WO2017026852A1

WO2017026852A1 - Method for service of recommending product through relative comparison, and recommendation service apparatus and program therefor

Info

Publication number: WO2017026852A1
Application number: PCT/KR2016/008915
Authority: WO
Inventors: 정진호
Original assignee: 정진호
Priority date: 2015-08-13
Filing date: 2016-08-12
Publication date: 2017-02-16
Also published as: JP6087467B2; JP5992122B1; JP2017037629A; KR20170116924A; KR101866514B1; JP2017037650A; KR20170019944A

Abstract

A method for a service of recommending a product through relative comparison and an apparatus for the recommendation service are provided. The method for a service of recommending a product of the present invention comprises: a product ranking configuration step for configuring a ranking of a product for each user on the basis of selection of preferred products of a plurality of users through relative comparison of a plurality of products according to tastes of the users (S100); a similar taste user determination step for cross-comparing a ranking of a product configured by a specific user and a ranking of a product configured by at least one different user to determine another user who has a taste similar to that of the specific user (S200); and a product recommendation step for recommending a product to the specific user on the basis of the ranking of the product configured by another user having the taste similar to the taste of the specific user (S300).

Description

Product recommendation service method through relative comparison and recommendation service device and program therefor

The present invention relates to a product recommendation service method through a relative comparison, and a recommendation service apparatus and a program therefor, and more particularly, a user determines a ranking of a product through a relative comparison according to his or her taste, and determines a ranking of products among users. The present invention relates to a method of recommending a product based on a ranking of a product determined by a user of a similar taste through cross comparison.

Abundant information is shared on the Internet, and users select products by referring to evaluations or recommendations on products experienced or possessed by others.

For example, when a product is searched on a search service, evaluation or recommendation for various products is presented as a search result, and a user often selects a product suitable for himself or herself based on such information.

However, there is a lack of advertising information and the like, and the credibility of the evaluation and recommendation of the product is deteriorated. Furthermore, since each person has a different personality, the taste of the product is different according to the taste of the preferred product. However, there is a problem in that when selecting a product according to the recommendation, a product that is not actually suitable for one's taste is selected.

In addition, most of the methods of recommending products by displaying the evaluation of products by ratings such as the number of stars are applied. In the case of evaluation using the number of stars, a user evaluates the feeling of only one product in the absence of a comparison target. Even though the same user is evaluating the same product, there is a limitation that it is difficult to consistently evaluate the product because the different feelings are reflected differently, resulting in different results. .

The present invention is to solve the problems of the prior art as described above, to solve the problem that the reliability of the evaluation or recommendation for the goods presented on the Internet due to advertising information and the like.

In particular, each person has a different personality, and accordingly, the tastes of the preferred products are different accordingly, so trust in the evaluation or recommendation of the product is selected so that a product that is practically unsuitable to one's taste is selected. To solve.

Furthermore, in the case of evaluating products using the number of stars, there is a limitation that the consistent evaluation is difficult even though the same user evaluates the same product, and there is a problem in that comparison between the products having the same number of stars is impossible. To provide a recommendation service that can be evaluated comparatively.

In order to achieve the above technical problem, a product recommendation service method according to the present invention and a product for setting a ranking of products for each user based on a user's preferred product selection through relative comparison with respect to a plurality of products according to tastes of a plurality of users. A ranking setting step; A similar taste user determination step of judging a ranking of a product set by a specific user and a ranking of a product set by at least one other user to determine another user having a similar taste with the specific user; And a product recommendation step of recommending the product to the specific user based on a ranking of the product set by another user having a similar taste with the specific user.

Preferably, the product ranking setting step may include a product list generation step of generating a product list by receiving a plurality of product information from the user; And setting a ranking of the plurality of products based on the user's preferred product selection through relative comparison according to the user's taste with respect to the plurality of products included in the product list, and based on this, corresponding to the user A product ranking list generating step of generating a product ranking list may be included.

Here, in the product ranking list generating step, a ranking may be set by sequential ordering of the plurality of products.

Furthermore, the similar taste user determining step may include a product recommendation request step of receiving a product recommendation request from a specific user among the plurality of users; A similarity calculating step of comparing a product ranking list corresponding to the specific user with a product ranking list corresponding to one or more other users of the plurality of users and calculating a similarity; And a similar taste user selection step of determining one or more other users having similar tastes with the specific user among the one or more other users based on the similarity.

Preferably, the calculating of the similarity may include extracting a product ranking list corresponding to one or more other users that include a product included in the product ranking list corresponding to the specific user and at least a predetermined reference number, and the specific user. The similarity may be calculated by cross comparing the product ranking list corresponding to and the product ranking list corresponding to the extracted other users.

Here, a ranking is set in the product ranking list by sequential ordering of a plurality of products, and the similarity calculating step includes a product ranking list corresponding to the specific user and a product ranking list corresponding to the other user. For each, classifying a plurality of products into a plurality of groups according to a sequential sequence order; By comparing the group of the product ranking list corresponding to the specific user and the group of the product ranking list corresponding to the other user for each group, the similarity of the other user with respect to the specific user based on the number of the same product for each group It may include the step of calculating.

Preferably, the product recommendation step may include: a list extraction step of extracting a product ranking list corresponding to the specific user and a product ranking list corresponding to the selected one or more similar taste users; A new product extraction step of extracting a new product from a product ranking list corresponding to the similar taste user not included in the product ranking list corresponding to the specific user; A recommendation product determining step of determining a recommendation product based on a ranking for a new product set by the user of similar taste; And providing recommendation information including the recommendation product to the specific user.

