Method and apparatus for determining skin type

Abstract

A method for determining skin type performed by an electronic device is disclosed. The method for determining skin type includes the steps of: receiving responses to a plurality of survey questions from a user; classifying the plurality of survey questions into a plurality of skin characteristic items and calculating a score corresponding to the response for each skin characteristic item to calculate a sum of scores for each skin characteristic item; determining a first skin type based on the sum of scores for each skin characteristic item; comparing the sum of scores for each skin characteristic item with pre-set threshold scores; providing an additional survey question to the user for at least one skin characteristic item among the plurality of skin characteristic items that exceeds the pre-set threshold scores based on the comparison result, and receiving a response to the additional survey question; changing the first skin type to a second skin type corresponding to the pre-set condition if the response result to the additional survey question satisfies a pre-set condition; and generating a skin type code for the user by combining the skin types determined for each of the plurality of skin characteristic items.

Description

Technology Field

[0001] The present disclosure relates to a method for determining skin type, a method for recommending products based on skin type, and an apparatus. Background Technology

[0002] Technologies for analyzing skin conditions and recommending suitable cosmetics have been developed in various forms. Representative examples include survey-based analysis methods utilized to evaluate a user's skin condition. Survey-based analysis may involve obtaining user responses through questions regarding skin characteristics such as oiliness, moisture, sensitivity, wrinkles, and acne, and determining skin type based on those responses.

[0003] Meanwhile, conventional skin type-based product recommendation technology has been implemented by utilizing the matching relationship between product information stored in a database and skin types. Conventional product recommendation technology has been implemented by pre-setting tags or classification criteria corresponding to specific skin types, selecting products that meet those criteria, and providing them to the user.

[0004] However, survey-based skin analysis technology may be limited in accuracy by the subjectivity of user responses, and simple score aggregation methods have the drawback of failing to adequately reflect the complex relationships between skin characteristics. Furthermore, it has been pointed out that when skin types fall within the boundary range, results determined by a single criterion do not sufficiently reflect the actual skin condition. Additionally, product recommendation technology has been limited in that it often relies on a simple matching relationship between pre-set skin types and products, making it difficult to precisely reflect individual user characteristics. The problem to be solved

[0005] One embodiment aims to provide a skin type determination method and apparatus that can improve the decrease in accuracy of skin type determination caused by the subjectivity of user survey responses.

[0006] One embodiment aims to provide a skin type determination method and apparatus capable of determining skin type more precisely by reflecting the complex relationship between a plurality of skin characteristic items.

[0007] One embodiment aims to provide a skin type determination method and apparatus capable of determining an accurate skin type by correcting the skin condition through additional surveys, even when the skin type is located in a boundary area.

[0008] One embodiment aims to provide a skin type determination method and apparatus that overcome the limitations of a simple matching method between skin type and product and can precisely recommend a product suitable for the user's skin characteristics.

[0009] One embodiment aims to provide a method and apparatus for determining skin type that can improve the accuracy of personalized product recommendations by reflecting the purchase history and product ingredient information of users of the same skin type.

[0010] However, the problems that the present invention aims to solve are not limited to those mentioned above, and may include problems that are not mentioned but can be clearly understood by those skilled in the art from the description below. means of solving the problem

[0011] A method for determining a skin type performed by an electronic device according to one or more embodiments of the present disclosure comprises: receiving responses to a plurality of survey questions from a user; classifying the plurality of survey questions into a plurality of skin characteristic items and calculating a score corresponding to the response for each skin characteristic item to calculate a sum of scores for each skin characteristic item; determining a first skin type based on the sum of scores for each skin characteristic item; comparing the sum of scores for each skin characteristic item with pre-set threshold scores; providing additional survey questions to the user for at least one skin characteristic item among the plurality of skin characteristic items that exceeds the pre-set threshold scores based on the comparison result and receiving a response to the additional survey questions; changing the first skin type to a second skin type corresponding to the pre-set condition when the response result to the additional survey questions satisfies a pre-set condition; and generating a skin type code for the user by combining the skin types determined for each of the plurality of skin characteristic items.

[0012] An electronic device according to one or more embodiments of the present disclosure comprises a processor and a memory connected to the processor, wherein the memory is configured to store a program and the processor is configured to execute the program, and when the program is executed, a skin type determination method is implemented. Effects of the invention

[0013] One embodiment can determine the user's skin condition more precisely by reflecting the survey response results and additional surveys in stages.

[0014] One embodiment can improve accuracy compared to the existing simple score-based method by determining skin type by considering the correlation between multiple skin characteristic items.

[0015] One embodiment can reduce unnecessary additional questions and decrease the burden of user input through a conditional hierarchical survey structure.

[0016] One embodiment can effectively select a product suitable for the user's skin characteristics by reflecting both a skin type code and product ingredient information.

[0017] One embodiment can improve the reliability of personalized product recommendations by utilizing the purchase history of users of the same skin type and AI-based analysis. Brief explanation of the drawing

[0018] FIG. 1 is a drawing for explaining a product recommendation process according to skin type according to one or more embodiments. FIG. 2 is a flowchart illustrating a method for determining skin type according to one or more embodiments. FIG. 3 is a diagram illustrating an example of setting scores corresponding to survey questions and responses according to one or more embodiments. FIG. 4 is a diagram illustrating the process of setting score ranges and threshold values ​​for each skin characteristic item according to one or more embodiments. FIG. 5 is a drawing for explaining skin type codes and skin type information corresponding to each code according to one or more embodiments. FIG. 6 is a diagram illustrating the criteria for determining a skin type code according to the score intervals for each skin characteristic item according to one or more embodiments. FIG. 7 is a drawing for explaining an example of the combination and interpretation of skin type codes according to one or more embodiments. FIG. 8 is a flowchart illustrating a product recommendation method based on a skin type code according to one or more embodiments. FIG. 9 is a drawing for explaining product-specific tag information and skin type suitability information according to one or more embodiments. FIG. 10 is a flowchart illustrating a method for filtering products based on ingredients according to a skin type code according to one or more embodiments. FIG. 11 is a drawing for explaining a product filtering process according to one or more embodiments. FIG. 12 is a drawing for explaining skin type suitability information by ingredient according to one or more embodiments. FIG. 13 is a flowchart illustrating a purchase history-based product sorting method according to one or more embodiments. FIG. 14 is a flowchart illustrating a method for generating multiple recommendation results based on skin type codes according to one or more embodiments. FIG. 15 is a drawing for explaining a skin type code-based recommendation table structure according to one or more embodiments. FIG. 16 is a flowchart illustrating a method for generating skin condition diagnosis text and providing a product according to one or more embodiments. FIG. 17 is a flowchart illustrating an artificial intelligence-based product recommendation method according to one or more embodiments. FIG. 18 is a diagram illustrating the learning and inference process of an artificial intelligence model according to one or more embodiments. FIG. 19 is a flowchart illustrating a method for determining skin type based on additional surveys according to one or more embodiments. FIG. 20 is a flowchart illustrating an additional survey-based skin type determination method according to one or more embodiments. FIG. 21 is a block diagram illustrating the configuration of an electronic device according to one or more embodiments. Specific details for implementing the invention

[0019] The various embodiments described in this specification are illustrative for the purpose of clearly explaining the technical concept of this disclosure and are not intended to limit it to specific embodiments. The technical concept of this disclosure includes various modifications, equivalents, alternatives, and embodiments optionally combined from all or part of each embodiment described in this specification. Furthermore, the scope of the technical concept of this disclosure is not limited to the various embodiments presented below or the specific descriptions thereof.

[0020] Terms used in this specification, including technical or scientific terms, may have the meaning generally understood by those skilled in the art to which this disclosure pertains, unless otherwise defined.

[0021] Expressions used herein such as “comprising,” “may compose,” “possessing,” “possessing,” “having,” and “possessing” imply the existence of the subject feature (e.g., function, operation, or component, etc.) and do not exclude the existence of other additional features. That is, such expressions should be understood as open-ended terms implying the possibility of including a second embodiment.

[0022] In this specification, singular expressions include plural expressions unless the context clearly specifies them as singular. Additionally, plural expressions include singular expressions unless the context clearly specifies them as plural. Throughout the specification, when a part is described as including a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0023] Additionally, the terms 'module' or 'part' as used in the specification refer to software or hardware components, and the 'module' or 'part' performs certain roles. However, the meaning of 'module' or 'part' is not limited to software or hardware. The 'module' or 'part' may be configured to reside in an addressable storage medium or configured to run on one or more processors. Thus, as an example, the 'module' or 'part' may include components such as software components, object-oriented software components, class components, and task components, and at least one of processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, or variables. The components and the functions provided within the 'module' or 'part' may be combined into a smaller number of components and 'modules' or 'parts', or further separated into additional components and 'modules' or 'parts'.

[0024] According to one embodiment of the present disclosure, a ‘module’ or ‘part’ may be implemented as a processor and memory. The term ‘processor’ should be broadly interpreted to include a general-purpose processor, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a controller, a microcontroller, a state machine, etc. In some environments, the term ‘processor’ may refer to an application-specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), etc. The term ‘processor’ may also refer to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors combined with a DSP core, or any other combination of such configurations. Additionally, the term ‘memory’ should be broadly interpreted to include any electronic component capable of storing electronic information. 'Memory' may refer to various types of processor-readable media, such as Random Access Memory (RAM), Read-Only Memory (ROM), Non-Volatile Random Access Memory (NVRAM), Programmable Read-Only Memory (PROM), Erasable-Programmable Read-Only Memory (EPROM), Electrically Erasable PROM (EEPROM), Flash Memory, Magnetic or Optical Data Storage Devices, Registers, etc. If a processor can read information from memory and / or write information to memory, the memory is said to be in an electronic communication state with the processor. Memory integrated into a processor is in an electronic communication state with the processor.

[0025] Expressions such as "first," "second," or "first," "second" as used in this specification are used to distinguish one object from another when referring to a plurality of objects of the same kind, unless otherwise indicated in the context, and do not limit the order or importance of said objects.

