Home Delivery Product Recommendation Device and Home Delivery Product Recommendation Program

Abstract

To further improve personalized recommendation technology, and thereby improve customer convenience and customer unit price, etc. 【Solution means】A home delivery product recommendation device for a product with a predetermined home delivery schedule, comprising: a current date and time acquisition means for acquiring information on the current date and time; an order date and time acquisition means for acquiring information on the order start date and time and the order closing date and time of the product; an order date and time acquisition means for acquiring information on the order start date and time and the order closing date and time of the product; a time interval calculation means for calculating the time interval between the current date and time and the order closing date and time; a product recommendation means for recommending a recommended product based on a plurality of past purchase histories of the user's product and the current date and time; and a display amount calculation means for calculating the number or ratio of the recommended products by the product recommendation means to be displayed in the recommended product display area according to the time interval.

Description

Technical Field

[0001] The present invention relates to a home delivery product recommendation device and a home delivery product recommendation program.

Background Art

[0002] As one of the recommendation technologies in online shopping and EC (electronic commerce) sites, personalized recommendation that analyzes a user's past purchase history and browsing history by an algorithm and recommends products that the user is likely to buy for the purpose of improving customer convenience and customer unit price is known (see, for example, Patent Documents 1 and 2).

[0003] Non-Patent Document 1 also describes a recommendation engine having a plurality of respective features. For example, in an EC site case of a regular home delivery service for organic vegetables, new products are recommended by a novelty engine according to the customer's preference, and then re-purchase is recommended by a repeatability engine. If a purchase is also made at this timing, the purchase interval is grasped, and the periodicity engine recommends at a timing considered to be optimal for each customer, and measures for leading to regular product purchases are described.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Non-Patent Documents

[0005]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] Personalized recommendation is a technology for predicting customer demand, i.e., "who, when, and what is likely to be purchased", and recommends products to customers at an effective timing. And continuous improvement and enhancement are required regarding its effectiveness.

[0007] The present invention is proposed in view of the above points, and in one aspect, it aims to further improve personalized recommendation technology, and thereby improve customer convenience and customer unit price.

Means for Solving the Problem

[0008] To solve the above problems, a home delivery product recommendation device according to the present invention is a home delivery product recommendation device for products with a predetermined order home delivery schedule, and includes a current date and time acquisition means for acquiring information on the current date and time, an order date and time acquisition means for acquiring information on the order start date and time and the order closing date and time of the product, a time interval calculation means for calculating the time interval between the current date and time and the order closing date and time, and based on a plurality of past purchase histories of the user for the product and the current date and time, Among the said products, those that are in the purchase cycle for the said user recommended products Determination product recommendation means for The shorter the said time interval is, the more the number or proportion of the said recommended products to be displayed in the recommended product display area is calculated display amount calculation means.

Effect of the Invention

[0009] According to an embodiment of the present invention, in one aspect, it is possible to further improve personalized recommendation technology, and thereby improve customer convenience and customer unit price.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7A

Figure 7B

Figure 7C

Figure 8A

Figure 8B

Figure 9

Embodiments for Carrying Out the Invention

[0011] Embodiments of the present invention will be described in detail with reference to the drawings. <System Configuration> FIG. 1 is a diagram showing a configuration example of a regular home delivery product recommendation system 100 according to the present embodiment. The regular home delivery product recommendation system 100 in FIG. 1 includes a home delivery online shopping server 10, a recommendation machine 20, various DBs (databases) 30, and a user terminal 40, and is connected via a network 50.

[0012] The home delivery online shopping server 10 is a web server that provides an EC site for an operator who operates a home delivery service for, for example, fresh food and daily necessities. Users (members) access and log in to the EC site using the user terminal 40 to select and order home delivery products. On the EC site, home delivery products are categorized and displayed in a predetermined product display column, and a recommended product display area where recommended products (suggested products) determined by the recommendation machine 20 are displayed is provided.