As an example, the list extracting step extracts a product ranking list corresponding to a similar taste user having the highest similarity with respect to the specific user, and the recommendation product determining step includes a new product having the highest rank or a ranking up to a predetermined number. New products can be selected as recommended products in ascending order.

As another example, the list extracting step extracts a product ranking list corresponding to a user of similar taste up to a predetermined value of similarity with respect to the specific user, and the recommendation product determining step includes a corresponding product for each new product. A recommendation score may be calculated based on the similarity of the similar taste user and the ranking set by the similar taste user, and a new product may be selected as the recommended product based on the recommendation score.

As another example, a ranking score is pre-assigned to each ranking of the product ranking list, and in the recommendation product determining step, the similarity degree of the similar taste user is converted into a ratio such that the sum is 1, and thus, Based on the ratio and the ranking score, it is possible to calculate a recommendation score for a new product.

As another example, the product ranking list may be a product ranking based on the relative preference based on the user's subjectivity.

As another example, the calculating of the similarity may include: classifying a product ranking list corresponding to the specific user and a product ranking list corresponding to the other user into a plurality of groups according to a ranking of a specific range; And a group consisting of a product having a ranking corresponding to a specific range among the product ranking list corresponding to the specific user and a group having a product having a ranking corresponding to a specific range or a range different from the specific range among the product ranking list corresponding to another user. Comparing the cross-group by group, calculating the similarity of the other user with respect to the specific user based on the number of the same product for each group.

As another example, the product recommendation step may include a pre-verification and evaluation request step of requesting pre-verification and evaluation of the designated product by designating a product that the specific user wants to pre-verify and evaluate; A recommendation score calculating step of calculating a recommendation score for the designated product based on similarity with the similar taste user for the designated product and ranking information set for the designated product by the similar taste user; And providing recommendation information including the calculated recommendation score to the specific user.

In addition, the recommendation service apparatus according to the present invention, a ranking setting unit for setting a ranking for the product for each user based on the user's preferred product selection through a relative comparison of a plurality of products according to the taste of each of the plurality of users; A similar user determination unit which cross compares a ranking of a product set by a specific user with a ranking of a product set by at least one other user, and determines another user having a similar taste with the specific user; And a product recommendation unit recommending a product to the specific user based on a ranking of the product set by another user having a similar taste with the specific user.

Preferably, the ranking setting unit generates a plurality of product lists for each user, and sets the ranking of the product on the product list on the basis of the user's preferred product selection through relative comparison according to each user's taste, the user And generating a product ranking list for each user, and the similar user determination unit cross compares the product ranking list corresponding to the specific user who requested the product recommendation with the product ranking list corresponding to the other user and compares the similarity of the other user to the specific user. Calculating and determining a similar taste user for the specific user based on the similarity, and the product recommendation unit is not included in the product ranking list corresponding to the specific user in the product ranking list corresponding to the similar taste user; To extract new products, the similar taste The recommended product may be determined based on the ranking of the new product set by the user.

In addition, the recommendation service apparatus according to the present invention may be a device in which a computer program for performing each step of the product recommendation service method according to the present invention is recorded.

In addition, the program according to the present invention, a ranking setting unit for setting a ranking for a product for each user based on the user's preferred product selection through relative comparison of a plurality of products according to the taste of each of the plurality of users ; A similar user determination unit which cross compares a ranking of a product set by a specific user with a ranking of a product set by at least one other user, and determines another user having a similar taste with the specific user; And a product recommendation unit for recommending a product to the specific user based on a ranking of the product set by another user having a similar taste as that of the specific user.

According to the present invention as described above, by recommending a product based on the evaluation of the product set by other users with similar tastes to the user, a product recommendation that can satisfy the user can be made.

In addition, since the product is recommended by combining evaluations of a plurality of other users who have similar tastes with the user who requested the product recommendation, advertisement information may be excluded, thereby increasing the reliability of the recommendation.

Furthermore, by overcoming the limitations of simple evaluation of products using the number of stars, the user can select products more effectively by recommendation through comparative evaluation among multiple products.

1 illustrates an embodiment of a recommendation service system for providing a product recommendation service according to the present invention;

2 is a block diagram of an embodiment of a recommendation service apparatus for providing a product recommendation service according to the present invention;

3 is a schematic flowchart of an embodiment of a product recommendation service method through relative comparison according to the present invention;

4 shows an embodiment of a product ranking setting process in the recommended service method according to the present invention,

5 illustrates an embodiment of a similar taste user determination process in the recommended service method according to the present invention.

6 illustrates an embodiment of a product recommendation process in the recommendation service method according to the present invention;

7 illustrates an embodiment of recommended product information in the recommendation service method according to the present invention.

8 shows a product recommendation process in the recommendation service method according to an improvement of the embodiment of the present invention,

9 shows a product recommendation process in the recommendation service method according to an improvement of the embodiment of the present invention,

10 is a block diagram of an embodiment of a recommendation service apparatus for providing a product recommendation service according to an improvement of the embodiment of the present invention;

11 shows a product recommendation process carried out in the recommended service method according to an improvement of the embodiment of the present invention,

12A to 12C illustrate an example of providing and displaying a result of verification and evaluation by the product recommendation process of FIG. 11 to the user.

FIG. 13 illustrates an example of displaying the result of verification and evaluation by the product recommendation process of FIG. 11 on a map.

In order to explain the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, the following describes exemplary embodiments of the present invention and looks at it with reference.

