KR20210094987A - Method and system for recommending products based on preference and presenting recommended products for customers - Google Patents

Abstract

The present invention relates to a method for recommending the most suitable products for a customer on the basis of preference elements, a method for presenting recommended products, and a product recommendation system using the same. According to the present invention, a method for displaying recommended products in a three-dimensional (3D) space comprises the following steps: analyzing preference elements into two emotional attributes to providing a preference map arranged two-dimensionally and a product map in which products are two-dimensionally arranged to correspond to the topography of the preference map; extracting products corresponding to at least one preference element selected by a customer in the preference map from the product map to determine the extracted products as recommended products; when first to third categories are determined among a plurality of categories, determining 3D coordinates for each recommended product, which have an azimuth component classified according to metainformation belonging to the first category, a vertical axis component aligned according to metainformation belonging to the second category, and a horizontal axis component aligned according to metainformation belonging to a third category; rendering a 3D matrix in which visualization display information for visualizing the recommended product is arranged in 3D coordinates; and displaying the 3D matrix on a screen.

Description

Recommendation method based on taste factors, method of proposing recommended products, and product recommendation system using them

The present invention relates to a product recommendation method, and more particularly, to a method and system for recommending a product based on taste factors selected by a user.

The product recommendation service is currently widely used in marketing. In particular, it is a method of analyzing a customer's purchasing behavior to recommend a product that the customer is likely to prefer, or a product that other customers who have similar purchasing behavior have purchased or shown interest in. is being developed mainly in the way of recommending to customers.

These purchasing behavior-based product recommendation services essentially operate on the assumption that the customer's purchasing behavior pattern will be consistent, and since they are recommended to customers based on past data, there is a high possibility that customers will not receive practical help at this time. . For example, a customer is dissatisfied with an existing product and wants a new style product. The conventional product recommendation service will recommend products that were previously preferred according to the purchasing behavior pattern of the past customer.

Another product recommendation service is simply a method of recommending popular products to a specific group of customers classified as belonging to a certain customer. However, this product recommendation service is based on the past purchase data of multiple customers under the assumption that customers belonging to a specific group will prefer similar products.

Therefore, there is a need for a service that can recommend the most suitable product based on the abstract taste factors that customers currently want to choose.

The problem to be solved by the present invention is to provide a method and apparatus for recommending the most suitable product based on the abstract taste elements currently possessed by the customer.

A three-dimensional display method for recommended products according to an aspect of the present invention,

providing a taste map two-dimensionally arranged by analyzing taste elements into two emotional attributes, and a product map in which goods are two-dimensionally arranged to correspond to the topography of the taste map;

extracting products corresponding to at least one taste element selected by the customer from the taste map from the product map and determining them as recommended products;

When the first to third categories are determined from among a plurality of categories of meta information describing the recommended products, the azimuth component classified according to the meta information belonging to the first category for each recommended product and belonging to the second category determining a three-dimensional coordinate having a vertical axis component aligned according to meta information and a horizontal axis component aligned according to meta information belonging to the third category;

Rendering a three-dimensional matrix in which the visualization display information for visualizing the recommended product is arranged in the three-dimensional coordinates; and

The method may include displaying the 3D matrix on a screen.

According to an embodiment, the taste element may include at least one of a color, a color combination, a phrase, and an associative image.

According to an embodiment, the meta information may include at least one of a product name, a brand, a price, a color, a material, a taste factor, popularity, a rating, and a product characteristic.

According to one embodiment, the three-dimensional display method of the recommended product,

The method may further include displaying the visualization display information and the meta information in an auxiliary display area when the customer selects a specific recommended product in the 3D matrix.

A three-dimensional display method of data according to another aspect of the present invention,

When first to third categories are determined from among a plurality of categories of meta information describing a data item, an azimuth component classified according to meta information belonging to the first category for each data item and meta information belonging to the second category determining a three-dimensional coordinate having a vertical axis component aligned according to information and a horizontal axis component aligned according to meta information belonging to the third category;

rendering a three-dimensional matrix in which the visualization display information for visualizing the data item is arranged in the three-dimensional coordinates; and

The method may include displaying the 3D matrix on a screen.