[0013] The recommendation machine 20 has a plurality of recommendation engines described later, and is a recommendation device that analyzes the past purchase history of each user, performs calculations, and outputs products that the user is likely to buy. The output recommended products are displayed in the recommended product display area for the user on the EC site.

[0014] The user terminal 40 is, for example, a PC (personal computer), a smartphone, a tablet terminal, etc., and is a terminal of a member user who uses the home delivery service. A predetermined application program, a general-purpose web browser, etc. for accessing and logging in to the EC site of the home delivery online shopping server 10 are installed in advance in the user terminal 40. The user accesses and logs in to the EC site using the user terminal 40 to select and order (purchase) desired products. The ordered products are delivered to the delivery destination such as the user's home according to the delivery schedule.

[0015] (Functional configuration) FIG. 2 is a diagram showing a functional configuration example of the recommendation machine 20 according to the present embodiment. The recommendation machine 20 mainly includes a current date and time acquisition unit 101, an order date and time acquisition unit 102, a time interval calculation unit 103, a product recommendation unit 104, and a display amount calculation unit 105.

[0016] The current date and time acquisition unit 101 has a function of acquiring information on the current date and time.

[0017] The order date / time acquisition unit 102 has a function of acquiring information on the order start date / time and the order deadline date / time of a product for which a predetermined order delivery schedule is determined.

[0018] The time interval calculation unit 103 has a function of calculating the time interval between the current date / time and the order deadline date / time (i.e., the remaining time until the order deadline date / time) based on the current date / time and the order deadline date / time.

[0019] The product recommendation unit 104 has a function of recommending recommended products based on a plurality of past purchase histories of the user's products and the current date / time. The product recommendation unit 104 corresponds to a recommendation engine including a novelty engine, a repeatability engine, a periodicity engine, and a preference engine. Among them, in particular, the engine that has a function of recommending recommended products based on a plurality of past purchase histories of the user's products and the current date / time corresponds to the periodicity engine.

[0020] The display quantity calculation unit 105 has a function of calculating the number or ratio of recommended products to be displayed in the recommended product display area according to the time interval between the current date / time and the order deadline date / time (the remaining time until the order deadline date / time).

[0021] Note that the recommendation machine 20 can be implemented by a general-purpose computer. Specifically, the recommendation machine 20 includes hardware such as an arithmetic processing unit (CPU, etc.), a memory, an input / output interface, and a communication interface. The functions of the recommendation machine 20 are realized by the arithmetic processing unit executing processing according to a computer program stored in the memory. That is, each functional unit is realized by a computer program executed on the hardware resources such as the arithmetic processing unit and the memory of the computer constituting the recommendation machine 20. Also, these functional units may be read as "means", "module", "unit", or "circuit". Further, the functional units may be partially arranged in the memory of the recommendation machine 20 or an external storage device on the network. Also, each functional unit of the recommendation machine 20 is not only realized by a single server device, but may also be realized as a system composed of a plurality of devices with functions distributed. Also, the above computer program may be stored in a computer-readable storage medium.

[0022] (Database) The DB 30 according to this embodiment includes a user DB, a product DB, a purchase history DB, and order delivery schedule information.

[0023] The user DB is a DB in which user information of members who use the delivery service is registered. For example, in addition to user information such as the user's name, age, address, family, and hobbies, it includes the login ID and login password of the EC site in the delivery online shopping server.

[0024] The product DB is a DB (product master) in which products sold as delivery products are registered. The product information includes, for example, a product code, a product name, a JAN code, a product category, a price, a content volume, an inventory quantity, a producer, etc. The products to be recommended are selected from the products registered in the product DB.

[0025] The purchase history DB is a DB in which the history of products purchased by the user in the past is recorded. It includes at least purchase history information such as, for example, the product code, product name, purchase date and time, purchase price, and purchase quantity for each user.