First, the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention, and singular forms may include plural forms unless the context clearly indicates otherwise. Also in this application, terms such as "comprise" or "have" are intended to indicate that there is a feature, number, step, action, component, part, or combination thereof described on the specification, one or more other It is to be understood that the present invention does not exclude the possibility of the presence or the addition of features, numbers, steps, operations, components, parts, or a combination thereof.

In describing the present invention, when it is determined that the detailed description of the related well-known configuration or function may obscure the gist of the present invention, the detailed description thereof will be omitted.

According to the present invention, a user determines a ranking of a product through relative comparison according to his or her preference, and determines a user having a similar taste through a cross comparison of the ranking of the product among users based on the ranking of a product determined by a user of the similar taste. Suggest ways to recommend products.

In the present invention, the product applied to the recommendation service may be a shaped article or an intangible information, for example, may include various items such as a smartphone, a car, and may include a place such as a restaurant or a cafe. It may also include content such as movies and music.

1 illustrates an embodiment of a recommendation service system for providing a product recommendation service according to the present invention.

The recommendation service apparatus 100 holds ranking list information of a product determined by each user according to the taste of each of the plurality of

users

10a, 10b, ... 10n, and is determined based on the ranking list information of the product for each user. By identifying other users with similar tastes as the user, a product recommendation service is provided to the specific user based on the ranking information of the product determined by the other users with similar tastes.

A plurality of users (10a, 10b, ... 10n) can be connected to the recommendation service device 100 by the communication terminal owned by the user to receive the recommended service, such as a smartphone, PC, laptop, etc. Various communication terminals known as may be applied.

The user may be connected to the recommended service device 100 through the network 30 as a communication terminal owned by the user, where the network 30 may include both a wired network and a wireless network as a general Internet line. Furthermore, when the recommendation service according to the present invention is provided in a limited range of a specific company or the like, the network may be a virtual private network or an intranet network accessible only to a limited area or an authorized person.

Further, the recommendation service apparatus 100 may provide a recommendation service to the

users

10a, 10b, ... 10n in association with the general search service server 50, for example, the

users

10a, 10b, ... 10n. The user may request a recommendation for a product by accessing a search service server 50 that he or she normally uses, and in this case, the search service server 50 requests a product recommendation in consideration of the user's taste to the recommendation service device 100. The recommendation service apparatus 100 may provide the search service server 50 with a product recommendation result in consideration of the taste of the corresponding user.

2 is a block diagram of an embodiment of a recommendation service apparatus for providing a product recommendation service according to the present invention.

The recommendation service apparatus 100 may include a ranking setting unit 110, a similar user determination unit 130, a product recommendation unit 150, and the like, and may further include a user information DB 170. have.

The ranking setting unit 110 generates a product list for each user about products experienced or possessed by each user, and selects a preferred product through a relative comparison according to his or her taste, and based on this, the product on the product list. Determine a ranking for, and matching the product list and the ranking for the product according to generate a product ranking list for each user. In this case, the ranking setting unit 110 may provide a user with a service page or a tool that allows a user to input a list of products experienced or possessed by the user and determine a preferred product through relative comparison for each product according to his or her taste. It may be.

Preferably, the product ranking list generated by the ranking setting unit 110 may be stored in the user information DB 170 corresponding to the identification information of the corresponding commercial user.

The similar user determination unit 130 determines a similar user based on the product ranking list for each user generated by the ranking setting unit 110. As an example, the product ranking list corresponding to a specific user and the product ranking corresponding to another user are determined. By cross comparing the list, other users having similar tastes with the specific user are determined. In this case, the similar user determination unit 130 may calculate the similarity according to the cross comparison of the product ranking list, and may select one other user having the highest similarity or select a plurality of other users within the similarity within the preset range. .

The product recommendation unit 150 identifies one or more other users having similar tastes with the corresponding user determined by the similar user determining unit 130, and experiences the corresponding user based on a product ranking list set by another user having similar tastes. Create and provide recommendation information for recommending products that you do not have.

With such a configuration, the recommendation service apparatus 100 may provide a recommendation service according to the present invention as a function of each configuration, where the recommendation service apparatus 100 may be implemented as one device including each configuration. Alternatively, each configuration may be configured by each individual device or individual module and selectively combined, or the recommended service device 100 may be implemented as a group of each configuration. Furthermore, a kind of server capable of performing a storage function and a calculation function, etc. may be applied to the recommendation service device 100, and a computer program for performing each step of the product recommendation service method according to the present invention to be described later may be recorded in the recommendation service device. have.

Through the recommendation service apparatus according to the present invention described above, it is possible to provide a product recommendation service through relative comparison considering the user's taste, and a method for providing a product recommendation service through relative comparison according to the present invention Let's look through.

3 is a schematic flowchart of an embodiment of a product recommendation service method through relative comparison according to the present invention.

Since the product recommendation service method according to the present invention is implemented in the recommendation service apparatus according to the present invention described above, it will be described with reference to the recommendation service apparatus.

The recommendation service method according to the present invention comprises: a product ranking setting process of setting a ranking for a product through relative comparison according to a user's preference (S100); A similar taste user determination process of determining one or more other users having similar tastes with the specific user by cross comparing the ranking of the product determined by the specific user requesting the recommendation service with the ranking of the product determined by the other user (S200); And a product recommendation process (S300) for recommending a product to the specific user based on a ranking of a product determined by another user of similar taste.

Each process described in the recommendation service method according to the present invention will be described in more detail through the following detailed embodiments.

First, the product ranking setting process S100 will be described with reference to an embodiment of the product ranking setting process in the recommendation service method according to the present invention shown in FIG. 4 together with FIG. 3.