Recommendation product three-dimensional display system according to an aspect of the present invention,

a map database including a taste map in which taste elements are analyzed and arranged two-dimensionally by two emotional attributes, and a product map in which goods are two-dimensionally arranged to correspond to the topography of the taste map;

a product recommendation unit for extracting products corresponding to at least one taste element selected by the customer from the taste map from the product map and determining them as recommended products; and

When the first to third categories are determined from among a plurality of categories of meta information describing the recommended products, the azimuth component classified according to the meta information belonging to the first category for each recommended product and belonging to the second category and a matrix generator for determining three-dimensional coordinates having a vertical axis component aligned according to meta information and a horizontal axis component aligned according to meta information belonging to the third category.

According to one embodiment, the recommended product three-dimensional display system,

It may further include a three-dimensional display unit for rendering a three-dimensional matrix in which the visualization display information for visualizing the recommended product is arranged in the three-dimensional coordinates.

According to one embodiment, the recommended product three-dimensional display system,

When the customer selects a specific recommended product from the three-dimensional matrix, the display unit may further include a two-dimensional display unit for displaying the visualization display information and the meta information in an auxiliary display area.

Recommendation product three-dimensional display system according to another aspect of the present invention,

A taste map is two-dimensionally arranged by analyzing taste elements into two emotional attributes, and a product map is two-dimensionally arranged to correspond to the topography of the taste map. Products corresponding to at least one selected taste element are extracted from the product map and determined as recommended products, and when first to third categories are determined from among a plurality of categories of meta information describing the recommended products, each For each recommended product of a product recommendation server for determining three-dimensional coordinates; and

It may include a terminal for rendering a three-dimensional matrix in which the visualization display information for visualizing the recommended product is arranged in the three-dimensional coordinates and displayed on the screen.

According to an embodiment, the terminal may be operable to display the taste map from the product recommendation server to the customer, and transmit a taste element selected by the customer in the taste map to the product recommendation server.

According to an embodiment, the terminal may further include a two-dimensional display unit for displaying the visualization display information and the meta information in an auxiliary display area when the customer selects a specific recommended product from the three-dimensional matrix.

A three-dimensional display system for data according to an aspect of the present invention

a database for storing data items; and

When first to third categories are determined from among a plurality of categories of meta information describing the data item, an azimuth component classified according to meta information belonging to the first category for each data item and belonging to the second category and a 3D matrix generator for determining 3D coordinates having a vertical component aligned according to meta information and a horizontal component aligned according to meta information belonging to the third category.

According to the method and system of the present invention, a customer can be recommended a plurality of products or services by selecting an abstract taste element that he currently has.

Furthermore, according to the method and system of the present invention, it is possible to display recommended goods or services in a three-dimensional matrix and allow the customer to browse and select.

With respect to the embodiments of the present invention disclosed in the text, specific structural or functional descriptions are only exemplified for the purpose of describing the embodiments of the present invention, and the embodiments of the present invention may be implemented in various forms and the text It should not be construed as being limited to the embodiments described in .

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. The same reference numerals are used for the same components in the drawings, and duplicate descriptions of the same components are omitted.

1, 2, 3 and 4 are diagrams illustrating a taste map that can be used in an embodiment of the present invention.

Referring to FIG. 1 , the taste map assigns emotional properties to the horizontal and vertical axes of the two-dimensional plane, respectively, and according to the intensity of the corresponding emotional property in the two-dimensional plane, a phrase, color, or color combination expressing emotion, which Taste elements such as images and materials that evoke emotions are arranged.

At this time, the emotional properties assigned to the horizontal axis and the vertical axis are 'orthogonal' properties to be expressed mathematically, and are concepts from which a third emotion can be derived at the point where they intersect each other. Conversely, a certain emotion is analyzed using two independent emotional attributes.

In the example of FIG. 1 , the horizontal axis is assigned dynamic properties. Moving to the left, dynamic words, colors, color combinations, and associative images are arranged, and moving to the right, static words, colors, color combinations, and associative images are arranged. On the other hand, the vertical axis is assigned strength and weakness properties. Words, colors, color combinations, and associative images that are strong and hard are arranged toward the bottom, and words, colors, combinations of colors, and associative images that are soft and soft are arranged toward the top.