[0026] Figure 3 is a diagram for explaining the order home delivery schedule information according to the present embodiment. The order home delivery schedule information is information regarding the period during which a product can be ordered and the delivery date. In the case of a general regular home delivery service, a period during which an order can be placed is determined, and orders are placed during a fixed period every week, and the product is delivered on a fixed day of the week.

[0027] For example, in the case of a weekly one-term regular home delivery service shown in Figure 3, during the orderable period from 12:00 am (order start date and time) on Tuesday of each week to 12:00 pm (order closing date and time) on Sunday, the user can place an order for a product. At the order closing date and time, the orders for that week are closed, and the ordered items are home-delivered, for example, on Wednesday of the following week. Needless to say, the order home delivery schedule information may be determined for each region based on the user's address. Also, it may be two or more terms per week.

[0028] (Recommendation engine) Figure 4 is a diagram (part 1) for explaining the recommendation engine according to the present embodiment. As shown in Figure 4, the recommendation machine 20 according to the present embodiment has a recommendation engine having a plurality of different algorithms including, for example, a novelty engine, a repeatability engine, a periodicity engine, and a preference engine.

[0029] · The novelty engine is a recommendation engine for promoting first-time purchases. The novelty engine has a function of recommending new products that the user has no purchase experience with, such as new products and promotional products, and contributes to the expansion of the purchase items.

[0030] · The repeatability engine is a recommendation engine for promoting second-time purchases. The repeatability engine has a function of recommending products that the user has purchased once in the past, and contributes to an increase in the purchase frequency.

[0031] · The periodic engine is a recommendation engine for promoting third or more purchases. The periodic engine has a function of recommending products that the user has purchased two or more times in the past, contributing to an improvement in the purchase frequency. Based on the purchase history of products repeatedly purchased in the past, the periodic engine grasps the purchase periodicity (purchase interval) of the product, and when the purchase cycle (for example, a cycle week) arrives, recommends the product to promote repeat purchases. The product recommendation by the periodic engine can also be said to be a function to prevent forgetting to buy periodic products.

[0032] · The preference engine is a recommendation engine for promoting the purchase of products that match the preferences of the user. In the case of the preference engine, products including new products and experience products can be target products for recommendation. For example, based on information such as preferences, age, family, etc. in the user information, and / or preference information based on the purchase history of the user, new products (scenarios where you want to promote first-time purchases) and experience products (scenarios where you want to promote purchases for the second time or more) that are matched can be recommended alone or in combination with other novelty engines or periodic engines.

[0033] Figure 5 is a diagram (part 2) for explaining the recommendation engine according to this embodiment. The target products for recommendation in the novelty engine are new products, and the target products for recommendation in the repeat engine and the periodic engine are experience products. However, the recommendation machine 20 calculates the score of each product's likelihood to be bought and the priority based on the score for each engine from among a plurality of target products for recommendation, and recommends the products with higher priority to be displayed with higher priority based on the calculation results shown in Figure 5 (each recommendation priority list for each recommendation engine). The score is calculated comprehensively by incorporating various parameters such as, for example, the user's preferences (preference engine), as well as the market sales performance (popularity) of the product, the time elapsed since release, staple products, seasonality, inventory level, promotion level (weight value indicating the degree that the home delivery service provider particularly wants to prioritize sales).

[0034] FIG. 6 is a diagram (Part 3) for explaining the recommendation engine according to the present embodiment. The recommendation process of the recommendation engine 20 will be described with the following model cases.

[0035] Step S1: The recommendation engine 20 recommends product A, which is a new product, based on the novelty engine (or preference engine). At this time, it is assumed that the user did not purchase the new product (n-th week).

[0036] Step S2: The recommendation engine 20 recommends product A again based on the novelty engine (or preference engine). At this time, it is assumed that the user purchased the new product (n + 1-th week). The information of the purchased product A is recorded as the first purchase history in the purchase history DB together with the purchase date and time.

[0037] Step S3: The recommendation engine 20 recommends product A from among the experience products (purchased only once in the past) based on the repeatability engine. At this time, it is assumed that the user did not purchase the new product (n + 2-th week).