First, the ranking setting unit 110 of the recommended service device 100 receives product information experienced or held by the user from the user (S110) and generates a product list (S120). 4 is an embodiment for setting the ranking for the restaurant, the ranking setting unit 110 of the recommended service device 100 receives the restaurant information experienced by the user from the user A, and collects the information ( As shown in a), a restaurant list 210 that combines the restaurant information 215 experienced by the user A is generated.

The ranking setting unit 110 of the recommendation service apparatus 100 sets a ranking of products based on the user's preference based on a comparison between products included in the product list according to the user's taste. In operation S130, the merchandise ranking list is generated by integrating them, in operation S140.

As shown in (b) of FIG. 4, the user A may perform a relative comparison according to the taste of the user A with respect to the restaurant 1 221 and the restaurant 2 223 on the user A's restaurant list 210. Relatively more preferred restaurant 2 225 may be selected, and the ranking setting unit 110 of the recommended service device 100 performs the relative comparison repeatedly so that the list of restaurants experienced by the user according to the taste of the user may be selected. A ranking is set based on the selection of the restaurants on 210 and the sum is generated to generate a restaurant ranking list. Here, the restaurant ranking list is a restaurant ranking list 230 in which all restaurants 215 on the restaurant list 210 are arranged in sequence (235) and the ranking 231 is set as shown in FIG. 4 (c). You can also create

In order to input a product and set a ranking of the user, the recommended service device 100 may provide a service page or an application for input and ranking of product information to the user's terminal.

Through the product ranking setting process (S100) as described above, according to the user's preference to set the ranking for the product through a relative comparison, a plurality of users by repeatedly performing the product ranking setting process (S100) a plurality of A user ranking list for each user may be generated.

The product ranking list for each user generated as described above may be stored in the user information DB 170 corresponding to each user.

In addition, the user may update the overall product ranking list by adjusting the preference selection of the product according to the re-experience of the existing products in the product ranking list, and the user may also update the existing product rankings for new experiences or newly owned products. The product ranking list can be updated by adding a new ranking list.

Secondly, the similar taste user determination process (S200) will be described with reference to an embodiment of the similar taste user determination process in the recommended service method according to the present invention shown in FIG. 5 along with FIG. 3.

In a state in which product ranking lists for a plurality of users are held through the product ranking setting process S100, the similar user determination unit 130 of the recommendation service apparatus 100 according to a product recommendation request from a specific user S210. Comparing a product ranking list corresponding to the specific user with a product ranking list corresponding to at least one or more other users (S220), a similarity degree between the specific user and other users' preferences is calculated (S230). In operation S240, the user may determine another user having a similar taste with the specific user based on the calculated similarity.

For example, as shown in (a) of FIG. 5, the restaurant ranking list 310a corresponding to the user N from the restaurant ranking list 310a in which the ranking is set by sequential ordering of the plurality of restaurants, respectively, corresponds to the user N ( If user A requests a restaurant recommendation in the state that 310n is retained, the restaurant ranking list corresponding to user N from the restaurant ranking list 310a corresponding to user A and the restaurant ranking list 310b corresponding to user B Cross comparison of up to 310n yields the similarity of each other user to user A.

In this case, the similarity may be calculated in consideration of only the restaurant ranking list including the same number of products and the same product included in the product ranking list corresponding to the user A among the users B to N.

Regarding the similarity calculation, referring to the example illustrated in (b) of FIG. 5, a plurality of restaurants are ranked according to a sequential order of products with respect to a restaurant ranking list in which a ranking is set by sequential ordering of a plurality of restaurants. In (b) of FIG. 5, restaurants included in the restaurant ranking list corresponding to user A are classified into

groups

320a and 330a, respectively, according to the sequence order. Ten restaurants in the ranking list were also classified into

groups

320b and 330b.

In addition, a similarity of user B with respect to user A is calculated through cross-comparison of restaurant groups included in the restaurant ranking list of user A and user B who have requested a restaurant recommendation, and the upper ranking group 320a corresponding to user A is selected as user B. Cross-compare with each of the upper ranking group 320b and the lower ranking group 330b corresponding to, and further compare the lower ranking group 330a corresponding to the user A with the upper ranking group 320b and the lower ranking group corresponding to the user B. 330b is cross compared to each other.

For example, as a grasp of similar taste between user A and user B, the restaurant 323a included in the upper ranking group 320a corresponding to user A and the restaurant 323b included in the upper ranking group 320b corresponding to user B may be used. The number of restaurants belonging to the upper ranking group is determined by comparing the same as restaurants 1 (325a and 325b). The restaurant 333a included in the lower ranking group 330a corresponding to user A is compared with the restaurant 333b included in the lower ranking group 330b corresponding to user B, such as restaurant 12 (335a, 335b). Similarly, determine the number of restaurants in the lower ranking group.

In addition, as a grasp of the opposite taste between the user A and the user B, the restaurant 323a included in the higher ranking group 320a corresponding to the user A and the restaurant 333b included in the lower ranking group 330b corresponding to the user B are selected. In comparison, the number of restaurants belonging to the opposite ranking group, such as restaurants 2 (327a and 337b), is determined. Similarly, the number of restaurants belonging to the opposite ranking group is compared by comparing the restaurants 333a included in the lower ranking group 330a corresponding to user A with the restaurants 323b included in the upper ranking group 320b corresponding to user B. Figure out.