[0026] Expressions used herein such as “A, B, and C,” “A, B, or C,” “A, B, and / or C,” or “at least one of A, B, and C,” “at least one of A, B, or C,” “at least one of A, B, and / or C,” “at least one selected from A, B, and C,” “at least one selected from A, B, or C,” “at least one selected from A, B, and / or C,” etc., may mean each of the listed items or all possible combinations of the listed items. For example, “at least one selected from A and B” may refer to (1) A, (2) at least one of A, (3) B, (4) at least one of B, (5) at least one of A and at least one of B, (6) at least one of A and B, (7) at least one of B and A, and (8) all of A and B.

[0027] As used herein, the expression “based on” is used to describe one or more factors affecting an act or action of a decision or judgment described in the phrase or sentence containing such expression, and such expression does not exclude additional factors affecting said act or action of a decision or judgment.

[0028] As used in this specification, the expression that a certain component (e.g., a first component) is "connected" or "connected" to another component (e.g., a second component) may mean that the said certain component is not only directly connected or connected to the said other component, but is also connected or connected through a new other component (e.g., a third component).

[0029] As used herein, the expression "configured to" may have meanings such as "set to," "capable of," "modified to," "made to," or "capable of." Such expression is not limited to the meaning of "specifically designed in hardware," and, for example, a processor configured to perform a specific operation may mean a generic-purpose processor capable of performing that specific operation by executing software.

[0030] Various embodiments of the present disclosure will be described below with reference to the accompanying drawings. In the accompanying drawings and the description thereof, identical or substantially equivalent components may be given the same reference numerals. Furthermore, in the description of the various embodiments below, the description of identical or corresponding components may be omitted, but this does not mean that such components are not included in the embodiments.

[0031] FIG. 1 is a drawing for explaining a product recommendation process according to skin type according to one or more embodiments.

[0032] Referring to FIG. 1, an electronic device according to one embodiment receives a survey response from a user, determines a skin type based on the received survey response, and can recommend a product (30) based on the determined skin type. Here, the electronic device can be implemented as various types of electronic devices such as a mobile phone, a smartphone, a personal computer (PC), a tablet PC, a wearable device, an artificial intelligence (AI) device, etc.

[0033] According to one embodiment, the electronic device may receive responses from a user to multiple survey questions (10) for each of the multiple skin characteristic items. The skin characteristic items serve as criteria for multi-faceted analysis of the user's skin condition and may include distinct characteristics such as oiliness (Oily / Dry), moisture, sensitivity, pores, pigmentation, wrinkles, and acne. Each skin characteristic item may be linked to multiple survey questions (10) for determining the degree of the corresponding characteristic. The skin characteristic items are not limited thereto and may additionally include skin characteristics such as skin elasticity, pore condition, and skin tone uniformity.

[0034] Skin characteristic items may be referred to in various ways, such as skin characteristic indicators, skin characteristic elements, skin characteristic parameters, and skin axes, but in this disclosure, they will be collectively referred to as skin characteristic items.

[0035] According to one embodiment, the electronic device may provide a survey question (10) to the user in correspondence with each skin characteristic item. The survey question (10) may consist of questions for quantitatively evaluating the user's skin condition, and each question may be provided in a manner that selects one or more of a plurality of response items.

[0036] For example, a survey question (10) may consist of a single-choice response or a multiple-choice response. For example, a specific survey question may consist of a question such as "Are pores visible on the skin of the face?" and may include multiple response items such as "Many are visible," "Some are visible," and "Almost not visible."

[0037] According to one embodiment, the electronic device receives multiple survey responses regarding skin condition from a user and can calculate scores for multiple skin characteristic items, such as skin oiliness, moisture, and sensitivity, based on the input responses.

[0038] According to one example, the electronic device determines an initial skin type based on scores for multiple skin characteristic items and can correct the skin type by performing additional surveys if specific conditions are met. For example, the electronic device can provide additional questions to distinguish cases with complex skin characteristics and determine the final skin type based on the response results.

[0039] According to one embodiment, the electronic device can generate a finally determined skin type in the form of a code and recommend a product (30) based on the generated skin type code (20).

[0040] The skin type code (20) may be a string generated by combining values ​​corresponding to each of a plurality of skin characteristic items. For example, the skin type code (20) may be generated by arranging characters or symbols representing the state of each skin characteristic item in a pre-set order. For example, if the skin characteristic items include oiliness, sensitivity, pores, wrinkles, and acne, a skin type code in the form of "DRNWA" may be generated by combining characters representing the state corresponding to each item.

[0041] According to one example, the electronic device may identify multiple products using a tag corresponding to a skin type code (20) and select a product (30) suitable for the user by considering the ingredient information of the products. For example, the electronic device may preferentially select a product containing ingredients suitable for a specific skin type and exclude a product containing unsuitable ingredients.

[0042] According to one embodiment, the electronic device may provide descriptive information regarding skin type. According to one example, the electronic device may generate diagnostic text describing the skin condition to the user based on skin characteristic items included in the skin type code (20) and display it along with recommended products (30). For example, the electronic device may provide descriptions related to moisturizing care or irritation relief for skin types that include characteristics such as dryness or sensitivity.

[0043] According to one embodiment, the electronic device can perform an efficient diagnosis by omitting unnecessary questions during a survey-based skin type determination process and performing only necessary additional surveys based on user responses. According to one example, the electronic device can reduce the burden of user input by controlling it so as not to perform additional questions when conditions are met.

[0044] In the case of conventional technology, there was a problem where the input process was lengthy and accuracy was reduced because all users had to go through the same survey procedure, but the present invention can improve diagnostic efficiency and recommendation accuracy by combining a conditional survey structure and data-based recommendation.

[0045] Below, with reference to the drawings, the electronic device will specifically explain the skin type determination process, the product recommendation process, and the recommendation score calculation process using an artificial intelligence model.

[0046] FIG. 2 is a flowchart illustrating a method for determining skin type according to one or more embodiments.

[0047] Referring to FIG. 2, an electronic device according to one embodiment receives survey responses from a user, calculates scores for each skin characteristic item, and can determine a skin type through additional surveys and condition judgments.

[0048] According to one embodiment, the electronic device can receive responses to multiple survey questions from a user (S210).

[0049] For example, the survey questions may be questions for determining oiliness, moisture level, sensitivity, pore condition, and wrinkle condition. For example, an electronic device may provide questions for checking the degree of skin shine and receive user input from a user for at least one of multiple response items.

[0050] According to one embodiment, the electronic device can classify a plurality of survey questions into a plurality of skin characteristic items, calculate a score corresponding to the response for each skin characteristic item, and calculate a sum of scores for each skin characteristic item (S220).

[0051] According to one embodiment, the electronic device can classify each survey question by mapping it to a predefined skin characteristic item. For example, the electronic device can map the survey question "Does your skin feel tight after washing your face?" to a moisture characteristic item, and the survey question "Does your face become shiny in the afternoon?" to an oil characteristic item.

[0052] For example, the electronic device can be configured to correspond a single survey question to a single skin characteristic item, or to correspond a single survey question to multiple skin characteristic items. For instance, the electronic device can be set to correspond the survey question "Is your skin dry and oily at the same time?" to both oil characteristic items and moisture characteristic items.

[0053] According to one embodiment, the electronic device may calculate the sum of scores for each skin characteristic item by summing the scores of survey questions corresponding to the same skin characteristic item. The sum of scores for each skin characteristic item may be a value representing the degree of the corresponding skin characteristic.

[0054] According to one example, the electronic device can calculate a total score for each skin characteristic item by accumulating the scores obtained from multiple survey questions. For example, if the electronic device obtains 2 points, 3 points, and 1 point respectively from three survey questions corresponding to the oil characteristic item, it can calculate the oil characteristic score as a total of 6 points.

[0055] According to one embodiment, the electronic device can determine a first skin type based on the sum of scores for each skin characteristic item (S230).

[0056] The first skin type may be a skin condition initially determined based on survey responses. The first skin type may be classified according to the magnitude of the score for each skin characteristic item.

[0057] According to one example, the first skin type can be determined in the form of dry, oily, combination, sensitive, etc. For example, an electronic device may determine the skin to be of the dry type if the score for the oil item is low and the score for the moisture item is low.

[0058] According to one embodiment, the electronic device can compare the sum of scores for each skin characteristic item with preset threshold scores (S240).

[0059] The threshold score may be a reference value for determining whether additional judgment is required for a specific skin characteristic item. The threshold score may be set differently for each skin characteristic item.

[0060] According to one embodiment, the electronic device can individually set a threshold score based on the score distribution, importance, or judgment criteria of each skin characteristic item.

[0061] For example, the electronic device may be configured to provide additional survey questions for a skin characteristic item if the sum of the scores for the skin characteristic items is greater than or equal to a threshold score.

[0062] For example, the electronic device may set a threshold score based on statistical information of a user group or the distribution of survey responses. For instance, the electronic device may set a relatively low threshold score for sensitivity characteristics and a relatively high threshold score for oil characteristics.

[0063] For example, the electronic device can be configured to perform an additional survey if the sum of the scores for the sensitivity characteristic items is 5 points or more, and to perform an additional survey if the sum of the scores for the oil characteristic items is 7 points or more.

[0064] According to one embodiment, the electronic device may provide additional survey questions for at least one skin characteristic item that exceeds preset threshold scores among a plurality of skin characteristic items based on comparison results, and receive responses to the additional survey questions (S250).

[0065] Additional survey questions may be designed to assess specific skin characteristic items more precisely. Additional survey questions may be provided when the sum of scores for each skin characteristic item exceeds a threshold score. The additional survey questions may consist of questions designed to differentiate the detailed condition of specific skin characteristic items.

[0066] Additional survey questions may be designed to assess the user's skin condition in greater detail. These additional questions may consist of multiple-choice responses. However, they are not limited to this, and it goes without saying that they may also be in the form of single-choice, ranked-choice, or free-response options.

[0067] For example, if the score of a sensitivity characteristic item exceeds a threshold score, the electronic device may provide additional survey questions to more specifically check for skin irritation reactions. For example, the electronic device may provide questions such as, "Do you experience stinging or redness of the skin when using a specific cosmetic product?"