[0038] Step S4: The recommendation engine 20 recommends product A from among the experience products (purchased only once in the past) based on the repeatability engine. At this time, it is assumed that the user purchased the new product (n + 3-th week). The information of the purchased product A is recorded as the second time in the purchase history DB together with the purchase date and time. Also, based on the purchase history DB of the first purchase date and time (n + 1-th week) and the second purchase date and time (n + 3-th week), the recommendation engine 20 determines that the purchase interval (purchase cycle) of product A for the user is two weeks, and associates and holds the information of the user, product A, and purchase cycle.

[0039] Step S5: Based on the periodic engine, the recommendation engine 20 recommends product A whose purchase cycle corresponds to the current week (this week) from among the experience goods (purchased two or more times in the past) (the (n + 5)th week). Note that the recommendation engine 20 did not recommend product A that did not correspond to the purchase cycle in the (n + 4)th week. In the (n + 4)th week, the recommendation engine 20 may recommend some other product, such as another product whose purchase cycle corresponds to that week.

[0040] Step S6: Based on the periodic engine, the recommendation engine 20 recommends product A whose purchase cycle corresponds to the current week (this week) from among the experience goods (purchased two or more times in the past) (the (n + 7)th week). In this way, the periodic engine recommends product A at a periodic timing when it is highly likely that the user will purchase product A, thereby associating product A with regular purchases.

[0041] <Home delivery online shopping EC site> The recommended products according to this embodiment are dynamically recommended and displayed on the EC site of home delivery online shopping according to the remaining time until the order deadline, that is, the current date and time and the order deadline. Hereinafter, it will be described in detail with reference to an example of the EC site screen. The EC site screen is generated by the home delivery online shopping server 10 and displayed on the screen of the user terminal 40.

[0042] (EC site screen immediately after the order start date and time) FIG. 7A is a diagram for explaining the EC site screen (immediately after the order start date and time) 1 according to this embodiment. The top screen of the EC site immediately after the order start date and time shown in FIG. 7A has, for example, the current logged-in user 401, the current date and time 402, the order home delivery schedule 403 in the current term, the content 404, and the recommended product display area 405.

[0043] Here, according to the order delivery schedule information or the order delivery schedule 403 shown in FIG. 3, the user can place an order during the orderable period from 12:00 am (order start date and time) on Tuesday every week to 12:00 pm (order deadline date and time) on Sunday. Also, since the current date and time 402 is "2024 / 10 / 1 (Tue) 14:00", it is immediately after the order start date and time, and there is still sufficient time allowance until the order deadline date and time.

[0044] The recommended product display area 405 shown in FIG. 7A includes a first recommended product display area 405a, a second recommended product display area 405b, and a third recommended product display area 405c. Also, in the first recommended product display area 405a, recommended products recommended by the novelty engine are displayed, in the second recommended product display area 405b, recommended products recommended by the preference engine are displayed, and in the third recommended product display area 405c, recommended products recommended by the repeatability engine are displayed.

[0045] FIG. 7B is a diagram for explaining the EC site screen (immediately after the order start date and time) 2 according to the present embodiment. Compared with FIG. 7A, in the recommended product display area 405 shown in FIG. 7B, recommended products recommended by the novelty engine, recommended products recommended by the preference engine, and recommended products recommended by the repeatability engine are displayed within one area. That is, the recommended product display area 405 may be provided with a recommended product display area divided for each engine according to the screen design specifications, etc., or a mixed recommended product display area that does not separate the recommended products from each engine may be provided.

[0046] FIG. 7C is a diagram for explaining the EC site screen (immediately after the order start date and time) 3 according to the present embodiment. The order cart screen of the EC site shown in FIG. 7C is a screen showing the products in the user's order cart (shopping basket). Also, in the order cart screen, a fourth recommended product display area 405d is provided, and recommended products recommended by the periodic engine are displayed in the fourth recommended product display area 405d. As described above, the periodic engine recommends the product when the period (for example, a weekly period) arrives, thereby promoting repeat purchases. In particular, in the order cart screen which is the final confirmation screen (immediately before purchase), by making a final push recommendation of the products that the user has forgotten to buy but are not yet in the order cart, it is to improve the further purchase frequency and unit price.