Based on this comparison, the similarity of user B with respect to user A is calculated. Similarly, restaurants belonging to a higher ranking group mean that users A and user B have similar tastes. The higher weight is applied to the number, and since the restaurants belonging to the lower ranking group mean that the tastes of user A and user B may be somewhat similar, the middle weight is applied to the number of restaurants belonging to the lower ranking group. Since the restaurants belonging to the opposite ranking group may have different tastes of the user A and the user B, the similarity is calculated by applying the low weight.

As an example, the similarity of user B to user A may be calculated by the following [Formula 1].

S _Aβ = (α × H + β × L + γ × R) / T; (α>β> γ) [Formula 1]

Where S _ΑΒ is the similarity of user B to user A, α is the weight for products belonging to the same high ranking group, β is the weight for products belonging to the same lower ranking group, and γ is the opposite ranking The weight of the products belonging to the group, H is the number of products belonging to the same high ranking group, L is the number of products belonging to the same lower ranking group, R is the number of products belonging to the opposite ranking group, T Represents the number of products in which H, L, and R are added together.

By repeating the above process, similarity with respect to user A is calculated for user B from user B to user N shown in (a) of FIG. 5.

Then, another user with high similarity is selected for the user A, in which only one other user having the highest similarity may be selected, or a plurality of other users having similarities up to a predetermined value may be selected.

Third, the product recommendation process (S300) will be described with reference to an embodiment of the product recommendation process in the recommendation service method according to the present invention shown in FIG.

When one or more similar taste users having similar tastes with the specific user who requested the product recommendation are selected through the similar taste user determination process (S200), the product recommendation unit 150 of the recommendation service apparatus 100 may correspond to the specific user. The product ranking list and the product ranking list corresponding to the selected one or more similar taste users are extracted (S310).

Next, the product recommendation unit 150 of the recommendation service apparatus 100 compares the product ranking list of the specific user with the product ranking list of the similar taste user (S320), and corresponds to the product ranking list corresponding to the similar taste user. In step S330, a new product is extracted from a product not included in the product ranking list corresponding to the specific user.

In addition, the product recommendation unit 150 of the recommendation service apparatus 100 determines a recommended product based on the ranking of the new product set by the user of similar preference (S340), and recommends information including the recommended product. Provided to a specific user (S350).

If only one other user who has the highest similarity with respect to a specific user who has requested a product recommendation is selected as a similar taste user, a new product is extracted from a ranking list corresponding to one similar taste user, and the new product of the highest ranking among the new products is extracted. The new products are selected as the recommended products in the order of increasing the ranking of the products or the predetermined number.

In addition, a case in which a plurality of different users up to a preset value of similarity is selected as a similar taste user will be described with reference to the embodiment illustrated in FIG. 6.

6 illustrates an example of extracting a recommended restaurant based on a restaurant ranking setting of a plurality of similar taste users in relation to the restaurant recommendation described above.

First, user B, user D selected as user similar to user A, user D,. A restaurant ranking list corresponding to the user C is extracted, and restaurants not included in the restaurant ranking list corresponding to the user A are extracted from each restaurant ranking list as new restaurants, and similarity of the similar taste user is illustrated in FIG. 6. A new restaurant table 410 is created in which new restaurants corresponding to each similar taste user are arranged in order. 6 is a portion included in the restaurant ranking list corresponding to user A and excluded from the new restaurant.

For each new restaurant based on the new restaurant table 410, a recommendation score is calculated based on the similarity of the similar taste user and the ranking set by the similar taste user. In the case of restaurant 3 (411, 413), In this regard, since the ranking set by the user B with the highest similarity is second and the ranking set by the user D with the highest similarity is first, the recommendation score is calculated high. In the case of restaurant 6 (415, 417), since the ranking set by the user B with the highest similarity is third and the ranking set by the user C with the lowest similarity is first, it is relatively lower than the restaurant 3 (411). Is calculated.

As an example, the recommendation score for the new product may be calculated by the following [Formula 2].

[Equation 2]

Here, SCORE is a recommendation score for the new product, M is the total number of similar taste users who set the ranking for the new product, SI is the similarity degree of the i-th similar taste user, and Rank is the i-th similar taste user. It is the ranking set to the said new product, and k is a weight set as needed.

As such, each recommendation score is calculated for each new product, and a product to be recommended is selected based on the calculated recommendation score. Only one new product having the highest recommendation score may be selected as a recommendation product, A plurality of new products may be selected as recommended products up to a higher order.

When the recommendation product is selected, the product recommendation unit 150 of the recommendation service apparatus 100 generates and provides recommendation information including the recommendation product to the user, for example, the recommendation service method according to the present invention shown in FIG. 7. As in the embodiment of the product information, the recommendation information 420 including recommendation restaurants in a sequential order together with the recommendation order is provided to the user terminal. In this case, various additional information may be included in the recommendation information. For example, in the case of the recommendation of the restaurant as illustrated in FIG. In addition, a URL for a web page on which such various information is presented may be included.

(Example 1) Here, the product recommendation process S300 can also be improved as follows. Hereinafter, the product recommendation process for performing the product recommendation process (S300) with the recommended service method according to an improved example of the embodiment with reference to Figures 3, 8 and 9 will be described.

The product recommendation unit 150 of the recommendation service apparatus 100 compares the product ranking list of the specific user with the product ranking list of the similar taste user (S320), and corresponds to the specific user in the product ranking list corresponding to the similar taste user. Products not included in the product ranking list are extracted as new products (S330).

Hereinafter, with reference to FIG. 8, the processing of such contrast (S320), and new product extraction (S330) will be described in more detail.