For example, 'light feeling' is classified as very dynamic in terms of dynamic attributes and slightly soft in terms of dynamic attributes. Accordingly, in the taste map of FIG. 1 , the 'light feeling' will be arranged in a coordinate region that is biased to the left on the horizontal axis and slightly upward on the vertical axis.

As such, in addition to the modifiers that describe emotion, color combinations, or color itself, can be similarly placed on the taste map if they are classified based on dynamic properties and dynamic properties. For example, in FIG. 1 , the combination of dark green, very light yellow, and slightly light green was evaluated to give a very dynamic and slightly soft feeling, and was placed in the middle region of the second quadrant of the taste map. Furthermore, in this way, when several color combinations are placed around the location of 'light feeling', the seller can represent those color combinations with one modifier of 'light feeling'.

Such a taste map may be used when recommending a product in which color matching plays an important role, for example, clothes, furniture, indoor/outdoor design, and the like to a customer.

The taste elements placed on the taste map may be varied. For example, by using a favorite entertainer as a taste element, the names or photos of the entertainers may be arranged on the taste map according to dynamic and strength properties from the image of the entertainer.

Methods of arranging colors, color combinations, associative images, or emotional modifiers on the taste map may vary, and the results of placement may vary according to subjective criteria. In addition, the shape, coordinates, or width of the area occupied by each taste element may be appropriately selected according to, for example, the results of a survey study or psychological study on customers, or according to the intention, strategy, and experience of the seller or producer. .

2 is a taste map that can be used in an embodiment of the invention, and illustrates a taste map in which modifiers expressing senses and characteristics are arranged.

Referring to FIG. 2 , various modifiers such as 'elegant', 'glamorous', 'classy', 'subtle', and 'modern' are arranged according to dynamic properties and strength properties that each makes people feel.

Meanwhile, one representative modifier may be grouped to represent several similar modifiers. For example, the modifier 'decent' can represent 'reasonable', 'formal', 'refined', traditional', 'dignified', etc. Furthermore, there may be modifiers represented by two or more representative modifiers. For example, the modifier 'luxurious' belongs to the group of the modifier 'elegant Han', but also belongs to the group of the modifier 'decent'. Since the task of generating a taste map is a task of classifying tastes and sensations, there may be no absolute standards.

3 is a taste map that can be used in an embodiment of the present invention, and illustrates a taste map in which the taste map is divided into a small number of blocks and modifiers are arranged.

Referring to FIG. 3 , the horizontal axis was based on the emotional property related to temperature, and the vertical axis was based on the emotional property related to strength and weakness.

The middle four blocks are moderate, and are connected with modifiers such as 'elegant' and 'refined'. The three blocks in the upper left are warm and soft, and are connected with modifiers such as 'pretty', 'light', and 'romantic'. The upper right three blocks are soft and cool, and are connected with modifiers such as 'clean' and 'natural'. The three blocks in the lower left are warm and hard, and are connected with modifiers such as 'sporty' and 'personal'. The three blocks in the lower right are cool and hard, and can be connected with modifiers such as 'calm' and 'dignified'.

1 to 3 exemplify the arrangement of modifiers for color or sensory expression on the taste map, but other concepts may also be arranged on the taste map.

4 is a taste map that can be used in an embodiment, and illustrates a taste map in which modifiers representing materials are arranged.

Referring to FIG. 4 , the horizontal axis is based on dynamic properties and the vertical axis is based on strength and weakness properties, but unlike the previous examples, the taste map contains modifiers indicating material or texture, for example, wood, iron, Words such as stone, cotton, and paper are arranged.

1 to 4 , the marks arranged at each coordinate of the taste map may be different from each other, but since the properties of the coordinate axes are common to each other, it can be seen that they share the same sense. For example, when a customer wants a product that is slightly soft, not too dynamic and not too static, the taste map of FIGS. 1 to 4 shows that the customer has a 'natural' taste and , can identify color combinations and materials suitable for such tastes.

In the present invention, based on this point, the seller creates a product map in which the seller classifies his/her products according to the overall feeling, color, material, etc. on at least one taste map, and places the customer's taste on the taste map. This is a method and system for recommending products on a product map corresponding to a taste element corresponding to the selected area when an area is selected.

