weight of the recommendation condition for selecting the recommended product according to the trend of the user. Present as a product.

  FIG. 1 is a diagram illustrating a configuration of an online shopping system 1 according to the present embodiment. As shown in FIG. 1, the online shopping system 1 includes an online shopping server 11, a database device 12, and a recommended product selection device 13, and is connected to terminal devices 2a to 2n via a network 3 such as the Internet. Is done. The terminal devices 2a to 2n are information processing devices operated by a user who uses an online shopping service, and correspond to, for example, a personal computer or a mobile phone terminal.

  The online shopping server 11 is a device that provides an online shopping service. Specifically, the online shopping server 11 generates a shopping screen that is a screen for inputting information for checking information related to a product or inputting information for purchasing the product, and the terminal device 2a to 2n. When the online shopping server 11 generates a shopping screen, the online shopping server 11 causes the recommended product selection device 13 to select a recommended product and includes the selected recommended product in the shopping screen.

  The online shopping server 11 inputs information for purchasing a product on the shopping screen, and when the input information is transmitted to the online shopping server 11, the transmitted information is registered in the database device 12, and the product is shipped. And processing such as payment. Information registered in the database device 12 by the online shopping server 11 is also used by the recommended product selection device 13 to adjust the weight of the recommendation condition.

  The database device 12 stores various information used in the online shopping server 11. The configuration of the database device 12 is shown in FIG. As shown in FIG. 2, the database device 12 stores product data 12a, user data 12b, purchase data 12c, weather data 12d, recommended condition data 12e, and the like.

  The product data 12a is information relating to products sold in the online shopping system 1. An example of the product data 12a is shown in FIG. As shown in FIG. 3, the product data 12a includes items such as a product ID, a type, a product name, and a price. The product ID is an identification number for identifying the product. The type represents the type of product. The product name is the name of the product. The price is the selling price of the product.

  In the first line of the example shown in FIG. 3, the product identified by the identification number “AAA” belongs to the type “boots”, and the product name and price are “boots A” and “35000”, respectively. It shows that there is. In the second row of the example shown in FIG. 3, the product identified by the identification number “BBB” belongs to the type “boots”, and the product name and price are “boots B” and “43000, respectively. ".

  The user data 12b is attribute information of the user of the online shopping system 1. An example of the user data 12b is shown in FIG. As shown in FIG. 4, the user data 12b includes items such as a user ID, sex, age, and address. The user ID is an identification number for identifying the user. The sex, age, and address are the user's sex, age, and address, respectively.

  The first line of the example shown in FIG. 4 indicates that the user identified by the identification number “XXX” has gender, age, and address “female”, “20”, and “Tokyo”, respectively. Show. In the second row of the example shown in FIG. 4, the user identified by the identification number “YYY” has a gender, age, and address of “male”, “31”, and “Osaka”, respectively. It is shown that.

  The purchase data 12c is information relating to products purchased in the online shopping system 1. An example of the purchase data 12c is shown in FIG. As shown in FIG. 5, the purchase data 12c includes items such as a product ID, a user ID, a region, and a purchase date / time. The product ID is an identification number for identifying the product, and corresponds to the product ID of the product data 12a. The user ID is an identification number for identifying the user, and corresponds to the user ID of the user data 12b.

  The region represents the current location of the user when the product is purchased. Specifically, in the case where the terminal device 2 operated by the user at the time of product purchase has position acquisition means such as GPS (Global Positioning System), the name of the area acquired by the position acquisition means Is set in this item, otherwise, the address of the person registered in the user data 12b is set. The purchase date and time is the date and time when the product is purchased.

  The first row of the example shown in FIG. 5 shows that the product identified by the identification number “AAA” is “Tokyo” by the user identified by the identification number “XXX”. / 10 15:12 "indicates that the purchase was made on the date. Further, the second line of the example shown in FIG. 5 shows that the product identified by the identification number “CCC” is identified by the user identified by the identification number “YYY” in the area “Osaka” in “2008”. / 9/12 10:45 "indicates that the purchase was made on the date and time.