First, a product ranking list corresponding to a user B, a user C, a user D, a user E, ... user F selected as a user with a similar taste user is extracted, and a product ranking list corresponding to the user A on each product ranking list. By extracting a product that is not included as a new product, to create a new product table 510 sorting the new products corresponding to each similar taste user in the order of similarity preference users. In FIG. 8, a part marked with * denotes a product included in a product ranking list corresponding to user A and is excluded from the new product.

Based on the new product table 510, a recommendation score is calculated for each new product based on the similarity of the similar taste user and the ranking set by the similar taste user. The calculation procedure is as follows.

Each ranking is assigned a ranking score in advance. In this example, a ranking score of 100 is placed in the first place, a ranking score of 90 in the second place, a ranking score of 80 in the third place, and a ranking score of 70 in the fourth place. Then, the similarity SIRawi of each comparison target person, ie, a similar taste user, is converted into a ratio SIi such that the sum is one. This conversion follows Equation 3 below.

, [Equation 3]

Here, SIi is the ratio of the similarity, SIRawi is the similarity of each similar taste user, M is the number of comparators.

After calculating the ratio SIi for the similarity as described above, the recommended score SCORE of the product to be evaluated is calculated by the following [Equation 4] based on the ratio of each user and the rank score RankScorei.

, [Equation 4]

Here, SIi is a ratio with respect to the similarity, RankScorei is a ranking score of products in the product list of each user, and M is the number of comparison targets.

For example, in Fig. 8, the recommendation score and the expected ranking of the product P12 are determined as follows.

In this way, the user A's product ranking list has the same recommendation score as the product P3, so the expected ranking is third.

8, the recommendation score and the expected ranking of the product P10 are determined as follows.

Here, the recommendation score SCOREP10 determines the expected ranking in comparison with the ranking score of the user A's product ranking list.

As such, each recommendation score is calculated for each new product, and a product to be recommended is selected based on the calculated recommendation score, and only one new product having the highest recommendation score may be selected as a recommendation product, and among the recommended scores. A plurality of new products may be selected as recommended products up to the set upper order.

When the recommendation product is selected, the product recommendation unit 150 of the recommendation service apparatus 100 generates and provides recommendation information including the recommendation product to the user, for example, as shown in FIG. 9. The recommendation information 520 is provided to the user terminal. At this time, the recommendation information may also include various additional information. For example, if the restaurant recommends, information such as menu, price, location, and phone number may be included, or various user reviews may be included. The URL for the web page where is presented may be included.

(Improvement Example 2) Further, the user may subjectively give a relative preference to his / her product, generate a product ranking list, and judge the similar taste user and the recommended product based on the product ranking list. In such a case, it is possible to determine similar users and recommended products on the basis of subjective relative preferences and minimal information on the user's products, thereby reducing the database capacity, calculation, and processing load required for providing information. .

In such a case, the similar taste user judgment process (S200) can be improved as follows. In the above method, a product ranking list of another user and user A to be compared is grouped as a comparison target for a user (for example, in order of increasing number of common products) including a product common to user A in the product ranking list. By dividing by, cross comparison was performed for each group to calculate similarity and to determine similar taste users.

On the other hand, a user who includes a large number of products belonging to the upper group of user A in the upper group may be determined as a similar taste user.

Alternatively, the ranking of each product included in the user A's product ranking list is checked in the product ranking list of the user to be compared, and the similarity is calculated by comprehensively considering the difference in the rankings, and the similar taste users are judged. Also good. This can be considered to be calculated by, for example, comparing the ranking of the product X in the product ranking list of the user A with the ranking of the product X in the product ranking list of another user.

(Improvement Example 3) Similar taste user determination process (S200) can also be improved as follows. Hereinafter, a similar taste user determination process (S200) of implementing the similar taste user determination process (S200) as a recommendation service method according to an improvement of the embodiment will be described.

In response to a request for recommending a product from a specific user, the similar user determination unit 130 of the recommendation service apparatus 100 may perform a product ranking list corresponding to a specific user and a product ranking list corresponding to another user according to a ranking of a specific range. Classified into a plurality of groups, and comparing the product ranking list between the plurality of classified groups to calculate the similarity of the taste only in each of the groups. In this case, the specific range may be set by a specific user or by a system (ie, a recommended service device).

That is, the similar user determination unit 130 of the recommended service apparatus 100 classifies each of the product ranking list corresponding to a specific user and the product ranking list corresponding to another user into a plurality of groups according to a sequential sequence order. Rather than a group consisting of a product of a ranking corresponding to a specific range (for example, top 1 to 10) of a product ranking list corresponding to a specific user and a specific range (for example, a product ranking list corresponding to another user) For example, the information is classified into a group consisting of products ranked in the top 1 to 10 ranks, or a group consisting of products in a rank (eg, 30 or less) that falls in a range different from a specific range. Similarity of tastes for a specific user and another user can be calculated through the cross comparison for each group.

At this time, the similar user determination unit 130 of the recommended service device 100 classifies a product ranking list of a specific user into two groups (for example, a group of top 1 to 10 and a group of 30 or less), and A user's product ranking list may also be classified into two groups (for example, a group of top 1 to 10 and a group of 30 or less), and then cross comparison for each of the classified groups may be performed.

For example, a group that is subject to cross comparison by group is a group consisting of products having a rank corresponding to a specific range among the ranking lists corresponding to a specific user, and a group that is different from a specific range in the product ranking list corresponding to a specific user. A group consisting of products of the corresponding ranking, a group consisting of products with a ranking corresponding to a specific range of the product ranking list corresponding to another user, and a product of a ranking falling within a specific range from the product ranking list corresponding to another user. It may include a group consisting of.