5 is a diagram illustrating a product map in which the product seller arranges products according to the geometry of at least one taste map in an embodiment of the present invention.

Referring to FIG. 5 , a product map created by sorting and arranging audio products that the product seller wants to sell according to color and design is exemplified. According to the coordinates and area where the product is arranged on the product map, corresponding taste element data is determined. The taste element data is data indicating which taste element is given to a certain product and is a taste element on a taste map on which the product map is based. Each product may be given one or more taste element data.

In this way, corresponding taste element data for all products is determined, and a product database is constructed using product data such as a product name, product description, sales conditions, and product image, and preference element data of the product.

What can be placed on the product map is not only tangible products, but also intangible goods and services such as movies and music.

6 is a conceptual diagram illustrating that a recommended product is determined by matching a product in a product map with respect to a taste selected by a customer from a taste map according to an embodiment of the present invention.

Referring to FIG. 6 , the map on the left is a taste selection map that asks customers for a desired taste, and the map on the right is a product map.

The customer may select at least one desired taste element on the taste selection map. At this time, since the product map and the taste selection map share the same emotional properties (that is, the horizontal axis and the vertical axis are the same as, for example, a dynamic property and a strength property, respectively), a product corresponding to the same coordinates or area as the taste element selected by the customer The product on the map can be determined as a recommended product suitable for the customer's taste.

Furthermore, if multiple product maps using the same emotional attributes as coordinate axes are built, various kinds of products corresponding to one taste element selected by the customer can be recommended from the multiple product maps.

In this case, depending on the embodiment, not only the taste element selected by the customer, but also products corresponding to the taste elements within a similar range may be recommended automatically or with the customer's consent. For example, when a customer selects a taste element of 'lovely' in a taste selection map such as the taste map of FIG. 2, the representative modifier 'cute' or ' All other modifiers in the group including the 'lovely' modifier are also selected as a taste factor so that corresponding products can be recommended.

On the other hand, in an embodiment of the present invention, after products corresponding to the taste element selected by the customer are recommended, a method for displaying the recommended products to the customer will be described next.

Data of each recommended product may be structured into meta information, arrangement information, and content information for describing the recommended product. However, the arrangement information is not an essential element of the data item, and may be information temporarily determined to display the data item on the screen.

In general, meta information is additional information for describing data. The utilization of data can be increased by recognizing what data is about and who created it when and how by meta-information.

In the present invention, the meta information includes a key word and a visual indication, and may further include taste element data. The keyword may be a product name, model name, brand, trademark, catch phrase, etc. that can represent the corresponding product. The subject word may be two or more phrases having different meanings, or may be several kinds of linguistic expressions having the same meaning. The visual display may be at least one two-dimensional image, an icon, or a three-dimensional model that can represent the corresponding product.

Meta information is displayed on the screen when the recommended products are arranged in a three-dimensional matrix. For example, the visual indication may be displayed in any one cell on the half-plane of the three-dimensional matrix, and the key word may be displayed in the two-dimensional auxiliary display area when the recommended product is selected, for example, by a mouse cursor.

The taste element data is data indicating at least one taste element of the product, and may belong to meta information describing the product. The taste element data may have a modifier, a color, a color combination, a texture, and an associative image.

The arrangement information is information for determining where to arrange the recommended products in the three-dimensional matrix. The arrangement information may be predetermined, but may be immediately determined immediately before displaying the recommended product on the screen after searching for the recommended product.

In the present invention, a three-dimensional matrix is a virtual structure for displaying data items on limited areas of radial virtual half-planes with respect to a certain vertical axis, that is, on a cell. The coordinates of a cell can be expressed in terms of horizontal axis, vertical axis, and azimuth component. Data items related to each other would preferably be displayed in cells on the same half-plane. Therefore, the most important coordinate component is the azimuth. For example, products belonging to one brand may be arranged on one virtual half plane, that is, represented in the same azimuth. Subsequently, in one half-plane, products may be arranged in order of price or popularity, and horizontal axis coordinates and vertical axis coordinates may be determined according to categories such as price order or popularity order. For example, it would be desirable to display data items from top to bottom, or from inside to outside, in ascending or descending order based on price or popularity. Therefore, the next important coordinate component is the vertical and horizontal coordinates.