[0047] From the viewpoint of the effect of improving the purchase frequency and unit price, although it is desirable that the recommended product display area 405d recommended by the periodic engine is provided on the order cart screen as one method, it is not necessarily limited to being displayed only on the order cart screen. Considering comprehensively the screen space constraints and the display balance of the recommended product display areas by other engines, it may be displayed on the top screen or other screens.

[0048] (EC site screen immediately before the order deadline) FIG. 8A is a diagram for explaining the EC site screen (immediately before the order deadline) 1 according to the present embodiment. In the EC site screen shown in FIG. 8A, since the current date and time 402 is "2024 / 10 / 6 (Sun) 10:00", there is little time remaining until the order deadline. That is, the EC site screen shown in FIG. 8A is the top screen of the EC site immediately before the order deadline.

[0049] Immediately after the order start date and time and until the order deadline date and time, there is still sufficient time allowance. Comparing with the top screen of FIG. 7A, in the top screen of FIG. 8A, in the recommended product display area 405a, the recommended products recommended by the novelty engine are switched to the recommended products recommended by the periodicity engine. That is, in the top screen immediately before the order deadline date and time, many recommended products recommended by the periodicity engine are displayed.

[0050] FIG. 8B is a diagram for explaining the EC site screen (immediately before the order deadline date and time) 2 according to the present embodiment. In the EC site screen shown in FIG. 8B, since the current date and time 402 is "2024 / 10 / 6 (Sun) 10:00", there is little time allowance until the order deadline date and time.

[0051] Immediately after the order start date and time and until the order deadline date and time, there is still sufficient time allowance. Comparing with the top screen of FIG. 7B, in the top screen of FIG. 8B, more recommended products recommended by the periodicity engine are displayed in the recommended product display area 405.

[0052] Thus, according to the present embodiment, on the screen of the EC site, the display ratio of the recommended products recommended by the periodicity engine is higher than that of the recommended products recommended by other engines according to the remaining time amount until the order deadline date and time, that is, the time distance between the current date and time and the order deadline date and time.

[0053] As described above, the periodicity engine promotes repeat purchases by recommending the product when the period (for example, the weekly period) arrives. In particular, in the order cart screen, which is the final confirmation screen (screen immediately before purchase), by recommending the products that the user may forget to buy, which are not yet in the order cart, one more time on the last screen, it is possible to further improve the purchase frequency and unit price.

[0054] Furthermore, in this embodiment, paying attention to the usefulness of the periodic engine itself that makes a final push recommendation for the user's forgotten-cycle products and its recommendation timing, especially at the point immediately before the order deadline time, by making a final push recommendation for the forgotten-cycle products that are not yet in the order cart, it is possible to further increase the purchase frequency and unit price.

[0055] In fact, according to the result comparison of the AB test using the actual e-commerce site by the applicant, by actively controlling the display of the forgotten-cycle products (recommended products by the periodic engine) that are not yet in the order cart at the last possible timing close to and immediately before the order deadline time, an effect of pushing up the purchase frequency and unit price was recognized compared to the case where no active display control was performed.

[0056] That being said, regardless of whether it is close to or immediately before the order deadline time, simply actively displaying the forgotten-cycle products that are not yet in the order cart from the beginning can increase the purchase frequency and unit price of the forgotten-cycle products themselves. However, under the constraints of the recommended product display area space or the number of recommended product displays on the screen, conversely, as the display of the recommended products by the novelty engine and the repeat engine decreases accordingly, the purchase frequency and unit price will be pushed down, which will interfere with the purchase suitability cycle of the model case shown in FIG. 6. That is, overall, in the long term, it will be impossible to expect an increase in the purchase frequency and unit price.