[0068] According to one embodiment, if the response result to an additional survey question satisfies a preset condition, the electronic device can change the first skin type to a second skin type corresponding to the preset condition (S260).

[0069] The second skin type may be a skin type corrected by reflecting additional survey results. The second skin type may be the result of more precisely modifying the first skin type determined based on initial survey responses. The second skin type may be determined by reflecting additional responses obtained regarding specific skin characteristic items.

[0070] According to one example, if an electronic device determines the first skin type as oily based on the score for oil characteristics, it may change the second skin type to combination type if it is confirmed that dry areas exist in the response results to additional survey questions. For example, the electronic device may determine the skin type as combination if the user simultaneously selects dryness and oiliness in a multiple-choice response.

[0071] According to one embodiment, the electronic device can generate a skin type code for a user by combining skin types determined for each of a plurality of skin characteristic items (S270).

[0072] For example, a skin type code can be generated by arranging the types of each skin characteristic item according to a pre-set order. For instance, an electronic device can generate a single code by combining characters corresponding to items such as oiliness, sensitivity, pores, wrinkles, and acne.

[0073] For example, if the electronic device determines that the oiliness characteristic item is dry (D), the sensitivity characteristic item is resistant (R), the pore characteristic item is non-pore (N), the wrinkle characteristic item is wrinkle-free (T), and the acne characteristic item is present (A), it can generate a skin type code as a string such as "DRNTA".

[0074] For example, an electronic device can generate a skin type code in the form of "OSPWF" by representing oily (O), sensitive (S), pores (P), wrinkled (W), and acne-free (F) as a single character, respectively.

[0075] FIG. 3 is a diagram illustrating an example of setting scores corresponding to survey questions and responses according to one or more embodiments.

[0076] Referring to FIG. 3, an electronic device according to one embodiment can set multiple response items and a score corresponding to each response item for a survey question corresponding to a specific skin characteristic item.

[0077] According to one embodiment, the electronic device can set a survey question as a question corresponding to pore visibility, which is one of the skin characteristic items.

[0078] For example, an electronic device may provide multiple response options for a survey question, such as "many pores are visible," "some pores are visible around the T-zone and cheeks," "pores are only in the T-zone," "none," or "I'm not sure."

[0079] According to one embodiment, the electronic device may set different scores for each response item. It goes without saying that the score for each response item may be set according to the device manufacturing stage or user input.

[0080] For example, the electronic device may assign 4 points for "many pores are visible," 3 points for "some pores are visible," 2 points for "pores are only in the T-zone," 1 point for "none," and 0 points for "I don't know." For example, the electronic device may reflect the score corresponding to the response item selected by the user as the score of the corresponding skin characteristic item.

[0081] According to one embodiment, the electronic device may obtain a score corresponding to a response item selected by a user and reflect the score in the calculation of the total score for each skin characteristic item. According to one example, the electronic device may calculate the total score for each skin characteristic item by accumulating the scores obtained from a plurality of survey questions.

[0082] FIG. 4 is a diagram illustrating the process of setting score ranges and threshold values ​​for each skin characteristic item according to one or more embodiments.

[0083] Referring to FIG. 4, an electronic device according to one embodiment can set survey questions, score ranges, and threshold values ​​for each skin characteristic item, and determine the skin type based on the set threshold values.

[0084] According to one embodiment, the electronic device may set corresponding survey questions and score ranges for each of a plurality of skin characteristic items. Skin characteristic items may include oiliness / dryness, sensitivity, pores, wrinkles / elasticity, and acne. Each skin characteristic item may be linked to one or more survey questions for evaluating the corresponding characteristic.

[0085] According to one example, the electronic device may set multiple survey questions (Q45 to Q49) for oil / dryness items and set minimum and maximum values ​​for response scores for the survey questions. For example, the electronic device may set the score range for oil / dryness items from 3 to 20 points.

[0086] According to one embodiment, the electronic device may set a threshold value for each skin characteristic item. The threshold value may be a reference value for distinguishing skin types based on the sum of scores for each skin characteristic item. The threshold value may be set to a different value depending on the characteristics of each skin characteristic item.

[0087] The threshold value may be used to determine skin type based on which range the sum of scores for each skin characteristic item falls into. The threshold value may be used as a criterion to determine whether additional judgment or further refinement is required for a specific skin characteristic item.

[0088] According to one example, the electronic device may set a threshold value of 13 for the oil / dry item, determine that it is dry (D) if the total score is less than 13, and determine that it is oily (O) if it is 13 or more. For example, the electronic device may determine that it is a dry type if the total score of the oil / dry item is 12, and an oily type if it is 14.

[0089] According to one embodiment, an electronic device can determine a skin type by comparing the sum of scores for each skin characteristic item with a threshold value. The threshold value comparison can be performed by determining whether the sum of scores is greater than or less than the threshold value. The result of the threshold value comparison can be used as a criterion for determining the skin type for each skin characteristic item.

[0090] According to one example, the electronic device may set a threshold value of 8.5 for the sensitivity item, determine that it is resistive (R) if the sum of the scores is 8.5 or less, and determine that it is sensitive (S) if it exceeds 8.5. For example, the electronic device may determine that it is a resistive type if the sum of the sensitivity scores is 7, and a sensitive type if it is 9.

[0091] According to one embodiment, the electronic device determines the state of a corresponding skin characteristic item by using the result of comparing threshold values ​​for each skin characteristic item, and can utilize the determined result to generate a skin type code. According to one example, the electronic device can also determine the skin condition for pore items, wrinkle / elasticity items, and acne items based on threshold values ​​set for each.

[0092] FIG. 5 is a drawing for explaining skin type codes and skin type information corresponding to each code according to one or more embodiments.

[0093] Referring to FIG. 5, an electronic device according to one embodiment can set each code included in the skin type code and the name and attribute information of the skin type corresponding to the code.

[0094] According to one embodiment, the electronic device may set a code value to indicate a skin type determined for each skin characteristic item. The code value may be a character or symbol for identifying each skin type. The code value may be a component for expressing the state of each skin characteristic item when generating a skin type code.

[0095] A skin type code may be a string generated by combining code values ​​corresponding to each of multiple skin characteristic items. The skin type code may be information intended to concisely represent the user's skin condition.

[0096] According to one example, the electronic device can set code values ​​such as oily (O), dry (D), combination (C), and dehydrated oily (M) for oil / dryness characteristic items. For example, the electronic device can assign "O" when the oiliness characteristic is high and "D" when it is dry.

[0097] According to one embodiment, the electronic device may set code values ​​such as sensitivity (S) and resistance (R) for a sensitivity characteristic item. For example, the electronic device may assign "S" when the skin reacts sensitively to external stimuli and assign "R" when the response to stimuli is low.

[0098] According to one embodiment, the electronic device may assign "P" to the pore characteristic item if the pore is prominent and "N" if the pore is not prominent.

[0099] According to one embodiment, the electronic device can set code values ​​for wrinkle / elasticity characteristic items such as "W" when there are many wrinkles, "T" when elasticity is high, and "E" when expression wrinkles are present.

[0100] According to one embodiment, the electronic device can set a code value for an acne characteristic item, such as "A" if acne is present and "F" if acne is not present.

[0101] For example, an electronic device can generate a skin type code by arranging the determined code values ​​for each skin characteristic item according to a preset order. For instance, the electronic device can generate a skin type code such as "OSPWA" by arranging the codes in the order of oiliness / dryness, sensitivity, pores, wrinkles / elasticity, and acne.

[0102] For example, an electronic device can generate a skin type code such as "DRNTF" if it corresponds to dry (D), resistant (R), non-pore (N), elastic (T), and acne-free (F). For instance, the electronic device can combine each code value to represent the user's skin condition as a single string.

[0103] FIG. 6 is a diagram illustrating the criteria for determining a skin type code according to the score intervals for each skin characteristic item according to one or more embodiments.

[0104] Referring to FIG. 6, an electronic device according to one embodiment can set a range of the sum of scores for each skin characteristic item and determine a corresponding skin type code and skin type according to the set score range.

[0105] According to one embodiment, the electronic device may define a score range for each skin characteristic item and set a type ID and code value corresponding to the score range. The score range may be a value representing a range that includes the sum of the scores of a specific skin characteristic item. The score range may be defined by a lower limit value (from) and an upper limit value (to).

[0106] A score range may be range information for classifying skin types based on the sum of scores for each skin characteristic item. Score ranges may be set differently depending on the characteristics of each skin characteristic item. However, not limited thereto, score ranges may be referred to as score ranges, judgment ranges, or classification ranges.

[0107] According to one example, the electronic device may set a score range such that it determines the oily (O) type when the total score for the oil / dry item is 13 or higher, and determines the dry (D) type when it is less than 13. For example, the electronic device may determine the oily type when the oil / dry score is 15, and the dry type when it is 10.

[0108] According to one embodiment, the electronic device may set a score range such that if the sum of the scores for the sensitivity item exceeds 8.5, it is determined to be sensitive (S), and if it is 8.5 or less, it is determined to be resistant (R). For example, the electronic device may determine the sensitive type when the sensitivity score is 9, and the resistant type when it is 7.

[0109] According to one embodiment, the electronic device may set a score range such that if the total score for the pore item exceeds 1.5, it is determined to be in a state where the pore is prominent (P), and if it is 1.5 or less, it is determined to be in a state where the pore is not prominent (N). According to an embodiment, the electronic device may also set a different reference value for the pore item.

[0110] According to one embodiment, the electronic device may set a score range such that if the total score for the wrinkle / elasticity item exceeds 3.5, it is determined to be in a state with many wrinkles (W), and if it is 3.5 or less, it is determined to be in a state with high elasticity (T).

[0111] According to one embodiment, the electronic device may set a score range such that if the total score for the acne item exceeds 0.5, it determines that there is acne (A), and if it is 0.5 or less, it determines that there is no acne (F).