As such, in order to display data items in a certain order, the arrangement information may have information belonging to a first category, a second category, and a third category determined based on the meta information, respectively. According to the first to third categories designated in advance or determined immediately, it is possible to determine at which coordinates, ie, in which cells, the data items in the three-dimensional matrix are to be displayed, respectively. The azimuth component, that is, the half-plane, is determined according to the information corresponding to the widest first category, the ordinate component is determined according to the information corresponding to the second intermediate category, and the abscissa component is determined according to the information corresponding to the narrowest third category. A coordinate component may be determined.

For example, a first category may be a brand, a second category may be a color, a third category may be a price, a first category may be a material, a second category may be a price, a third category may be a product name, and the first category may be a taste factor, the second category may be a material, and the third category may be a price.

On the other hand, only the first category, which determines which half-plane to belong to, is definitive, and the second and third categories may be unspecified. In this case, data items in one half plane do not overlap each other and may be randomly arranged.

Finally, the content information may include information such as text, hypertext, documents, photos and other multimedia attached to the corresponding data item. For example, the content information included in the data item of the recommended product may be a product manual, a picture of use, sales conditions, reference materials, and the like.

7 is a diagram illustrating an example of arranging and displaying recommended products in a three-dimensional matrix on the basis of recommended product data according to an embodiment of the present invention.

Referring to FIG. 7 , the 3D matrix 72 is positioned in the 3D display area 71 at the center of the screen. In the vicinity of the 3D matrix 72 , there may be further 2D auxiliary display areas 73 to 76 to increase search and browsing efficiency. Depending on the embodiment, the three-dimensional matrix 72 is rotated at a predetermined rotation speed so that the customer may browse the three-dimensional matrix 72 as a whole, or the customer may drag the three-dimensional matrix 72 using a mouse or the like. The dimensional matrix 72 may be rotated at any angle and direction.

The three-dimensional matrix 72 has anti-planes 721 radially formed at a predetermined azimuth. A plurality of cells 722 and 723 are arranged in each of the semi-planes 721 . Recommended products are arranged on the plurality of cells (722, 723).

According to the embodiment, the azimuth of the recommended products is determined by using a color as a first category among the taste elements, and then, in the corresponding half-plane 721 , each according to the average brightness value (vertical axis) and the saturation value (horizontal axis) of the product image It may be displayed in the three-dimensional matrix 72 in such a way as to be arranged in the cells 722 and 723 .

In another embodiment, the recommended products are azimuthally determined with the brand as the first category, and then, in the corresponding half-plane 721 , cells according to price (second category, vertical axis) and color (third category, horizontal axis) It can be represented in the three-dimensional matrix 72 in such a way as to be arranged at (722, 723).

In another embodiment, the recommended products are azimuth determined by using the taste element selected by the customer as the first category, and accordingly, recommended products corresponding to the same taste element are arranged on each half-plane 721, and then the corresponding half-plane ( Each 721 may be displayed in the three-dimensional matrix 72 in such a way that it is arranged in the respective cells 722 and 723 according to popularity or brand name.

As such, within the three-dimensional matrix 72, each recommended product may be classified and arranged according to various criteria. The classification criteria and arrangement method exemplified above are merely exemplary application examples of the present invention and are not limited to such application examples.

Cells 723 and 724 positioned at respective coordinates of each half-plane 721 of the three-dimensional matrix 72 are exemplified as squares, but may be, for example, a circular planar figure or a solid shape such as a sphere.

Furthermore, the size of each cell 722 is illustrated as being substantially the same and may appear different depending on perspective. However, depending on the embodiment, some cell 722 may be relatively larger than other cells 723. For example, the size of the cell 722 may be large depending on the number of duplicates of the searched data, importance, search accuracy, etc. can

One or more 2D display areas 73 to 76 may be formed around the 3D matrix display area 71 to facilitate browsing of the 3D matrix 72 .

For example, on the left side of the 3D matrix 72 , there is a display mode selection window 73 that can selectively hide or show each half-plane of the 3D matrix 72 on the screen. Also, if the customer selects a visual indication, ie, an icon, in any one cell 722 of the three-dimensional matrix 72 (eg, clicking on the cell 722 with the mouse cursor or hovering the mouse cursor over the cell 722 ). uploading operation), meta information display windows 74 and 75 for displaying meta information of data items arranged in the selected cell, that is, a key word and an icon image, are further displayed.