[0057] <Recommended Product Display Control Process> FIG. 9 is a flowchart showing the recommended product display control process according to the present embodiment. This flowchart is executed, for example, when a user accesses and logs in to an EC site, and the delivery online shopping server 10 receives a display request for an EC site screen from the user terminal 40, and the recommendation machine 20 receives a request for obtaining recommended product information from the delivery online shopping server 10. Also, by reading and executing a program that can be realized by the arithmetic processing unit of the recommendation machine 20, the following steps (hereinafter referred to as "S") can be realized.

[0058] S1: The recommendation machine 20 obtains information on the current date and time from a clock (timekeeping means).

[0059] S2: The recommendation machine 20 obtains information on the start date and time and the deadline date and time of the current order from the order delivery schedule information.

[0060] S3: The recommendation machine 20 determines whether the current date and time is a time between the start date and time and the deadline date and time of the current order, thereby determining whether the current date and time is an orderable time. If it is determined that the current date and time is an orderable time, the process proceeds to S4. If it is determined that the current date and time is not an orderable time, since it is outside the orderable time, the process proceeds to END.

[0061] S4: The recommendation machine 20 calculates the time interval (remaining time) until the deadline date and time from the current date and time and the start date and time of the current order. Note that the remaining time until the deadline date and time may also be referred to as the time distance between the current date and time and the deadline date and time.

[0062] S5: Based on the time interval (remaining time) until the order deadline time, when the remaining time until the order deadline time is short (when the time distance between the current date and time and the order deadline time is small), compared with the case where the remaining time until the order deadline time is long (when the time distance between the current date and time and the order deadline time is large), the recommendation machine 20 calculates the display ratio of recommended products such that the display ratio of recommended products recommended by the periodic engine is higher than that of recommended products recommended by other engines. Note that the recommended products recommended by the periodic engine are products determined to be in the purchase cycle (for example, the cycle week) based on the current date and time.

[0063] (Calculation example 1) When dividing the orderable period into the first half and the second half at a predetermined date and time Order start date and time: 2024 / 10 / 1 12:00 Order deadline date and time: 2024 / 10 / 6 12:00 Orderable time: 120H Number of recommended product displays: 14 · First half Current date and time: 2024 / 10 / 1 12:00 Remaining time until the order deadline time: 120H Recommended products of the novelty engine: 4 pieces Recommended products of the preference engine: 3 pieces Recommended products of the repeatability engine: 3 pieces Recommended products of the periodic engine: 4 pieces Display ratio of recommended products of the periodic engine: 4 / 14 · Second half (within 24 hours until the order deadline time) Current date and time: 2024 / 10 / 6 12:00 Remaining time until the order deadline time: 24H Recommended products of the novelty engine: 0 pieces Recommended products of the preference engine: 3 pieces Recommended products of the repeatability engine: 3 pieces Recommended products of the periodic engine: 8 pieces Display ratio of recommended products of the periodic engine: 8 / 1

[0064] (Calculation Example 2) When dividing the orderable period into multiple intervals at a predetermined date and time Order start date and time: 2024 / 10 / 1 12:00 Order deadline date and time: 2024 / 10 / 6 12:00 Orderable time: 120H Number of recommended products to be displayed: 14 · First period Current date and time: 2024 / 10 / 1 12:00 Remaining time until the order deadline date and time: 120H Recommended products from the novelty engine: 4 Recommended products from the preference engine: 3 Recommended products from the repeatability engine: 3 Recommended products from the periodicity engine: 4 Display ratio of recommended products from the periodicity engine: 4 / 14 · Second period (within 48 hours remaining until the order deadline date and time) Current date and time: 2024 / 10 / 5 12:00 Remaining time until the order deadline date and time: 48H Recommended products from the novelty engine: 3 Recommended products from the preference engine: 2 Recommended products from the repeatability engine: 2 Recommended products from the periodicity engine: 7 Display ratio of recommended products from the periodicity engine: 7 / 14 · Third period (within 24 hours remaining until the order deadline date and time) Current date and time: 2024 / 10 / 1 12:00 Remaining time until the order deadline date and time: 24H Recommended products from the novelty engine: 0 Recommended products from the preference engine: 2 Recommended products from the repeatability engine: 2 Recommended products from the periodicity engine: 10 Display ratio of recommended products from the periodicity engine: 8 / 14