[0112] According to one embodiment, the electronic device may determine the score range to which the sum of scores for each skin characteristic item belongs and select a code value corresponding to the score range. According to one example, the electronic device may generate a skin type code using the selected code value.

[0113] According to one embodiment, the electronic device can consistently determine skin condition using skin type determination criteria based on score ranges. For example, the electronic device can assign the same skin type code to users belonging to the same score range.

[0114] FIG. 7 is a drawing for explaining an example of the combination and interpretation of skin type codes according to one or more embodiments.

[0115] Referring to FIG. 7, an electronic device according to one embodiment can interpret a skin type code generated by combining code values ​​corresponding to each of a plurality of skin characteristic items and provide skin condition information corresponding to the code.

[0116] According to one embodiment, the electronic device can describe the user's skin condition by sequentially interpreting each code value constituting the skin type code.

[0117] Skin condition information may be information indicating the meaning of the skin type corresponding to each code value. Skin condition information may include the correspondence relationship between the code value and the skin condition.

[0118] For example, an electronic device can receive a skin type code such as "ORPTF" and interpret it by separating it into individual code values. For instance, the electronic device can interpret "O" as oily, "R" as resistant, "P" as pore, "T" as tight, and "F" as free of acne.

[0119] According to one embodiment, the electronic device can describe the user's skin condition by combining the skin conditions corresponding to each code value. According to one example, the electronic device can interpret the "ORPTF" code as "oily, resistant, pore-containing, elastic, and acne-free condition."

[0120] For example, an electronic device can interpret a skin type code such as "DSNWF" as "dry, sensitive, poreless, wrinkled, and acne-free." For instance, the electronic device can combine the meanings of each code value to provide the skin condition in the form of a single sentence.

[0121] According to one embodiment, an electronic device can interpret a code combination representing a complex skin condition to provide intuitive skin condition information to the user. According to one example, the electronic device can interpret a skin type code such as "MSPWA" as "a condition of dehydrated, oily, sensitive, pore-filled, wrinkle-filled, and acne-prone skin."

[0122] FIG. 8 is a flowchart illustrating a product recommendation method based on a skin type code according to one or more embodiments.

[0123] Referring to FIG. 8, an electronic device according to one embodiment can identify a product using a tag corresponding to a skin characteristic item included in a skin type code, filter the product by reflecting ingredient information, and then provide it to a user.

[0124] According to one embodiment, the electronic device can identify a plurality of products including tags corresponding to skin characteristic items included in a skin type code from a database (S810).

[0125] The database may be a data structure that stores correspondence relationships between skin characteristic items, tag information, and product information. The database may be a database in which one or more tag information corresponding to each of the multiple skin characteristic items included in the skin type code and multiple product information corresponding to each of the tag information are mapped and stored.

[0126] For example, the electronic device can set tags such as “moisturizing,” “hydrating,” and “hypoallergenic” for dry (D) skin characteristics and map multiple product information, such as creams, serums, and lotions containing the tags, to a database.

[0127] According to one example, the electronic device can set tags such as "fragrance-free," "hypoallergenic," and "skin soothing" for the sensitive (S) skin characteristic item, and map product information containing the corresponding tags to store them in a database.

[0128] According to one example, an electronic device can look up tags corresponding to each skin characteristic item included in a skin type code and identify products containing those tags from a database. For example, the electronic device can identify products containing tags corresponding to "oily (O)" and products containing tags corresponding to "sensitive (S)" together.

[0129] According to one embodiment, the electronic device can filter multiple products based on a skin type code and ingredient information of multiple products (S820).

[0130] Ingredient information may be data containing information on the types and amounts of ingredients included in each product. Ingredient information may serve as reference information for selecting products suitable for a user's skin type.

[0131] For example, an electronic device may set included and excluded ingredients corresponding to skin characteristic items included in a skin type code, and filter products according to the criteria. For instance, for sensitive skin types, the electronic device may exclude products containing irritating ingredients and select products containing moisturizing ingredients.

[0132] According to one embodiment, the electronic device can provide a filtered product to the user (S830).

[0133] The electronic device can display filtered products on a screen or provide them through a user interface. The electronic device can sort and provide filtered products according to certain criteria.

[0134] For example, an electronic device may provide skin condition description information corresponding to a skin type code along with filtered products. For instance, the electronic device may display recommended products along with text describing characteristics based on the user's skin type code.

[0135] FIG. 9 is a drawing for explaining product-specific tag information and skin type suitability information according to one or more embodiments.

[0136] Referring to FIG. 9, an electronic device according to one embodiment can set a plurality of tag information for each product and determine whether it is suitable for a skin type code based on the set tags.

[0137] According to one embodiment, the electronic device may assign one or more tags per product. The tag may be identification information indicating the characteristics, functions, or ingredients of the product. The tag may be reference information for determining which skin characteristic category the product is suitable for.

[0138] Tag information may be data representing the relationship between a product and skin characteristic items. Tag information may be expressed in the form of numbers, characters, or symbols. However, it is not limited thereto, and tag information may be referred to as a classification code, attribute information, or feature identifier.

[0139] According to one example, an electronic device may set multiple tags for a specific product. For example, the electronic device may set multiple tag values ​​for the "HEARTLEAF 77% SOOTHING TONER" product and configure the product to correspond to various skin characteristic items.

[0140] According to one embodiment, the electronic device can set a suitable skin type for each product based on tag information. For example, the electronic device can set a specific product based on tag information to be suitable for oily (O), combination (C), sensitive (S), wrinkles (W), and pores (P) characteristics, and not suitable for dry (D) and acne (A) characteristics. For example, the electronic device may include a product in the recommendation list if it satisfies some characteristic items included in a specific skin type code according to a tag combination.

[0141] According to one embodiment, the electronic device can set a product containing tags corresponding to all skin characteristic items as a product suitable for all skin types. According to one example, the electronic device can classify a product containing multiple tags, such as a "SOOTHING TRIAL KIT," as a product suitable for all skin types.

[0142] For example, an electronic device may preferentially select products containing tags that match skin characteristic items included in a skin type code.

[0143] FIG. 10 is a flowchart illustrating a method for filtering products based on ingredients according to a skin type code according to one or more embodiments.

[0144] Referring to FIG. 10, an electronic device according to one embodiment can set included and excluded ingredients based on skin characteristic items included in a skin type code, and filter products according to the criteria.

[0145] According to one embodiment, the electronic device can set the included and excluded ingredients based on skin characteristic items corresponding to a skin type code (S1010).

[0146] Included ingredients may refer to ingredients suitable for the user's skin type. Excluded ingredients may refer to ingredients unsuitable for the user's skin type.

[0147] Each of the included and excluded ingredients may serve as reference information to distinguish between recommended and non-recommended ingredients based on each skin characteristic category.

[0148] According to one example, the electronic device may set moisturizing ingredients as included ingredients and irritating ingredients as excluded ingredients for dry (D) skin types. For example, the electronic device may set moisturizing ingredients such as hyaluronic acid as included ingredients and alcohol as excluded ingredients.

[0149] According to one embodiment, the electronic device can exclude products containing excluded components from a plurality of products identified from a database and select and filter products containing included components (S1020).

[0150] Filtering can be performed by comparing the product's ingredient information with established ingredient standards.

[0151] For example, an electronic device may remove a product from the filtering list if that product contains an excluded ingredient. For instance, the electronic device may exclude products containing irritating ingredients for sensitive (S) skin types.

[0152] For example, an electronic device can prioritize the selection of products containing included ingredients and set them as recommended products. For instance, the electronic device can select a product containing moisturizing ingredients as a suitable product for users with dry skin.

[0153] According to one embodiment, the electronic device can provide a product suitable for the user's skin type by selecting products based on criteria for included and excluded ingredients.

[0154] FIG. 11 is a drawing for explaining a product filtering process according to one or more embodiments.

[0155] Referring to FIG. 11, an electronic device according to one embodiment can set ingredient standard information such as the number of included ingredients, the number of product name keywords, and the number of excluded ingredients for each tag and type corresponding to each skin characteristic item.

[0156] According to one embodiment, the electronic device may set the number of included ingredients corresponding to each skin characteristic item or type. The number of included ingredients may be a value representing the number of ingredients suitable for the corresponding skin characteristic item. The number of included ingredients may be used as a criterion for included ingredients when filtering products.

[0157] The number of included ingredients may be information indicating the size of the ingredient list suitable for a specific skin type. The number of included ingredients may be set differently for each skin characteristic category. However, it is not limited to these, and the number of included ingredients may be referred to as the recommended number of ingredients, the standard number of ingredients, or the standard number of ingredients.

[0158] For example, the electronic device can set a relatively large number of included ingredients for the dry type. For instance, the electronic device can set a number of ingredients related to moisturizing components as included ingredients.

[0159] According to one embodiment, the electronic device may set the number of product name keywords. The number of product name keywords may be a value representing the number of specific keywords included in the name of the product. The number of product name keywords may be reference information for reflecting the function or characteristics of the product.

[0160] For example, an electronic device can set the number of product name keywords to prioritize the selection of products containing keywords such as "soothing" or "calming" for the Sensitive type. For instance, the electronic device may determine a product as suitable if its product name contains specific keywords.

[0161] According to one embodiment, the electronic device may set the number of excluded ingredients. The number of excluded ingredients may be a value representing the number of ingredients unsuitable for a specific skin type. The number of excluded ingredients may be used as an exclusion criterion when filtering products.

[0162] The number of excluded ingredients may be information indicating the range of ingredients restricted for specific skin characteristic categories. The number of excluded ingredients may be set differently for each skin characteristic category. However, not limited thereto, the number of excluded ingredients may be referred to as the number of prohibited ingredients, the number of exclusion criteria, or the number of restricted ingredients.

[0163] For example, an electronic device may set multiple ingredients that can cause irritation to the sensitive type as excluded ingredients. For instance, the electronic device may determine a product containing ingredients such as alcohol or fragrances as an excluded product.

[0164] According to one embodiment, the electronic device can filter products using the number of included ingredients, the number of product name keywords, and the number of excluded ingredients set for each skin characteristic item. According to one example, the electronic device can select and provide to the user a product that satisfies the criteria for included ingredients and does not satisfy the criteria for excluded ingredients.