In addition, the similar user determination unit 130 of the recommended service apparatus 100 classifies each of the product ranking list corresponding to a specific user and the product ranking list corresponding to another user into a plurality of groups according to a sequential sequence order. Next, by selectively applying each of the categorized groups, a product ranking list between the selected groups is compared, and similarity of taste is calculated using only the respective groups.

That is, the similar user determination unit 130 of the recommendation service apparatus 100 classifies each of the product ranking list corresponding to a specific user and the product ranking list corresponding to another user into a plurality of groups according to a sequential sequence order. The similarity is calculated by comparing the first group and the second group corresponding to the specific user among the plurality of groups with the first group and the second group corresponding to the other user. Accordingly, the third group, which is composed of products having a rank corresponding to a range different from a range corresponding to the first group and the second group, is not included in the comparison process.

(Example 4) The product recommendation process (S300) can be improved as follows. Hereinafter, a product recommendation process for implementing the product recommendation process (S300) by the recommended service method according to the improved example of the embodiment will be described.

For this improvement, the recommended service device 100 may further include a pre-validation evaluation unit 160 as shown in FIG.

The pre-validation evaluation unit 160 of the recommended service device 100 designates a product for which a specific user wants to know whether or not it matches his or her taste, and provides evaluation information of the similar taste user with respect to the designated product. Accordingly, the user may perform verification and evaluation of the corresponding product in advance. At this time, the designated product designated by the specific user is a product which is not included in the product ranking list of the specific user that the specific user has not experienced.

To this end, as shown in FIG. 11, with respect to a product ranking list generated for each user (S410), a specific user may specify a product that he / she is curious about among products that he or she has not experienced (S420). By specifying a product, a pre-verification and evaluation of a specified product is requested to the pre-validation evaluation unit 160 of the recommended service device 100 (S430). Then, the similar user determination unit 130 of the recommended service device 100 determines a similar taste user for the designated product (S440), and the similar taste user with the similar taste with the similar taste user is set by the similar taste user. Based on the rank information, the recommendation score for the designated product is calculated in the same manner as in the improved example 1 (that is, using Equations 3 and 4) (S450). Then, the calculated recommendation score is provided to a specific user through the product recommendation unit 150 of the recommendation service apparatus 100 (S460).

The product recommendation process (S300) in the present modified example is to determine whether or not the inexperienced product that a specific user encounters on the road, TV, newspaper / magazine, friend, SNS, Internet, catalog, menu (restaurant), etc. fits the taste of the specific user. It can be used to verify and evaluate in advance.

Such verification and evaluation results may be provided in the same manner as in FIGS. 12A-12C.

The verification and evaluation results are displayed on the user A's ranking list as shown in FIG. 12A, as shown on the menu of the restaurant as shown in FIG. 12B, or as recommended in the product as shown in FIG. 12C. It can be displayed as an information center. At this time, the verification and evaluation results may also display the expected ranking on the user A's ranking list.

For example, the product recommendation process (S300) in the improved example, if a particular user wants to check whether the menu "Ganjikin" of the chicken restaurant "Canggu Chicken" seen on the street fits his or her taste, By using the communication terminal of the "Changu Chicken" designation of the "Gan Gan chicken" menu request the verification and evaluation results to the recommendation service device 100, and receives the verification and evaluation from the recommendation service device 100, The expected ranking and recommendation on the ranking list may be displayed.

In addition, if you want to select the menu that suits your taste in the restaurant "Halbum Chicken", which is the first visit by a specific user, you can use the whole communication terminal or the menu of your choice, "Salmon Dried Chicken" or "Fried Whole Chicken Chicken". "," Clean fried chicken "and" goso cheese "to request the verification and evaluation results to the recommended service device 100, and receives the verification and evaluation from the recommended service device 100 to The expected ranking and recommendation on the product ranking list may be displayed.

In addition, when a specific user wants to display an estimated rank on his ranking list on a map for a nearby chicken house while moving, a menu within a specific range of areas is provided in conjunction with map information provided or executed on his communication terminal. Search for "chicken", request verification and evaluation results from the recommended service device 100 with the information of the found "chicken" (ie, a unique number of each chicken menu, etc.), and verify and evaluate from the recommended service device 100. As shown in FIG. 13, the estimated ranking and recommendation on the product ranking list for each "chicken" menu may be displayed on the map.

According to the present invention as described above, by recommending a product based on the evaluation of the product set by the other users with similar tastes to the user, a product recommendation that can satisfy the user can be made.

The above description is merely illustrative of the technical idea of the present invention, and those skilled in the art to which the present invention pertains may make various modifications and changes without departing from the essential characteristics of the present invention. Therefore, the embodiments described in the present invention are not intended to limit the technical idea of the present invention but to explain, and the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

For example, the present invention can also be realized as a program that allows a computer to function as a ranking setting unit 110, a similar user determination unit 130, and a product recommendation unit.