[0065] Note that the recommendation machine 20 calculates the "display count" of recommended products such that, based on the remaining time until the order deadline, when the remaining time until the order deadline is short, the display count of the recommended products recommended by the periodic engine is larger than that of the recommended products recommended by other engines as compared to when the remaining time until the order deadline is long.

[0066] Needless to say, the above calculation examples 1 and 2 are illustrative explanations for facilitating the understanding of the process of S5. In actuality, the number of recommended products and the display ratio recommended by the periodic engine are dynamically calculated according to a given calculation formula for satisfying S5, based on the remaining time until the order deadline.

[0067] S6: The recommendation machine 20 obtains information on the recommended products to be displayed in the recommended product display area from the calculation results (for example, each recommended priority list in FIG. 5) according to the calculated display ratio, and transmits it to the delivery online shopping server 10. For example, the information on the recommended products includes the product ID, product name, product image, product price, display position indicating which recommended product display area to display in (including the priority order of display), and the like.

[0068] As described above, the recommendation machine 20 according to this embodiment increases the display ratio of recommended products recommended by the periodic engine compared to the display ratio of recommended products recommended by other engines when the remaining time until the order deadline is short, and even more so when the remaining time until the order deadline is long, in an online shopping EC site for home delivery. As a result, when the remaining time until the order deadline is short, by actively recommending the forgotten cycle products that have not yet been added to the order cart at the last possible timing close to or immediately before the order deadline time, it is possible to further improve the purchase frequency and unit price. That is, according to the regular home delivery product recommendation system 100 according to this embodiment, on one hand, it is possible to further improve the personalized recommendation technology, and as a result, improve customer convenience and customer unit price, etc.

[0069] The following points are also mentioned. · The periodic products according to this embodiment are different from the regular purchase products that the user reserves for purchase by specifying every week, every month, a specified date, etc. In the case of regular purchase products, when the purchase date specified by the user arrives, they are automatically ordered and purchased without the user's order cart operation, etc.

[0070] · In S6, the recommendation machine 20 can exclude the products already in the order cart from the recommended products so that the recommended products to be displayed in the recommended product display area do not overlap with the products already in the order cart by obtaining the product information in the order cart from the home delivery online shopping server 10. Alternatively, when the recommended products received from the recommendation machine 20 overlap with the products already in the order cart on the home delivery online shopping server 10 side, another set of recommended products may be obtained anew.

[0071] · The calculation process of recommended products by the recommendation machine 20 (for example, the recommended priority list creation process and its update process in FIG. 5) can be executed, for example, at the timing when the user accesses and logs in to the EC site, or at the timing when the home delivery online shopping server 10 receives a display request for the EC site screen from the user terminal 40.

[0072] · However, the calculation timing of recommended products is not limited to these. For example, from the perspective of reducing the load cost associated with increasing the calculation frequency, it may be earlier than the timing when the user accesses and logs in to the EC site or the timing when the home delivery online shopping server 10 receives a display request for the EC site screen from the user terminal 40. Also, for example, it may be at timing such as every hour, every day, every week, or every term. Even if it is earlier than the timing when the user accesses and logs in to the EC site or the timing when the home delivery online shopping server 10 receives a display request for the EC site screen from the user terminal 40, if there is a pre-calculated result (for example, the recommended priority list in FIG. 5) at that time, in S6, the recommendation machine 20 will obtain the number of products corresponding to the display ratio for each recommendation engine in priority order from the recommended priority list, and transmit it to the home delivery online shopping server 10 as information on the recommended products to be displayed in the recommended product display area. For example, in the above calculation example 1, from the recommended priority list, in the first half, a total of 4 products with priority ranks 1 to 4 by the periodic engine, and in the second half, a total of 8 products with priority ranks 1 to 8 by the periodic engine are obtained as information on the recommended products to be displayed in the recommended product display area.