  The weather data 12d holds information related to the weather. An example of the weather data 12d is shown in FIG. As shown in FIG. 6, the weather data 12d holds weather and temperature (for example, maximum temperature) for each region and for each date.

  The first line of the example shown in FIG. 6 shows that in the area “Tokyo”, the weather is “sunny” and the temperature is “32 ° C.” on the date “2008/9/1”. The date “/ 2” indicates that the weather was “cloudy” and the temperature was “28 ° C.”. In the second row of the example shown in FIG. 6, in the area “Osaka”, the weather is “rain” and the temperature is “25 ° C.” on the date “2008/9/1”. The date “/ 9/2” indicates that the weather was “rain” and the temperature was “24 ° C.”.

  The recommendation condition data 12e holds recommendation conditions for selecting a product as a recommended product for a product that is a candidate for a recommended product. An example of the recommendation condition data 12e is shown in FIG. As shown in FIG. 7, the recommendation condition data 12e has items such as a product ID, type, recommendation condition, number of times of satisfaction, and weight, and a plurality of combinations of the recommendation condition, the number of times of satisfaction, and the weight for one product ID. It is configured to be able to hold.

  The product ID is an identification number for identifying the product, and corresponds to the product ID of the product data 12a. The type represents the type of product and corresponds to the type of product data 12a. The recommendation condition is a condition for selecting the product indicated by the corresponding product ID as the recommended product. The number of fulfillment times is the number of times that the corresponding recommendation condition is satisfied when the product indicated by the product ID is purchased. The weight is a value indicating the importance of the corresponding recommendation condition, and is set based on the number of fulfillment times. Here, the value of the weight is any one of “1” to “5”, and the greater the number, the higher the importance of the corresponding recommendation condition.

  In the example shown in FIG. 7, the product identified by the identification number “AAA” has “sex” as “female”, “age” is “10”, and “age” is “ “20” generation, “age” is “30” generation, “address” is “Tokyo”, “address” is “Osaka”, and “address” is “Kanagawa”. This indicates that the recommendation condition that “address” is “Aichi” is set.

  Further, in the example shown in FIG. 7, the product identified by the identification number “BBB” has the same “sex” as “female”, similar to the product identified by the identification number “AAA”. “Age” is “10” generation, “age” is “20” generation, “age” is “30” generation, “address” is “Tokyo”, “address” Indicates that the recommended condition is set to be “Osaka”, “Address” is “Kanagawa”, and “Address” is “Aichi”.

  As described above, the same recommendation condition is set for the product identified by the identification number “AAA” and the product identified by the identification number “BBB”, but the weight of the recommendation condition is different because the number of fulfillment is different. Is different.

  Specifically, regarding the age, the product identified by the identification number “AAA” has the highest number of satisfaction of the recommended condition that the “age” is in the “20” generation, and the weight of the recommended condition is the highest. It is getting bigger. This is because the product identified by the identification number “AAA” sold best to users in their 20s among those in their 10s and 30s. It means that the importance is the highest.

  On the other hand, regarding the age, the product identified by the identification number “BBB” has the highest number of satisfaction of the recommended condition that the “age” is in the “30” generation, and the weight of the recommended condition is the largest. Yes. This is because the product identified by the identification number “BBB” sold best to users in their 30s out of their teens to 30s. It means that the importance is the highest.

  As for the address, the product identified by the identification number “AAA” has the largest number of times the recommendation condition that the address is “Tokyo” is satisfied, and the weight of the recommendation condition is the largest. This is because the product identified by the identification number “AAA” sold best to users residing in Tokyo among the users at the address set as the recommendation condition. This means that the importance of recommendation conditions is the highest.

  On the other hand, regarding the address, the product identified by the identification number “BBB” has the largest number of satisfaction of the recommendation condition that the address is “Osaka”, and the weight of the recommendation condition is the largest. This is because the product identified by the identification number “BBB” sold best to users residing in Osaka among the users at the address set as the recommendation condition. This means that the importance of recommendation conditions is the highest.