[0165] FIG. 12 is a drawing for explaining skin type suitability information by ingredient according to one or more embodiments.

[0166] Referring to FIG. 12, an electronic device according to one embodiment can set an inclusion type and an exclusion type for each component and configure product filtering criteria using the information.

[0167] According to one embodiment, the electronic device can set an inclusion type for each component. For example, the electronic device can set the Rice component to include types of complex (C), dry (D), and wrinkle (W). For example, if the Rice component has a moisturizing effect, the electronic device can set it as a component suitable for dry and wrinkle-related skin types.

[0168] According to one embodiment, the electronic device can set exclusion types for each component. For example, the electronic device can set dry (D) and wrinkle (W) types as exclusion types for the Teatree component. For example, if the Teatree component has a sebum-regulating function, the electronic device can set it as a component unsuitable for dry skin.

[0169] According to one embodiment, the electronic device may set semantic information for each component. The semantic information may be information explaining the function, effect, or purpose of application of the corresponding component. The semantic information may be explanatory information providing the basis for setting inclusion types and exclusion types.

[0170] For example, an electronic device may set Retinol as an ingredient with anti-aging effects and determine that it may irritate sensitive skin, and store the corresponding semantic information. For instance, the electronic device may set Retinol as an ingredient suitable for wrinkle improvement but unsuitable for sensitive skin.

[0171] According to one embodiment, an electronic device can determine whether the composition of ingredients in a product is compatible with a skin type code by using information on the inclusion type and exclusion type for each ingredient. According to one example, the electronic device can select a product if the ingredients included in a specific product correspond to the inclusion type of the user's skin type, and exclude the product if they correspond to the exclusion type.

[0172] FIG. 13 is a flowchart illustrating a purchase history-based product sorting method according to one or more embodiments.

[0173] Referring to FIG. 13, an electronic device according to one embodiment can sort and provide filtered products using purchase history data of users corresponding to the same skin type code as the user and purchase history data of all users.

[0174] According to one embodiment, the electronic device can calculate the purchase rate of filtered products based on purchase history data of users corresponding to the same skin type code as the skin type code (S1310).

[0175] The purchase rate can be a value representing the proportion of users corresponding to a specific skin type code who have purchased a particular product. The purchase rate can be an indicator reflecting product preference within a group of users of the same skin type.

[0176] The purchase rate may be a value calculated by dividing the number of purchases or purchasing users of a specific product by the total number of users of the same skin type. The purchase rate may be calculated differently for each product. However, it is not limited to these methods, and the purchase rate may be referred to as a purchase ratio by type, a user group-based purchase indicator, or a type-based preference indicator.

[0177] For example, if 30 out of 100 users of a specific skin type code purchase a specific product, the electronic device can calculate the purchase rate of that product as 0.3.

[0178] According to one embodiment, the electronic device can calculate the popularity of filtered products based on the purchase history data of all users (S1320).

[0179] Popularity can be a value indicating how much a specific product has been purchased by the entire user group. Popularity can be an indicator reflecting the general preference for a product.

[0180] Popularity may be a value calculated by dividing the total number of purchases or purchasing users of a specific product by the total number of users. Popularity may be expressed quantitatively for the purpose of comparison between products. However, it is not limited to these, and popularity may be referred to as the total purchase ratio, overall user preference, or product popularity indicator.

[0181] According to one example, if 200 out of 1,000 total users purchase a specific product, the electronic device can calculate the popularity of that product as 0.2.

[0182] According to one embodiment, the electronic device may normalize the purchase rate and popularity, respectively, and apply a first weight and a second weight to the normalized purchase rate and normalized popularity, respectively, to calculate a weighted sum score (S1330).

[0183] Normalization can be the process of converting values ​​with different ranges into a fixed range. A weighted sum score can be a value that calculates a final score by combining multiple indicators.

[0184] The weighted sum score may be calculated by summing the value obtained by multiplying the normalized purchase rate by the first weight and the value obtained by multiplying the normalized popularity by the second weight. The first weight and the second weight may be values ​​reflecting different levels of importance. However, not limited thereto, the weighted sum score may be referred to as a combined score, a composite score, or a final evaluation score.

[0185] According to one example, an electronic device can calculate a weighted sum score by applying a first weight of 0.9 to the purchase rate and a second weight of 0.1 to the popularity.

[0186] According to one embodiment, the electronic device may set the first weight to be greater than the second weight so that the purchase history data of users corresponding to the same skin type code as the user is reflected more preferentially. Since the purchase rate may be an indicator reflecting the actual purchasing behavior of users with the same skin type, it can be used as a criterion for providing products more suitable for individual users.

[0187] For example, electronic devices may prioritize the purchasing trends of a group of users with the same skin type over popularity among the entire user base. For instance, even if a particular product is popular overall, an electronic device may set a lower recommendation ranking if the purchase rate among users with the same skin type is low.

[0188] According to one embodiment, the electronic device may set a first weight greater than a second weight to perform personalized recommendations that reflect skin type specificity. The electronic device may prioritize providing products that are more suitable for the user and relatively reduce the influence of generally popular products regardless of skin type.

[0189] According to one embodiment, the electronic device can sort and provide filtered products to the user based on a weighted sum score (S1340).

[0190] Sorting can be the process of arranging products in order of highest weighted sum score. The sorting results can be provided to the user in the form of a recommended product list.

[0191] For example, electronic devices can prioritize exposure to users by placing products with high weighted sum scores at the top.

[0192] FIG. 14 is a flowchart illustrating a method for generating multiple recommendation results based on skin type codes according to one or more embodiments.

[0193] Referring to FIG. 14, an electronic device according to one embodiment can generate multiple recommendation results based on the entire skin type code and individual type codes for each skin characteristic item, and can provide the multiple recommendation results to a user by distinguishing them.

[0194] According to one embodiment, the electronic device can identify a first product purchased by users having the same skin type code as the skin type code among filtered products based on purchase history data of users corresponding to the same skin type code as the skin type code, and generate the first product as a first recommendation result (S1410).

[0195] The first recommendation result may be a recommendation result that reflects the purchasing trends of a user group with the same entire skin type code. The first recommendation result may be a recommendation result based on purchase history at the skin type code unit, which is a combination of multiple skin characteristic items. The first recommendation result may be a recommendation result that reflects actual purchase data of a user group having the same skin type as the user. However, not limited thereto, the first recommendation result may be referred to as an overall type-based recommendation result, an integrated type recommendation result, or a user group-based recommendation result.

[0196] For example, if a specific skin type code is "DSNWF", the electronic device can generate a first recommendation result for products purchased by users with the same code value.

[0197] For example, an electronic device can select products based on the purchase history data of users who satisfy all of the conditions, considering that the "DSNWF" code is a combination corresponding to dry (D), sensitive (S), non-pore (N), wrinkles (W), and acne-free (F). For instance, the electronic device can prioritize identifying products that contain moisturizing ingredients, are low-irritation, do not clog pores, and have wrinkle-improving functions.

[0198] According to one example, the electronic device may configure a first recommendation result by selecting top products based on the number of purchases or the purchase ratio of products repeatedly purchased within a group of users with the same skin type code. For example, the electronic device may place a moisturizing cream or a hypoallergenic serum purchased by a majority of users among 200 users with the same "DSNWF" as a top item in the first recommendation result.

[0199] For example, an electronic device can highlight products in a specific category by reflecting the purchasing patterns of users with the same skin type code. For instance, if "DSNWF" users purchase more moisturizing products than cleansers, the electronic device can configure the first recommendation result centered on moisturizing products.

[0200] According to one embodiment, the electronic device can identify a second product among filtered products based on purchase history data of users corresponding to each individual type code corresponding to each skin characteristic item included in the skin type code, and generate the second product as a second recommendation result (S1420).

[0201] The second recommendation result may be a recommendation result that reflects purchasing trends at the level of skin characteristic items. The second recommendation result may be a recommendation result generated independently for each skin characteristic item included in the skin type code. The second recommendation result may be a recommendation result generated based on users' purchase history regarding specific skin characteristic items. However, it is not limited thereto, and the second recommendation result may be referred to as an individual item-based recommendation result, a detailed type recommendation result, or an item-specific recommendation result.

[0202] According to one example, the electronic device can generate a product suitable for sensitive skin as a second recommendation result based on purchase history data of users corresponding to sensitivity (S) characteristics.

[0203] According to one example, the electronic device may identify products repeatedly purchased by users with sensitivity (S) characteristics and include those products preferentially in the second recommendation results. For example, the electronic device may generate toner, serum, and cream products containing hypoallergenic ingredients or having skin-soothing functions as the second recommendation results.

[0204] For example, an electronic device can set recommendation priorities based on products that are frequently purchased or have a high purchase rate among users with sensitive skin. For instance, the electronic device can place fragrance-free products or products with low skin irritation that are purchased by a large number of 150 users with sensitive skin as top recommendation items.

[0205] For example, an electronic device can select products based on specific ingredients related to sensitivity characteristics. For instance, the electronic device may determine that a product containing skin-soothing ingredients such as centella, aloe, or panthenol is suitable for sensitive skin and generate a second recommendation result.

[0206] According to one embodiment, the electronic device may provide the user with a first recommendation result and a second recommendation result separately (S1430).

[0207] According to one example, the electronic device may display a first recommendation result as "recommendation for users with the same skin type as me" and provide a second recommendation result classified as "recommendation by skin characteristics."

[0208] According to one example, the electronic device may provide a first recommendation result and a second recommendation result by separating them into different display areas. For example, the electronic device may place the first recommendation result in the upper area and the second recommendation result in the lower area on the user interface to provide them in a separated form.

[0209] According to one example, the electronic device may provide a first recommendation result and a second recommendation result with different labels or titles. For example, the electronic device may display the title "Recommendation based on all skin types" on the first recommendation result and the title "Personalized recommendation by skin characteristics" on the second recommendation result, so that the user can intuitively distinguish them.