Claims

A product ranking setting step of setting a ranking for products for each user based on a user's preferred product selection through relative comparison with respect to a plurality of products according to tastes of the plurality of users;

A similar taste user determination step of judging a ranking of a product set by a specific user and a ranking of a product set by at least one other user to determine another user having a similar taste with the specific user; And

And a product recommendation step of recommending the product to the specific user based on a ranking of products set by other users having similar tastes with the specific user.
The method of claim 1,

The product ranking setting step,

A product list generation step of receiving a plurality of product information from the user and generating a product list; And

A ranking of the plurality of products is set based on the user's preferred product selection through relative comparison according to the user's taste with respect to the plurality of products included in the product list, and based on the product corresponding to the user And a product ranking list generating step of generating a ranking list.
The method of claim 2,

The product ranking list generation step,

The product recommendation service method of claim 1, wherein the ranking is set by sequential ordering of the plurality of products.
The method of claim 2,

The similar taste user determination step,

A product recommendation request step of receiving a product recommendation request from a specific user among the plurality of users;

A similarity calculating step of comparing a product ranking list corresponding to the specific user with a product ranking list corresponding to one or more other users of the plurality of users and calculating a similarity; And

And a similar taste user selecting step of determining one or more other users having similar tastes with the specific user among the one or more other users based on the similarity.
The method of claim 4, wherein

The similarity calculating step,

Extracts a product ranking list corresponding to one or more other users that include the same product as the product included in the product ranking list corresponding to the specific user or more than a predetermined reference number, and extracts the product ranking list corresponding to the specific user from the other And comparing the product ranking list corresponding to the user to calculate the similarity.
The method of claim 4, wherein

The product ranking list is set by ranking the sequential sequence of each of a plurality of products,

The similarity calculating step,

Classifying a plurality of products into a plurality of groups according to a sequential order for each of the product ranking list corresponding to the specific user and the product ranking list corresponding to the other user; And

By comparing the group of the product ranking list corresponding to the specific user and the group of the product ranking list corresponding to the other user for each group, the similarity of the other user with respect to the specific user based on the number of the same product for each group A product recommendation service method comprising the step of calculating.
The method of claim 4, wherein

The product recommendation step,

A list extraction step of extracting a product ranking list corresponding to the specific user and a product ranking list corresponding to the selected one or more similar taste users;

A new product extraction step of extracting a new product from a product ranking list corresponding to the similar taste user not included in the product ranking list corresponding to the specific user;

A recommendation product determining step of determining a recommendation product based on a ranking for a new product set by the user of similar taste; And

And providing recommendation information including the recommendation product to the specific user.
The method of claim 7, wherein

The list extraction step,

Extracting a product ranking list corresponding to a similar taste user having the highest similarity with respect to the specific user;

The recommended product determination step,

A product recommendation service method comprising selecting a new product as a recommended product in order of increasing the ranking of a new product or a predetermined number of the highest ranking.
The method of claim 7, wherein

The list extraction step,

Extracting a product ranking list corresponding to a similar taste user up to a preset value of the similarity with respect to the specific user;

The recommended product determination step,

For each new product, a recommendation score is calculated based on the similarity of the similar taste user and the ranking set by the similar taste user, and the new product is selected as the recommended product based on the recommendation score. Recommended service method.
The method of claim 9,

A ranking score is pre-assigned to each ranking of the product ranking list, and in the recommendation product determining step, the similarity degree of the similar taste user is converted into a ratio such that the sum is 1, and the ratio of the similarity and the ranking A product recommendation service method for calculating a recommendation score of a new product based on a score.
The method of claim 7, wherein

The product ranking list is a product recommendation service method of ranking products based on a relative preference based on the user's subjectivity.
The method of claim 4, wherein

The similarity calculating step,

Classifying a product ranking list corresponding to the specific user and a product ranking list corresponding to the other user into a plurality of groups according to a ranking of a specific range; And

Group a group consisting of products of a ranking corresponding to a specific range of the product ranking list corresponding to the specific user and a group consisting of products of a ranking corresponding to a specific range or a range different from a specific range of the product ranking list corresponding to another user And comparing each other, and calculating a similarity degree of the other user to the specific user based on the number of identical products for each group.
The method of claim 4, wherein

The product recommendation step

A pre-verification and evaluation request step of requesting pre-verification and evaluation of the designated product by specifying a product which the specific user wants to pre-verify and evaluate;

A recommendation score calculating step of calculating a recommendation score for the designated product based on similarity with the similar taste user for the designated product and ranking information set for the designated product by the similar taste user; And

And providing recommendation information including the calculated recommendation score to the specific user.
A ranking setting unit configured to set a ranking of products for each user based on the user's preferred product selection through relative comparison of a plurality of products according to tastes of a plurality of users;

A similar user determination unit which cross compares a ranking of a product set by a specific user with a ranking of a product set by at least one other user, and determines another user having a similar taste with the specific user; And

And a product recommendation unit for recommending a product to the specific user based on a ranking of products set by other users having similar tastes to the specific user.
The method of claim 14,

The ranking setting unit,

Generates a plurality of user-specific product list, and generates a product ranking list for each user by setting a ranking for the product on the product list based on the user's preferred product selection through relative comparison according to each user's taste,

The similar user determination unit,

Comparing the product ranking list corresponding to the specific user who requested the product recommendation with the product ranking list corresponding to the other user, calculating the similarity of the other user to the specific user, and based on the similarity, the similarity to the specific user. Judge your taste buds,

The product recommendation part,

From the product ranking list corresponding to the user with similar tastes, a product which is not included in the product ranking list corresponding to the specific user is extracted as a new product, and the recommended product is selected based on the ranking of the new product set by the user with the similar taste. Recommended service device, characterized in that judging.
A recommendation service apparatus in which a computer program for performing each step of the product recommendation service method according to any one of claims 1 to 13 is recorded.
Computer,

A ranking setting unit configured to set a ranking for products for each user based on the user's preferred product selection through relative comparison of a plurality of products according to tastes of a plurality of users;

A similar user determination unit which cross compares a ranking of a product set by a specific user with a ranking of a product set by at least one other user, and determines another user having a similar taste with the specific user; And

And a product recommendation unit for recommending a product to the specific user based on a ranking of the product set by another user having a similar taste with the specific user.