  In the example shown in FIG. 7, the product identified by the identification number “CCC” has “weather” as “sunny”, “weather” as “cloudy”, and “weather”. “Rain”, “Temperature” is “30 ° C. or higher”, “Temperature” is “Less than 30 ° C.”, “Current location” is “Hokkaido”, “Current location” is “Tohoku” ”,“ Current location ”is“ Kanto ”,“ Current location ”is“ Chubu ”,“ Current location ”is“ Kansai ”,“ Current location ”is“ China ”,“ This indicates that a recommendation condition that “present location” is “Shikoku” and that “present location” is “Kyushu” is set.

  Each recommendation condition set for the product identified by the identification number “CCC” is also set with a weight corresponding to the number of fulfillment times.

  As described above, in the online shopping system 1 according to the present embodiment, the recommended product is selected based on the user's attributes and the recommended conditions according to the conditions such as the weather, the temperature, and the current location. For this reason, it is possible to select an appropriate recommended product even for a user who has never purchased a product with the online shopping system 1.

  Further, in the online shopping system 1 according to the present embodiment, even when the recommended conditions set for the product are the same, the importance of each recommended condition is changed according to the sales situation of each product. That is, even if the manager does not carefully set the recommendation conditions, the importance of the recommendation conditions is automatically optimized according to the sales situation.

  Returning to the description of FIG. 1, the recommended product selection device 13 executes various processes for selecting a recommended product based on the recommendation condition data 12e. The configuration of the recommended product selection device 13 is shown in FIG. As shown in FIG. 8, the recommended product selection device 13 includes a totaling unit 13a, a weight evaluation unit 13b, a recommended product selection unit 13c, and a sorting unit 13d.

  The totaling unit 13a updates the number of satisfaction of the recommendation condition data 12e based on the purchase data 12c. The weight evaluation unit 13b evaluates the weight of the recommendation condition based on the number of fulfillment times. The recommended product selection unit 13c selects a recommended product for the designated user. The sorting unit 13d sorts the products searched based on the search conditions designated by the designated user based on the recommendation condition data 12e.

  Next, the processing procedure of the recommended product selection device 13 will be described. FIG. 9 is a diagram illustrating a processing procedure of the satisfaction number updating process by the counting unit 13a. The fulfillment count update process is periodically executed while gradually shifting the target period.

  As shown in FIG. 9, the counting unit 13a first acquires a target period (step S101). And total part 13a acquires one piece of information on a target period from purchase data 12c (Step S102).

  Here, when the information of the target period can be acquired (No at Step S103), the counting unit 13a acquires the attribute information of the corresponding user from the user data 12b (Step S104), and purchase data 12c. The weather information on the day of purchase of the product indicated by the information acquired from is acquired from the weather data 12d (step S105). The date on which the product is purchased can be specified from the purchase date and time included in the information acquired from the purchase data 12c.

  Subsequently, the totaling unit 13a acquires one recommendation condition from the recommendation condition data 12e for a product having the same product ID as the product ID included in the information acquired from the purchase data 12c (step S106). Here, when the recommendation condition can be acquired (No at Step S107), the tabulation unit 13a acquires whether the recommendation condition is satisfied from the information acquired at Step S104 and Step S105 and the purchase data 12c. A determination is made based on the value of the area included in the information (step S108).

  If it is determined that it is satisfied (Yes at Step S109), 1 is added to the number of satisfaction of the recommended condition (Step S110), and the processing procedure is restarted from Step S106. In this way, it is determined whether or not all the recommended conditions are satisfied. If all the recommended conditions have been acquired in step S106 (Yes in step S107), the processing procedure is restarted from step S102.

  Then, when all the information on the target period has been acquired from the purchase data 12c and there is no information to be acquired in step S102 (Yes in step S103), the counting unit 13a ends the satisfaction count update process.