[0210] For example, the electronic device may provide second recommendation results subdivided by skin characteristic categories. For instance, the electronic device may organize separate lists for each skin characteristic category, such as sensitivity (S), wrinkles (W), and acne (A), and display products suitable for the corresponding characteristic in each list.

[0211] According to one example, the electronic device may provide a first recommendation result and a second recommendation result by applying different sorting criteria to each other. For example, the electronic device may sort the first recommendation result based on the purchase rate of users of the same skin type, and sort the second recommendation result based on the purchase frequency for a specific skin characteristic item.

[0212] FIG. 15 is a drawing for explaining a skin type code-based recommendation table structure according to one or more embodiments.

[0213] Referring to FIG. 15, an electronic device according to one embodiment can configure a plurality of recommendation tables using different keys depending on the expression method of the skin type code.

[0214] According to one embodiment, the electronic device may configure a first recommendation table using the entire skin type code as a key. The first recommendation table may be a data structure that stores the correspondence relationship between the entire skin type code and a product. The first recommendation table may be a structure for matching products at the code unit where a plurality of skin characteristic items are combined.

[0215] The first recommendation table may include a key such as a skin type code combined with multiple skin characteristic items.

[0216] For example, an electronic device can use a skin type code, such as "ORPTA" ​​or "DSNWF," as a key to store product information that matches the code. For instance, the electronic device can store products preferred by users with the same skin type code in a first recommendation table.

[0217] According to one embodiment, the electronic device may configure a second recommendation table using individual skin characteristic items included in a skin type code as a key. The second recommendation table may be a data structure that stores the correspondence relationship between individual skin characteristic items and products. The second recommendation table may be a structure for matching products at the unit of specific skin characteristic items.

[0218] The second recommendation table may include keys representing individual skin characteristic items.

[0219] For example, an electronic device can store product information suitable for the corresponding skin characteristic item using individual type codes such as "O", "S", "P", "W", and "A" as keys. For example, the electronic device can store products corresponding to the sensitivity (S) code in a second recommendation table.

[0220] According to one embodiment, the electronic device may provide customized recommendations to a user by utilizing a first recommendation table and a second recommendation table together. According to one example, the electronic device may provide precise recommendations for all skin types using the first recommendation table and provide complementary recommendations for specific skin characteristic items using the second recommendation table.

[0221] FIG. 16 is a flowchart illustrating a method for generating skin condition diagnosis text and providing a product according to one or more embodiments.

[0222] Referring to FIG. 16, an electronic device according to one embodiment can generate a diagnostic text describing the skin condition of a user based on a skin type code and provide it to the user along with a filtered product.

[0223] According to one embodiment, the electronic device can generate a diagnostic text describing the skin condition of the user based on a skin characteristic item included in a skin type code (S1610).

[0224] The diagnostic text may be text that expresses the user's skin condition in sentence form by interpreting the type of each skin characteristic item included in the skin type code. The diagnostic text may be structured to include the characteristics of the skin condition, care requirements, and precautions.

[0225] For example, an electronic device can generate natural language sentences by combining types for each skin characteristic item. For instance, if the skin type code is "ORPTF," the electronic device can generate diagnostic text such as, "The electronic device determines that the user's skin is oily, resistant, has pores, is elastic, and is acne-free."

[0226] According to one example, the electronic device may pre-store different descriptive sentences for each skin characteristic item and generate diagnostic text by combining the corresponding descriptive sentences. For example, the electronic device may generate a single integrated diagnostic sentence by combining a descriptive sentence corresponding to sensitivity (S) and a descriptive sentence corresponding to wrinkles (W).

[0227] According to one embodiment, the electronic device can provide filtered product and diagnostic text to the user (S1620).

[0228] For example, an electronic device can display diagnostic text along with product recommendation results on a user interface. For instance, the electronic device can display diagnostic text describing the user's skin condition at the top of the screen and provide a list of products suitable for that skin condition at the bottom of the screen.

[0229] For example, an electronic device can provide diagnostic text linked with product information. For instance, the electronic device can highlight and provide hypoallergenic products suitable for sensitive skin along with a diagnostic statement such as "Determined to be sensitive skin."

[0230] For example, an electronic device can extend diagnostic text into user-customized guidance information. For instance, the electronic device can provide the user with care guides, such as "Since you have dry skin, the use of moisturizing products is recommended."

[0231] FIG. 17 is a flowchart illustrating an artificial intelligence-based product recommendation method according to one or more embodiments.

[0232] Referring to FIG. 17, an electronic device according to one embodiment can calculate a product recommendation score using an artificial intelligence model with skin type codes and filtered product ingredient information, and sort products based on the calculated recommendation score and provide them to the user.

[0233] According to one embodiment, the electronic device can input skin type code and ingredient information of filtered products into an artificial intelligence model to obtain a recommendation score for each of the filtered products (S1710).

[0234] The artificial intelligence model may be a model that calculates recommendation scores by reflecting the relationship between skin type codes and product ingredients. The artificial intelligence model may be a model configured to reflect the degree of fit between skin characteristic items and ingredients based on training data.

[0235] For example, an electronic device may apply a machine learning model or a deep learning model as an artificial intelligence model. For instance, the electronic device may use a multilayer neural network structure to receive skin type codes and ingredient vectors as inputs and calculate the suitability of each product in the form of a score.

[0236] A detailed explanation of the artificial intelligence model will be provided in Fig. 18.

[0237] According to one embodiment, the electronic device can sort filtered products based on recommendation scores and provide the sorted products to the user (S1720).

[0238] For example, an electronic device can present products to a user by sorting them in order of highest recommendation score. For instance, the electronic device can display products with higher recommendation scores preferentially to the user by placing them at the top of the list.

[0239] For example, an electronic device may classify and provide products based on recommendation scores. For instance, the electronic device may classify and display the top 10% of products as "Recommended Products," middle-tier products as "General Recommendations," and lower-tier products as "Reference Products."

[0240] For example, an electronic device can provide the user with a reason for recommendation along with a recommendation score. For instance, the electronic device can display a description for a specific product such as "contains ingredients suitable for sensitive skin."

[0241] FIG. 18 is a diagram illustrating the learning and inference process of an artificial intelligence model according to one or more embodiments.

[0242] Referring to FIG. 18, an electronic device according to one embodiment can use skin type code (1830) and product ingredient information (1840) as input data (1810), use purchase history data as correct answer data (1820) to train an artificial intelligence model (200), and use the trained artificial intelligence model (200) to calculate a product recommendation score (1850).

[0243] According to one embodiment, the electronic device can input input data (1810) including a skin type code (1830) and product ingredient information (1840) into an artificial intelligence model (200) to train it.

[0244] The input data (1810) may be a data structure for simultaneously reflecting user skin characteristics and product characteristics.

[0245] According to one example, the electronic device may convert the skin type code (1830) into a vector form and configure it as input data (1810). For example, the electronic device may input a skin type code such as "OSPTA" by converting it into a binary or numeric value for each characteristic item.

[0246] According to one example, the electronic device may quantify the product's component information (1840) and configure it as input data (1810). For example, the electronic device may generate and input a component vector based on whether a specific component is included or its content.

[0247] According to one embodiment, the electronic device can train an artificial intelligence model (200) using correct answer data (1820) including purchase history data.

[0248] The correct answer data (1820) may be data that reflects product information actually purchased by users with a specific skin type.

[0249] Depending on the example, the electronic device can set the purchase status, purchase frequency, or repurchase status as the correct answer value. For instance, the electronic device can assign a high correct answer value if a specific product is frequently purchased by users of the same skin type.

[0250] For example, an electronic device can obtain the number of purchases per product by utilizing purchase history data of users with the same skin type code. For instance, the electronic device can aggregate the number of purchases per product for a group of users belonging to a specific skin type code.

[0251] According to one example, an electronic device can calculate a purchase rate per product based on the number of purchases per product and the total number of purchases for each product. For example, the electronic device can calculate the purchase rate by dividing the number of purchases of a specific product within a group of users with the same skin type code by the total number of purchases of the product.

[0252] For example, an electronic device can acquire popularity for each product. Popularity may be a value representing the degree of product response within the entire user group or the overall sales trend.

[0253] For example, an electronic device can generate a standard recommendation score for each product based on purchase rate and popularity. For instance, the electronic device can generate the standard recommendation score for each product by normalizing the purchase rate and popularity and then weighting and summing them. Accordingly, products that are purchased relatively frequently within a user group of the same skin type and simultaneously have high popularity may be assigned a higher standard recommendation score.

[0254] According to one example, an electronic device may generate a training dataset by mapping a skin type code (1830), product ingredient information (1840), and product-specific standard recommendation scores to each other. In this case, the skin type code (1830) and product ingredient information (1840) may constitute input data (1810), and the product-specific standard recommendation scores may constitute correct answer data (1820).

[0255] According to one embodiment, the electronic device can train an artificial intelligence model (200) using input data (1810) and correct answer data (1820).

[0256] The artificial intelligence model (200) may be a model that predicts product suitability by learning the relationship between the skin type code (1830) and product ingredient information (1840). The artificial intelligence model (200) may be a model trained to numerically evaluate whether a specific product is suitable for the skin condition by reflecting the correlation between each skin characteristic item included in the skin type code (1830) and the product ingredient information (1840).

[0257] Here, the fact that the artificial intelligence model (200) is trained means that a basic artificial intelligence model (e.g., an artificial intelligence model containing arbitrary random parameters) is trained using a number of training data by a learning algorithm, thereby creating a predefined operation rule or artificial intelligence model set to perform a desired characteristic (or purpose). The training may be performed through a separate server and system, but is not limited thereto and may also be performed on an electronic device. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples mentioned above.