[0073] Although the present invention has been described by showing specific examples according to a preferred embodiment of the present invention, it is obvious that various modifications and changes can be made to these specific examples without departing from the broad spirit and scope of the present invention defined in the claims. That is, the present invention should not be construed as being limited by the details of the specific examples and the attached drawings.

Description of Reference Numerals

[0074] 10 Delivery Online Shopping Server 20 Recommendation Machine 30 DB 40 User Terminal 50 Network 100 Delivery Product Recommendation System 101 Current Date and Time Acquisition Unit 102 Order Date and Time Acquisition Unit 103 Time Interval Calculation Unit (Remaining Time Calculation Unit) 104 Product Recommendation Unit 105 Display Quantity Calculation Unit

Claims

1. A home delivery product recommendation device for products with a specified order home delivery schedule, a current date and time acquisition means for acquiring information on the current date and time, an order date and time acquisition means for acquiring information on the order start date and time and the order deadline date and time of the product, a time interval calculation means for calculating the time interval between the current date and time and the order deadline date and time, a product recommendation means for determining, from among the products, recommended products that are in the purchase cycle for the user based on a plurality of past purchase histories of the user for the product and the current date and time, a display amount calculation means for calculating a larger number or ratio of the recommended products to be displayed in the recommended product display area as the time interval is shorter, A home delivery product recommendation device characterized by comprising the above.

2. A home delivery product recommendation device for products with a specified order home delivery schedule, a current date and time acquisition means for acquiring information on the current date and time, an order date and time acquisition means for acquiring information on the order start date and time and the order deadline date and time of the product, a time interval calculation means for calculating the time interval between the current date and time and the order deadline date and time, a product recommendation means for determining, from among the products, recommended products that are in the purchase cycle for the user based on a plurality of past purchase histories of the user for the product and the current date and time, a display amount calculation means for calculating a first number or first ratio of the recommended products to be displayed in the recommended product display area in the case of a first time interval, and calculating a second number larger than the first number or a second ratio larger than the first ratio of the recommended products to be displayed in the recommended product display area in the case of a second time interval shorter than the first time interval, A home delivery product recommendation device characterized by comprising the above.

3. A home delivery product recommendation program for causing a computer to function as a current date and time acquisition means for acquiring information on the current date and time, an order date and time acquisition means for acquiring information on the order start date and time and the order deadline date and time of a product with a specified order home delivery schedule, a time interval calculation means for calculating the time interval between the current date and time and the order deadline date and time, a product recommendation means for determining, from among the products, recommended products that are in the purchase cycle for the user based on a plurality of past purchase histories of the user for the product and the current date and time, a display amount calculation means for calculating a larger number or ratio of the recommended products to be displayed in the recommended product display area as the time interval is shorter. ​

4. A computer, a current date and time acquisition means for acquiring information on the current date and time, an order date and time acquisition means for acquiring information on the order start date and time and the order deadline date and time of a product for which a predetermined home delivery schedule is determined, a time interval calculation means for calculating the time interval between the current date and time and the order deadline date and time, a product recommendation means for determining a recommended product that is in the purchase cycle for the user from the product based on a plurality of past purchase histories of the product by the user and the current date and time, a display quantity calculation means for calculating a first number or a first ratio of the recommended products to be displayed in a recommended product display area in the case of a first time interval, and calculating a second number larger than the first number or a second ratio larger than the first ratio of the recommended products to be displayed in the recommended product display area in the case of a second time interval shorter than the first time interval, a home delivery product recommendation program for causing it to function.

Images