  For example, when the information on the first line of the purchase data 12c shown in FIG. 5 is acquired in step S102, the information on the first line of the user data 12b shown in FIG. 4 is acquired in step S104. In S105, the weather and temperature in Tokyo on September 9, 2008 are acquired. In this case, in step S106, each recommendation condition of the product with the product ID “AAA” shown in FIG. 7 is acquired, and in step S108, the following determination is made.

“Gender” is “female”: satisfied “age” is “10” generation: not satisfied “age” is “20” generation: satisfied “age” is “30” Being a teenager: Not satisfied “Address” is “Tokyo”: Satisfied “Address” is “Osaka”: Not satisfied “Address” is “Kanagawa”: Not satisfied “Address” Is "Aichi": not satisfied

  FIG. 10 is a diagram illustrating a processing procedure of weight evaluation processing by the weight evaluation unit 13b. The weight evaluation process is periodically executed. As illustrated in FIG. 10, the weight evaluation unit 13b first determines a weight evaluation criterion for each type of the recommendation condition data 12e (step S201).

  Some types of merchandise are sold in large numbers, while other types have a relatively small number of sales. For this reason, when the weight is simply evaluated based on the number of fulfillment, the weight of the recommendation condition of the type to be sold is greatly evaluated, and the product of the type having a relatively small number of sales is difficult to be selected as the recommended product. Therefore, the weight evaluation unit 13b determines, for each product type, an evaluation criterion as to which weight is given when the number of times of satisfaction is high.

  Subsequently, the weight evaluation unit 13b acquires one recommendation condition from the recommendation condition data 12e (step S202). If the recommendation condition can be acquired (No at Step S203), the weight is set according to the evaluation criterion for each type (Step S204), and the processing procedure is restarted from Step S202.

  If all the recommended conditions have been acquired from the recommended condition data 12e and there are no recommended conditions to be acquired in step S202 (Yes in step S203), the weight evaluation unit 13b ends the weight evaluation process.

  FIG. 11 is a diagram illustrating a processing procedure of recommended product selection processing by the recommended product selection unit 13c. The recommended product selection process is executed in accordance with a request from the online shopping server 11 when it is necessary to display the recommended product on the shopping screen.

  As illustrated in FIG. 11, the recommended product selection unit 13 c first acquires the user ID and position information of the user of the terminal device 2 that displays the shopping screen from the online shopping server 11 (step S <b> 301). Then, the recommended product selection unit 13c acquires user attribute information corresponding to the user ID from the user data 12b (step S302), and acquires current weather information of the current position of the user from the weather data 12d. (Step S303).

  In addition, when a user's terminal device cannot acquire a user's positional information for the reason of not having GPS, the user's attribute information corresponding to an address is acquired from the user data 12b.

  Subsequently, the recommended product selection unit 13c acquires information related to one product from the recommended condition data 12e (step S304). If the information can be acquired (No at Step S305), the recommended product selection unit 13c satisfies the information acquired at Steps S301 to S303 among the recommended conditions included in the acquired information. The total value of the corresponding weights is calculated (step S306).

For example, when the attribute information acquired in step S302 is information on the first line of the user data 12b shown in FIG. 4, the product with the product ID “AAA” shown in FIG.
The recommendation condition that “sex” is “female”, “age” is “20” generation, and “address” is “Tokyo” is satisfied, and the total weight becomes 13 (5 + 4 + 4). Further, when the user ID is the same, the product with the product ID “BBB” shown in FIG.
“Gender” is “female” “Age” is “20s” Recommended condition that “Address” is “Tokyo” is satisfied, but the total weight is 5 + 2 + 1 .

  Thereafter, the recommended product selection unit 13c restarts the processing procedure from step S304. If the information of all the products has been acquired from the recommended condition data 12e and there is no information to be acquired in step S304 (Yes in step S305), the recommended product selection unit 13c increases the order of the total weight value. A predetermined number of products are selected, and the selection result is notified to the online shopping server 11 (step S307).