[0258] Here, the artificial intelligence model (200) may be a generative artificial intelligence. For example, the artificial intelligence model may be implemented as a Transformer-based Large Language Model (LLM). The artificial intelligence model may be implemented as a server that communicates remotely with the electronic device or as an on-device implementation within the electronic device. According to an embodiment, a skin type determination method, a product recommendation method, or a recommendation score calculation method may be implemented using at least one artificial intelligence model (200). The artificial intelligence model (200) may be implemented as a regression model, a classification model, a neural network-based recommendation model, etc., and is not necessarily limited to a specific architecture. Depending on the purpose, the artificial intelligence model (200) may be implemented as a generative artificial intelligence, a model having a natural language recognition processing structure, a generative AI prompt engine, or a combination thereof. For example, AI models such as YOLO (You Only Look Once) based Detection-based OCR models, CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), RBM (Restricted Boltzmann Machine), DBN (Deep Belief Network), BRDNN (Bidirectional Recurrent Deep Neural Network), or Deep Q-Networks can be further utilized.

[0259] According to one example, the electronic device may adjust the parameters of the model to minimize the error between the input data (1810) and the correct data (1820) during the learning process. For example, the electronic device may perform learning using a loss function to reduce the difference between the predicted value and the correct value.

[0260] According to one embodiment, the electronic device can obtain a recommendation score for each product using a learned artificial intelligence model (200).

[0261] The artificial intelligence model may be a model trained to calculate the number of purchases and purchase ratios for each product based on the purchase history data of users with the same skin type code as the skin type code, and to calculate recommendation scores for each product based on the number of purchases and purchase ratios.

[0262] For example, an artificial intelligence model can aggregate the number of purchases per product based on a user group belonging to a specific skin type code, and calculate the purchase ratio by calculating the ratio of an individual product to the total number of purchases. For example, if a specific product is purchased 100 times and the total number of purchases is 1,000 in a user group with the skin type code "OSPTA", the electronic device can set the purchase ratio of that product to 0.1.

[0263] According to one example, an electronic device can train an artificial intelligence model using product-specific standard recommendation scores generated based on the purchase frequency, purchase rate, and popularity as correct answers. For example, the electronic device can assign higher standard recommendation scores to products that have a high purchase frequency, a high purchase rate, and high popularity within a group of users with the same skin type code.

[0264] According to one example, the artificial intelligence model can be trained to predict the standard recommendation score for each product by receiving the skin type code (1830) and product ingredient information (1840) as input.

[0265] For example, an artificial intelligence model can be trained to reflect the relative importance of purchase frequency, purchase ratio, and popularity reflected in the generation of the above-mentioned standard recommendation score.

[0266] For example, an artificial intelligence model can be trained to calculate recommendation scores by utilizing purchase frequency and purchase rate as input features. For instance, an electronic device model can be configured to assign high recommendation scores to products with a high purchase frequency and high purchase rate.

[0267] For example, an artificial intelligence model can calculate a recommendation score by reflecting the relative importance between the number of purchases and the purchase rate. For instance, an electronic device can give higher weight to the number of purchases to prioritize recommending products that have been repeatedly selected by multiple users.

[0268] For example, an artificial intelligence model can be trained to reflect changes in purchase history over time. For instance, an electronic device can generate recommendation scores that reflect the latest user preference trends by assigning higher importance to products purchased within a recent period.

[0269] According to one example, the electronic device can input a skin type code (1830) and product ingredient information (1840) into an artificial intelligence model (200) to calculate a product recommendation score (1850). For example, when a specific skin type code and specific product ingredient information are input, the electronic device can output a score indicating the degree to which the product is suitable for the user.

[0270] According to one example, the product recommendation score (1850) may be expressed as a normalized value. For example, the electronic device may be configured to set the product recommendation score (1850) to a range of 0 to 1 to facilitate comparison between products.

[0271] According to one example, the product recommendation score (1850) may be calculated by reflecting the contribution of each skin characteristic item. For example, the electronic device may calculate a suitability score for the sensitivity item, a suitability score for the oiliness item, and a suitability score for the acne item, respectively, and determine the final product recommendation score (1850) by weighted summing them.

[0272] According to one example, the product recommendation score (1850) may be determined differently depending on the type, content, and combination of ingredients included in the product. For example, even if the electronic device contains the same moisturizing ingredients, it may give a higher score to a product with a higher content.

[0273] According to one example, the product recommendation score (1850) may be a value that reflects user group-based information. For example, the electronic device may reflect a higher score for products that are highly preferred in the purchase history data of users with the same skin type code.

[0274] According to one example, the product recommendation score (1850) can be used as a reference value for determining priority among products. For example, the electronic device can sort products in order of highest product recommendation score (1850) and provide them to the user.

[0275] According to one example, the electronic device can compare multiple products using the calculated product recommendation score (1850) and provide the product suitable for the user in priority.

[0276] FIG. 19 is a flowchart illustrating a method for determining skin type based on additional surveys according to one or more embodiments.

[0277] Referring to FIG. 19, an electronic device according to one embodiment can determine whether to apply additional survey questions based on the sum of scores for each skin characteristic item, determine the skin type based on the additional survey results, and control whether to provide a subsequent survey.

[0278] According to one embodiment, the electronic device can determine whether the sum of the scores for each skin characteristic item satisfies the entry condition corresponding to the additional survey question (S1910).

[0279] The entry condition may be a criterion for determining whether it is necessary to conduct additional survey questions regarding specific skin characteristic items. The entry condition may be a condition satisfied when the sum of the scores for each skin characteristic item falls within a specific range.

[0280] For example, the electronic device may apply additional survey questions when the sum of the scores for specific skin characteristic items falls within a threshold range. For instance, the electronic device may be configured to provide additional survey questions when the sensitivity score falls within a threshold range.

[0281] According to one embodiment, when an entry condition is satisfied and additional survey questions are composed of multiple-choice responses, the electronic device can determine whether a preset condition is satisfied based on whether the number of selected items exceeds a preset number (S1920).

[0282] According to one example, the electronic device may determine a specific skin condition if the number of selected items exceeds a standard number. For example, the electronic device may determine sensitive skin if three or more items related to irritation response are selected.

[0283] According to one embodiment, the electronic device can determine the second skin type as the final skin type for the skin characteristic item when the preset conditions are satisfied (S1930).

[0284] The second skin type may be a skin condition determined by reflecting the results of responses to additional survey questions. The second skin type may be a type designed to provide more accurate results by correcting the previously calculated skin type.

[0285] In one example, the electronic device can change the existing skin type based on additional survey results. For instance, even if the initial judgment result is resistant (R), the electronic device can change it to sensitive (S) if the additional survey results satisfy the sensitivity criteria.

[0286] According to one embodiment, when a second skin type is determined, the electronic device may be controlled not to provide and evaluate subsequent additional survey questions scheduled for skin characteristic items (S1940).

[0287] For example, the electronic device can be configured not to provide additional questionnaires regarding a specific skin characteristic item once a final determination has been made. For instance, if the sensitivity item is confirmed, the electronic device can skip additional questions regarding the same item and proceed to the next step.

[0288] FIG. 20 is a flowchart illustrating an additional survey-based skin type determination method according to one or more embodiments.

[0289] Referring to FIG. 20, an electronic device according to one embodiment can control the skin type to be corrected or maintained by evaluating the response results to additional survey questions in stages.

[0290] According to one embodiment, the electronic device can determine whether the first condition is satisfied based on whether the number of selected items in the response result to the first item among additional survey items exceeds a preset first number (S2010).

[0291] The first condition may be a judgment criterion set in correspondence with the first item among the additional survey items. The first condition may be a condition for determining that the corresponding skin characteristic exists if the number of items selected by the user among the multiple selection items presented in the first item exceeds a pre-set first number.

[0292] According to one example, the electronic device may determine that the sensitivity condition is satisfied when multiple stimulation-related items are selected in the first item. For example, the electronic device may determine that the first condition is satisfied when the user simultaneously selects redness, stinging, and itching items.

[0293] According to one embodiment, if the first condition is satisfied, the electronic device can determine the skin type as a second skin type corresponding to the first condition (S2020).

[0294] According to one example, the electronic device can change the skin type to sensitive (S) if the first condition is a sensitive condition. For example, the electronic device can determine the skin type to sensitive (S) even if the existing skin type is resistant (R) if the first condition is satisfied.

[0295] According to one embodiment, if the first condition is not satisfied and the sum of the scores for each skin characteristic item satisfies the entry condition corresponding to the second item, the electronic device may determine whether the second condition is satisfied based on whether the number of selected items in the response result to the second item among additional survey items exceeds a preset second number (S2030).

[0296] The second condition may be a judgment criterion established in correspondence with the second item among the additional survey items. The second condition applies when the first condition is not satisfied and may be a condition for determining the presence of a specific skin characteristic based on whether the number of items selected in the second item exceeds a pre-set second number.

[0297] The second item may be a set of questions designed to evaluate skin characteristic items different from the first item. The second item may be configured to correspond to a single skin characteristic, and each selection item may be an item representing a symptom or condition related to that skin characteristic.

[0298] For example, the electronic device may determine that the corresponding skin characteristic condition is satisfied if a certain number or more of items related to specific symptoms are selected in the second item. For instance, the electronic device may determine that the dry skin condition is satisfied if items such as dryness, flaking, and tightness are selected in a standard number or more.

[0299] According to one embodiment, if the second condition is satisfied, the electronic device can determine the skin type as a third skin type corresponding to the second condition (S2040).

[0300] The third skin type may be a skin type further corrected by reflecting the second condition. The third skin type may be the final skin condition determined by reflecting the results of the multi-stage survey.

[0301] According to one example, the electronic device can determine the skin type as dry (D) if the second condition is a dry condition. For example, even if the existing skin type is combination (C), the electronic device can change it to dry (D) if the second condition is satisfied.

[0302] According to one embodiment, the electronic device can maintain the first skin type when the second condition is not satisfied (S2050).

[0303] For example, the electronic device may not change the existing skin type if the additional survey results do not meet any conditions. For instance, the electronic device may maintain the initial skin type if the number of selected items falls below a threshold value.

[0304] For example, the electronic device can perform an additional survey when User A's skin type is determined to be complex (C) in the initial analysis results. The electronic device can provide a survey corresponding to sensitivity as the first item.

[0305] For example, the electronic device may determine that the first condition is satisfied when User A selects three items from redness, stinging, and itching, and the first number is set to 2. As the first condition is satisfied, the electronic device may change User A's skin type to a skin type that includes sensitivity (S).