On the right side of the three-dimensional matrix 72, there is a taste display selection window 76 for selecting whether or not to display a taste in the three-dimensional matrix 72. As shown in FIG. For example, the taste display selection window 76 includes a button for reselection from among various taste elements selected by the customer. If the number of recommended products searched for according to the taste element selected by the customer is too large, and it is desired to hide some of the recommended products, if the customer deselects some taste elements, the related products may be removed from the screen. In this case, the products to be removed may be displayed only in cells of the 3D matrix 72 , or the 3D matrix 72 may be rebuilt by excluding the removed products.

Further, although not shown in the drawing, auxiliary information such as a search word used to generate the currently displayed three-dimensional matrix 72, the number of searched data items or the number of data items displayed in the three-dimensional matrix, statistics, and related information are displayed. Two-dimensional auxiliary display areas to be used may be further prepared.

8 is a flowchart illustrating a method of recommending a product, structuring the recommended product data for three-dimensional display, and generating a three-dimensional matrix according to an embodiment of the present invention.

Referring to FIG. 8 , in step S81, various taste elements are analyzed as two emotional attributes with respect to a predetermined kind of taste, for example, color, color combination, emotional modifier, associative image, and texture. A taste map arranged dimensionally is provided, and a product map in which products are arranged on the topography of the taste map is provided.

In step S82, products on the product map corresponding to the taste elements selected by the customer in the taste map are extracted, and the extracted products are determined as recommended products.

In step S83, three categories are selected from categories of meta information describing recommended products. The three categories may be a relatively more important first category, a middle second category and a relatively less important third category. The types of meta-information describing recommended products may be, for example, product name, brand, price, color, material, taste factor, popularity, rating, main characteristics, and the like.

In step S84, with respect to a three-dimensional matrix including virtual half-planes radially formed with respect to the vertical axis, the virtual half-plane determined according to the first category, the vertical axis value determined according to the second category, and the third category determined according to the The three-dimensional matrix is rendered by arranging a visual display of each recommended product in a cell whose coordinates are the horizontal axis values.

9 is a block diagram schematically illustrating a three-dimensional display system for recommended products according to an embodiment of the present invention.

Referring to FIG. 9 , the recommended product 3D display system 100 includes a map DB 101 , a taste selection unit 102 , a product recommendation unit 103 , a data structuring unit 104 , and a matrix generation unit 105 . , a three-dimensional display unit 106 , a two-dimensional display unit 107 , a graphic overlay unit 108 , a display 109 , and a user interface 110 .

The map DB 101 stores a taste map and a product map. The taste selection unit 101 presents an appropriate taste map to the customer through the display 109 , and receives an input of the taste element selected by the customer through the user interface 111 . The product recommendation unit 103 extracts products corresponding to the taste element determined by the taste selection unit 101 from the product map of the map DB 101 and determines the extracted products as recommended products.

The data structuring unit 104 determines the first category, the second category, and the third category based on the categories of meta information describing the recommended products. The matrix generator 105 assigns meta information of each recommended product to cells in the three-dimensional matrix having azimuth, vertical axis, and horizontal axis values as coordinates based on the determined first, second, and third categories. create

The matrix generator 105 transmits the three-dimensional matrix and the key words and visual display of the recommended products assigned to each cell to the three-dimensional display unit 106 and the two-dimensional display unit 107, respectively.

The three-dimensional display unit 106 renders a three-dimensional matrix image in which the visual display information of recommended products is arranged in each cell of the three-dimensional matrix, and the two-dimensional display unit 107 displays meta information corresponding to the coordinates of the cell selected by the customer. Render a 2D auxiliary image based on

The 3D matrix image of the 3D display unit 106 and the 2D auxiliary image of the 2D display unit 107 are synthesized by the graphic overlay unit 108 and displayed on the display 109 .

10 is a system diagram illustrating a three-dimensional mobile recommended product display system according to an embodiment of the present invention.

Referring to FIG. 10 , the mobile recommended product 3D display system 200 includes a terminal 210 , a product recommendation server 220 , and a wired/wireless network 230 .