  FIG. 12 is a diagram illustrating a processing procedure of sorting processing by the sorting unit 13d. The sort process is executed in accordance with a request from the online shopping server 11 when it is necessary to sort a list of products displayed on the shopping screen.

  As shown in FIG. 12, the sorting unit 13d first acquires the user ID and position information of the user of the terminal device 2 that displays the shopping screen from the online shopping server 11 (step S401). Further, the sorting unit 13d acquires a list of product IDs of products to be sorted from the online shopping server 11 (step S402).

  Then, the sorting unit 13d acquires user attribute information corresponding to the user ID from the user data 12b (step S403), and acquires current weather information of the current position of the user from the weather data 12d (step S403). S404). In addition, when a user's terminal device cannot acquire a user's positional information for the reason of not having GPS etc., the user's attribute information corresponding to an address is acquired from the user data 12b.

  Subsequently, the sorting unit 13d selects one product ID from the list of product IDs (step S405). If the product ID can be acquired (No at Step S406), the sorting unit 13d acquires information corresponding to the product ID from the recommendation condition data 12e (Step S407). Then, the sorting unit 13d calculates the total value of the weights corresponding to those satisfied by the information acquired in steps S401 to S404 among the recommended conditions included in the acquired information (step S408).

  After that, the sorting unit 13d restarts the processing procedure from step S405. If all the product IDs have been acquired from the product ID list and there are no product IDs to be acquired in step S405 (Yes in step S406), the sorting unit 13d determines that the product ID list has the largest total weight. Are sorted, and the sorting result is notified to the online shopping server 11 (step S409).

  As described above, in this embodiment, the recommendation condition is set based on the user's attribute and the like, and the importance evaluation of each recommendation condition is changed according to the sales situation. Appropriate products can be recommended even for those who have never used the system, and the recommendation conditions are automatically optimized, thereby reducing the man-hours required for the administrator to set the recommendation conditions.

  Various configurations of the online shopping system described above can be variously changed without departing from the spirit of the present invention. For example, in the above-described embodiment, an example in which the online shopping system is configured by combining three devices has been shown. However, these devices are integrated to form an online shopping system with one or two devices. It is good as well.

  The function of the recommended product selection device 13 may be realized by software. In this case, the recommended product selection device 13 is configured as a computer having hardware resources such as a processor and a memory. The recommended product selection program for realizing the function of the recommended product selection device 13 described above includes a hard disk device in the recommended product selection device 13, a storage medium readable by the recommended product selection device 13, a recommended product selection device 13, and the like. It is stored in another network-connected computer or the like, read out by the processor of the recommended product selection device 13, and developed in the memory of the recommended product selection device 13, so that it becomes an execution state and exhibits its function. .

  In the above embodiment, the recommended product selection device 13 is used to select the recommended product posted on the shopping screen. However, the recommendation product posted on the e-mail distributed to the user of the online shopping system. The recommended product selection device 13 may be used for other purposes such as selecting a product.

  The function of selecting recommended products by the recommended product selection device 13 described in the above embodiment is applied to a product search device that searches for products that match the type (category) selected by the user on the shopping screen. You can also. Specifically, the product search device to which the function of selecting recommended products is applied, among the products that match the type selected by the user, the importance of the recommendation condition that matches the user attribute information is weighted. Evaluate and search for products.

It is a figure which shows the structure of the online shopping system which concerns on a present Example. It is a block diagram which shows the structure of a database apparatus. It is a figure which shows an example of merchandise data. It is a figure which shows an example of user data. It is a figure which shows an example of purchase data. It is a figure which shows an example of weather data. It is a figure which shows an example of recommendation condition data. It is a block diagram which shows the structure of a recommended goods selection apparatus. It is a figure which shows the process sequence of the frequency | count fullness update process by a total part. It is a figure which shows the process sequence of the weight evaluation process by a weight evaluation part. It is a figure which shows the process sequence of the recommended goods selection process by a recommended goods selection part. It is a figure which shows the process sequence of the sort process by a sort part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Online shopping system 11 Online shopping server 12 Database apparatus 12a Product data 12b User data 12c Purchase data 12d Weather data 12e Recommended condition data 13 Recommended product selection apparatus 13a Aggregation part 13b Weight evaluation part 13c Recommended product selection part 13d Sort part 2a- 2n terminal device 3 network