[0306] For example, the electronic device can perform a second item in the case of User B, for whom the first condition is not satisfied. The electronic device can provide a questionnaire corresponding to dryness as the second item.

[0307] For example, the electronic device may determine that the second condition is satisfied when User B selects three items from dryness, peeling, and tightness, and the second number is set to 2. As the second condition is satisfied, the electronic device may change User B's skin type to dry (D).

[0308] For example, the electronic device may determine that neither the first condition nor the second condition is satisfied if, in the case of User C, the number of selected items in both the first and second questions does not exceed the respective standard number.

[0309] For example, if User C selects only one item related to sensitivity and one item related to dryness, the electronic device can determine that the number of each criterion is not exceeded. The electronic device can maintain User C's skin type as the existing skin type.

[0310] FIG. 21 is a block diagram illustrating the configuration of an electronic device according to one or more embodiments.

[0311] Referring to FIG. 21, the electronic device (100) includes a memory (2110) and a processor (2120) connected to the memory (2110). However, it is not limited thereto, and the electronic device (100) may be implemented with some components excluded or with other components included.

[0312] The memory (2110) can store at least one instruction, data, program, etc. required for the operation of the electronic device (100). Depending on the purpose of data storage, the memory (2110) may be implemented in the form of a memory embedded in the electronic device (100) or in the form of a memory that is detachable from the electronic device (100). For example, data for operating the electronic device (100) may be stored in a memory embedded in the electronic device (100), and data for the expansion function of the electronic device (100) may be stored in a memory that is detachable from the electronic device (100).

[0313] The processor (2120) controls the overall operation of the electronic device (100). Specifically, the processor (2120) is connected to each component of the electronic device (100) to control the overall operation of the electronic device (100). The processor (2120) may be configured to execute a program stored in memory (2110). When the program is executed, a skin type determination method according to the embodiments may be implemented.

[0314] According to one embodiment, when a program stored in memory (2110) is executed, the electronic device (100) receives responses to a plurality of survey questions from a user, classifies the plurality of survey questions into a plurality of skin characteristic items, calculates a score corresponding to the response for each skin characteristic item to calculate a sum of scores for each skin characteristic item, determines a first skin type based on the sum of scores for each skin characteristic item, compares the sum of scores for each skin characteristic item with preset threshold scores, and based on the comparison result, for at least one skin characteristic item among the plurality of skin characteristic items that exceeds the preset threshold scores, provides additional survey questions to the user and receives responses to the additional survey questions, and if the response result to the additional survey questions satisfies a preset condition, changes the first skin type to a second skin type corresponding to the preset condition, and generates a skin type code for the user by combining the skin types determined for each of the plurality of skin characteristic items.

[0315] The electronic device according to the present invention can perform personalized product recommendations by generating a skin type code that reflects the user's skin characteristics and combining product ingredient information and user purchase history data based on the said skin type code. The electronic device can provide products that are suitable for the user's skin condition, rather than simple general recommendations, and can effectively reduce the selection of unsuitable products that may occur due to a mismatch between skin characteristics and the product.

[0316] The electronic device according to the present invention can compensate for the limitations of initial score-based judgment by performing judgments on skin characteristic items stepwise through additional surveys and a conditional hierarchical survey structure, and by dynamically correcting the skin type according to the survey response results. The electronic device can more accurately determine a skin type suitable for user characteristics through the conditional hierarchical survey structure.

[0317] It is obvious that each step or operation of the method according to the embodiments of the present disclosure may be performed by a computer comprising one or more processors in accordance with the execution of a computer program stored in a computer-readable recording medium.

[0318] The computer-executable instructions stored on the aforementioned recording medium can be implemented through a computer program programmed to perform each corresponding step, and such a computer program can be stored on a computer-readable recording medium and executed by a processor. The computer-readable recording medium may be a non-transitory readable medium. In this case, a non-transitory readable medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short moment, such as a register, cache, or memory. Specifically, programs for performing the various methods described above may be provided by being stored on a non-transitory readable medium, such as semiconductor memory devices including erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), and flash memory devices; magnetic disks such as internal hard disks and removable disks; optical-magnetic disks; and non-volatile memory including CD-ROMs and DVD-ROMs.

[0319] Methods according to the various examples disclosed in this document may be provided by being included in a computer program product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or online through an application store (e.g., Play Store™). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created on a storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0320] As explained above, a person skilled in the art to which this disclosure pertains will understand that this disclosure may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of this disclosure is defined by the claims set forth below rather than by the detailed description, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts should be interpreted as being included within the scope of this disclosure.

[0321] The features and advantages described herein are not all included, and in particular, many additional features and advantages will become apparent to those skilled in the art by considering the drawings, the specification, and the claims. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and instructional purposes and may not be chosen to describe or limit the subject matter of this disclosure.

[0322] The foregoing description of the embodiments of the present disclosure is provided for illustrative purposes only. It is not intended to limit the present disclosure to the exact form disclosed or to make it incomplete. Those skilled in the art will understand that many modifications and variations are possible in light of the foregoing disclosure.

[0323] Therefore, the scope of the present disclosure is not limited by the detailed description but by any of the claims of the application based thereon. Accordingly, the disclosure of embodiments of the present disclosure is illustrative and does not limit the scope of the present disclosure as set forth in the following claims.

Claims

Claim 1 A method for determining skin type performed by an electronic device, comprising: receiving responses to a plurality of survey questions from a user; classifying the plurality of survey questions into a plurality of skin characteristic items and calculating a score corresponding to the response for each skin characteristic item to calculate a sum of scores for each skin characteristic item; determining a first skin type for each of the plurality of skin characteristic items based on the sum of scores for each skin characteristic item; and comparing the sum of scores for each skin characteristic item with a pre-set threshold score for each item. A step of determining to provide additional survey questions for a specific skin characteristic item that exceeds the threshold score among the plurality of skin characteristic items based on the above comparison results; when the provision of additional survey questions is determined, a first question is provided for the specific skin characteristic item, and if the number of items selected in the response to the first question exceeds a preset first number, the skin type for the specific skin characteristic item is changed from the first skin type to the second skin type, and if it is less than or equal to the first number, a second question is provided for the specific skin characteristic item, and if the number of items selected in the response to the second question exceeds a preset second number, the skin type for the specific skin characteristic item is changed from the first skin type to the third skin type, and if it is less than or equal to the second number, the skin type for the specific skin characteristic item is maintained as the first skin type; a step of generating a skin type code by combining characters corresponding to the finally determined skin type for each of the plurality of skin characteristic items according to a preset order; and based on the purchase history data of users in which all characters included in the skin type code are identical A step of generating a first recommendation result; a step of classifying users' purchase history data by individual type codes corresponding to each character included in the skin type code, and generating a second recommendation result based on the classified purchase history data;A method for determining skin type, comprising the step of providing the first recommendation result and the second recommendation result to the user; wherein the first recommendation result is a recommendation based on overall skin type and the second recommendation result is a customized recommendation based on skin characteristics. Claim 2 A method for determining skin type according to claim 1, further comprising: a step of identifying a plurality of products from a database that include tags corresponding to skin characteristic items included in the skin type code; a step of filtering the plurality of products based on the skin type code and ingredient information of the plurality of products; and a step of providing the filtered products to the user. Claim 3 A method for determining skin type according to claim 2, wherein the step of filtering the plurality of products based on the skin type code and the ingredient information of the plurality of products comprises: a step of setting included ingredients and excluded ingredients based on a skin characteristic item corresponding to the skin type code; and a step of excluding products containing the excluded ingredients among the plurality of products identified from the database, and selecting and filtering products containing the included ingredients. Claim 4 In paragraph 2, the step of providing the filtered product to the user comprises: a step of calculating a purchase rate for the filtered product based on purchase history data of users corresponding to a skin type code identical to the skin type code; a step of calculating a popularity for the filtered product based on purchase history data of all users; a step of normalizing each of the purchase rate and the popularity, and applying a first weight and a second weight to each of the normalized purchase rate and normalized popularity to calculate a weighted sum score; and a step of sorting and providing the filtered product to the user based on the weighted sum score; wherein the first weight is a value greater than the second weight. Claim 5 delete Claim 6 A method for determining skin type according to paragraph 2, wherein the step of providing the filtered product to the user comprises: generating a diagnostic text describing the skin condition of the user based on a skin characteristic item included in the skin type code; and providing the filtered product and the diagnostic text to the user. Claim 7 A method for determining skin type according to claim 2, wherein the step of providing the filtered products to the user comprises: a step of inputting the skin type code and the ingredient information of the filtered products into an artificial intelligence model to obtain a recommendation score for each of the filtered products; and a step of sorting the filtered products based on the recommendation score and providing the sorted products to the user; wherein the artificial intelligence model is trained using a training dataset that includes the user's skin type code and product ingredient information as input data and the user's purchase history data as correct answer data, and is a model trained to receive the skin type code and product ingredient information as input and output a recommendation score indicating the purchase suitability of the products. Claim 8 A method for determining skin type according to claim 7, wherein the learning dataset is generated by obtaining the number of purchases per product from purchase history data of users having the same skin type code as the skin type code, calculating the purchase rate per product based on the number of purchases per product and the total number of purchases for each product, obtaining the popularity for each product, generating a standard recommendation score per product based on the purchase rate and the popularity, and mapping the skin type code, the ingredient information of the product, and the standard recommendation score per product to each other; and the artificial intelligence model is trained to predict the standard recommendation score per product by receiving the skin type code and the ingredient information of the product as input. Claim 9 A skin type determination method according to paragraph 2, wherein the database is a database in which one or more tag information corresponding to each of a plurality of skin characteristic items included in the skin type code and a plurality of product information corresponding to each of the tag information are mapped and stored. Claim 10 delete Claim 11 delete Claim 12 delete Claim 13 An electronic device comprising a processor and a memory connected to the processor, wherein the memory is configured to store a program and the processor is configured to execute the program, and when the program is executed, any one of the methods of claims 1 to 4 and 6 to 9 is implemented.

Images