Through the terminal 210 connected to the product recommendation server 220, the customer can search for a product and browse the recommended product in a three-dimensional matrix.

The terminal 210 may include a 3D display unit 211 , a 2D display unit 212 , a graphic overlay unit 213 , a display 214 , a local recommended product DB 215 , and a user interface 216 .

The terminal 210 first accesses the product recommendation server 220, downloads a taste map, displays it to the customer through the display 214, and when the customer selects a taste element from the taste map through the user interface 216, it It is transmitted to the recommendation server 220 . The three-dimensional display unit 211 renders a three-dimensional matrix image according to the cell coordinates and meta information of the recommended products received through the product recommendation server 220, and the two-dimensional display unit 212 displays the two-dimensional auxiliary display according to the meta information. render the The graphic overlay unit 213 synthesizes the 3D matrix image of the 3D display unit 211 and the 2D auxiliary display of the 2D display unit 212 and displays them on the display 214 . In addition, the recommended products once searched may be stored and managed in the local recommended products DB (215).

The product recommendation server 220 may provide a product recommendation service to a customer connected through the terminal 210 . The product recommendation server 220 may include a map DB 221 , a taste selection unit 222 , a product recommendation unit 223 , a data structuring unit 224 , and a matrix generation unit 225 .

The map DB 221 provides a taste map and a product map, and the taste selection unit 222 transmits the taste map to the terminal 210 and also uses the customer's taste element transmitted from the terminal 210 to the product recommendation unit ( 223) are provided. The product recommendation unit 223 extracts products corresponding to the customer's taste elements from the product map and provides them to the data structuring unit 224 .

The data structuring unit 724 structures the data of the recommended product through the structuring method according to the embodiment of the present invention and provides it to the matrix generating unit 225 . The matrix generator 225 allocates each cell of the three-dimensional matrix to the recommended products according to three categories.

The product recommendation server 220 transmits meta information, cell coordinates, and content information of the recommended products structured in this way to the terminal 210 through the wired/wireless network 230 . The searched and structured meta information of the recommended products may be stored in the recommendation result DB 223 for later use.

As described above, although the present invention has been described with reference to the limited embodiments and drawings, the present invention is not limited to the above embodiments, which are various modifications and Transformation is possible. Therefore, the spirit of the present invention should be understood only by the claims described below, and all equivalent or equivalent modifications thereof will fall within the scope of the spirit of the present invention.

In addition, the device according to the present invention can be implemented as computer-readable codes on a computer-readable recording medium. The computer-readable recording medium includes all types of recording devices in which data readable by a computer system is stored. Examples of the recording medium include ROM, RAM, optical disk, magnetic tape, floppy disk, hard disk, non-volatile memory, etc., and also includes those implemented in the form of a carrier wave (eg, transmission through the Internet). In addition, the computer-readable recording medium is distributed in a network-connected computer system so that the computer-readable code can be stored and executed in a distributed manner.

Claims (4)

providing a taste map two-dimensionally arranged by analyzing taste elements into two emotional attributes, and a product map in which goods are two-dimensionally arranged to correspond to the topography of the taste map;

extracting products corresponding to at least one taste element selected by the customer from the taste map from the product map and determining them as recommended products;

When the first to third categories are determined from among a plurality of categories of meta information describing the recommended products, the azimuth component classified according to the meta information belonging to the first category for each recommended product and belonging to the second category determining a three-dimensional coordinate having a vertical axis component aligned according to meta information and a horizontal axis component aligned according to meta information belonging to the third category;

Rendering a three-dimensional matrix in which the visualization display information for visualizing the recommended product is arranged in the three-dimensional coordinates; and

A three-dimensional display method for recommended products comprising the step of displaying the three-dimensional matrix on a screen.
The method according to claim 1, wherein the taste element comprises at least one of a color, a color combination, a phrase, and an associative image.
The method according to claim 1, wherein the meta information comprises at least one of a product name, brand, price, color, material, taste factor, popularity, rating, and product characteristics.
The three-dimensional display of a recommended product according to claim 1, further comprising the step of displaying the visualization display information and the meta information in an auxiliary display area when the customer selects a specific recommended product in the three-dimensional matrix method.