Claims (9)

As a recommendation condition for determining whether or not a product is a recommended product, a database that stores a plurality of conditions related to user attribute information for each product and the attribute information of the user who purchased the product are collated. A counting means for counting the number of times the recommended condition matches the attribute information of the user who purchased the product for each recommended condition;
A weight evaluation unit that determines a weight indicating the importance of each recommendation condition based on the totaling result of the totaling unit;
A recommended product selection device comprising: recommended product selection means for selecting a recommended product by evaluating the importance of a recommendation condition that matches the attribute information of a specified customer with the weight.   The recommended product selection according to claim 1, further comprising: a sorting unit that sorts the designated products by evaluating the importance of the recommended condition that matches the attribute information of the designated customer by the weight. apparatus. The database further stores conditions for the weather for each product as a recommendation condition for determining whether or not the product is a recommended product.
The counting means further collates the weather information when the user purchases the product with the database, and determines the number of times the weather information when the user purchases the product matches the recommended condition for each recommended condition. Tally and
The recommended product selection device according to claim 1, wherein the recommended product selection unit further selects the recommended product by further evaluating the importance of the recommended condition that matches the current weather information with the weight. The database further stores a condition regarding the current position for each product as a recommendation condition for determining whether or not the product is a recommended product,
The counting means further collates the current position when the user purchases the product with the database, and determines the number of times the current position when the user purchased the product matches the recommended condition for each recommended condition. Tally and
The said recommended product selection means selects the recommended product by further evaluating the importance of the recommended condition that matches the current position of the user with the weight. Recommended product selection device. As a recommendation condition for determining whether or not a product is a recommended product, a database that stores a plurality of conditions related to user attribute information for each product and the attribute information of the user who purchased the product are collated. A counting procedure for counting the number of times the recommended condition matches the attribute information of the user who purchased the product for each recommended condition;
A weight evaluation procedure for determining a weight indicating the importance of each recommendation condition based on the counting result of the counting procedure;
A recommended product selection program for causing a computer to execute a recommended product selection procedure for selecting a recommended product by evaluating the importance of recommended conditions that match specified customer attribute information with the weights.   6. The recommendation according to claim 5, further comprising: causing the computer to further execute a sorting procedure for sorting the designated products by evaluating the importance of the recommendation condition that matches the attribute information of the designated customer by the weight. Product selection program. The database further stores conditions for the weather for each product as a recommendation condition for determining whether or not the product is a recommended product.
The tabulation procedure further compares the weather information when the user purchases the product with the database, and determines the number of times the weather information when the user purchases the product matches the recommended condition for each recommended condition. Tally and
The recommended product selection program according to claim 5 or 6, wherein the recommended product selection procedure further selects the recommended product by further evaluating the importance of the recommended condition that matches the current weather information with the weight. The database further stores a condition regarding the current position for each product as a recommendation condition for determining whether or not the product is a recommended product,
The counting procedure further compares the current position when the user purchases the product with the database, and determines the number of times the current position when the user purchased the product matches the recommended condition for each recommended condition. Tally and
The said recommended goods selection procedure further evaluates the importance of the recommendation conditions corresponding to a user's present position with the said weight, and selects a recommended goods, The said any one of Claims 5-7 characterized by the above-mentioned. Recommended product selection program. As a product search index, a database that stores a plurality of conditions relating to user attribute information for each product is compared with the attribute information of the user who purchased the product, and the condition and the attribute information of the user who purchased the product And a counting means for counting the number of times that matches with each condition,
A weight evaluation unit that determines a weight indicating the importance of the condition based on a totaling result of the totaling unit;
A product search apparatus comprising: recommended product selection means for selecting a search result product by evaluating the importance of the condition matching the specified customer attribute information with the weight.

Images