Method and apparatus for recommending user-customized scent

Abstract

Disclosed are a method and an apparatus for recommending a user-customized scent. The method for recommending a user-customized scent, according to an embodiment of the present invention, may comprise the steps of: collecting, from a first user, first use environment information of a first scent diffusion device and first scent information corresponding to a first scent diffused by the first scent diffusion device; performing training on an artificial intelligence learning model on the basis of the first use environment information and the first scent information; collecting second use environment information of a second scent diffusion device from a second user; on the basis of the artificial intelligence learning model, generating second scent information corresponding to a recommended scent associated with the second use environment information; and providing the second scent information to the second user.

Description

Method and device for recommending customized fragrances

[0001] The present invention relates to a method and device for recommending a user-customized fragrance.

[0002] The content described below merely provides background information related to one embodiment of the present invention and does not constitute prior art.

[0003] The environment can be viewed as a natural condition that can directly or indirectly influence living organisms. This environment can include gaseous particles distributed in the air. Humans inhale these gaseous particles through their noses, chemically stimulating olfactory receptors in the nose, allowing them to be recognized as odors through the olfactory nerves and identified. This olfactory nerve stimulation can elicit emotional responses and a sense of mental and emotional stability. It can also promote deep sleep and reduce physical fatigue.

[0004] Commonly referred to as "fragrance," fragrance refers to the pleasant odors emanating from flowers, perfumes, and other sources. The ability of certain scents to influence human psychology and emotions has long been known. Consequently, people use scent diffusers for various purposes, including mood-altering, psychological stability, and ventilation.

[0005] As an example of a fragrance device, perfume generally refers to a liquid or solid fragrance substance, which can be stored and used in a specific container equipped with a device capable of spraying the fragrance substance.

[0006] In particular, recently, control technologies that can control the operation of such emitting devices through digitalized control devices such as mobile terminals have emerged, and the emitting devices are generally equipped with fragrance capsules, etc., and emit a predetermined fragrance substance into the air according to a certain period of time or cycle.

[0007] However, the fragrance capsules provided in the fragrance device can have a wide variety of combinations depending on the number of cases, and due to the large number of variables, it is difficult for users to appropriately select them or directly control the fragrance intensity.

[0008] Furthermore, since the efficacy and purpose can both vary depending on the combination of fragrances, this must be finely adjusted to obtain the desired efficacy. However, there is a problem that it is impossible for general users to select these fragrance combinations or the intensity of each fragrance for their own purposes or desired efficacy without specialized knowledge, and it takes a lot of time.

[0009] The purpose of the present invention is to provide a method and device capable of recommending a user-customized fragrance by taking into account the user's individual preferences and the usage environment, such as the location and place of use.

[0010] In addition, the present invention aims to provide a method and device for providing a TTS (Text To Scent) function that recommends a scent based on text input by a user.

[0011] According to one embodiment of the present invention for achieving the above object, a user-customized fragrance recommendation method may include a step of collecting first usage environment information of a first fragrance diffusion device from a first user and first fragrance information corresponding to a first fragrance diffusion from the first fragrance diffusion device, a step of performing learning on an artificial intelligence learning model based on the first usage environment information and the first fragrance information, a step of collecting second usage environment information of a second fragrance diffusion device from a second user, a step of generating second fragrance information corresponding to a recommended fragrance associated with the second usage environment information based on the artificial intelligence learning model, and a step of providing the second fragrance information to the second user.

[0012] The first usage environment information may include at least one of the age of the first user, the gender of the first user, preference survey information of the first user, location information of the first induction device, purpose of the space where the first induction device is placed, usage time zone, and weather.

[0013] The first scent information may include at least one of information on the relative intensity of the emission of fragrance substances from each of the first to nth scent capsules mounted on the first scenting device to emit the first scent, the duration of emission of the first scent, and the intensity of emission of the first scent.

[0014] The second usage environment information may include information on fragrance capsules mounted on the second fragrance emitting device, and the step of generating the second fragrance information may include a step of selecting a first recommended fragrance that can be emanated only by a combination of fragrance capsules mounted on the second fragrance emitting device among recommended fragrance candidates associated with the second usage environment information.

[0015] The second usage environment information may further include information on fragrance capsules held by the second user, and the step of generating the second fragrance information may include a step of selecting a second recommended fragrance that can be diffused by combining fragrance capsules mounted on the second fragrance diffusion device and fragrance capsules held by the second user among recommended fragrance candidates associated with the second usage environment information, and the step of providing the second fragrance information may include a step of providing guidance to the second user indicating a replacement target among the fragrance capsules mounted on the second fragrance diffusion device for diffusion of the second recommended fragrance.

[0016] The step of generating the second fragrance information may include a step of selecting a third recommended fragrance that can be diffused by combining fragrance capsules that the second user does not have among the recommended fragrance candidates associated with the second usage environment information, and the step of providing the second fragrance information may include a step of providing the second user with purchase guidance for fragrance capsules that the second user does not have in order to diffuse the third recommended fragrance.

[0017] The step of collecting the second usage environment information may include a step of collecting at least one user keyword indicating a scent or mood from the second user, and the step of generating the second scent information may include a step of selecting a fourth recommended scent generated based on the user keyword.

[0018] The step of selecting the fourth recommended fragrance may include the step of determining a fragrance style parameter set including at least one fragrance style parameter based on the user keyword and the step of setting a default value for each of the fragrance style parameters included in the fragrance style parameter set.

[0019] The step of providing the second fragrance information may include the step of providing a user interface for adjusting the values ​​of each fragrance style parameter defining the fourth recommended fragrance, and the step of providing the second fragrance information by converting the fragrance style parameter set and the fragrance style parameter values ​​into information on the type of fragrance capsule and the relative intensity of the fragrant substance emitted from each fragrance capsule.

[0020] In addition, a user-customized fragrance recommendation device according to one embodiment of the present invention includes a memory having at least one program recorded thereon and a processor for executing the program, wherein the program may include instructions for performing a step of collecting first usage environment information of a first fragrance diffusion device from a first user and first fragrance information corresponding to a first fragrance diffusion from the first fragrance diffusion device, a step of performing learning on an artificial intelligence learning model based on the first usage environment information and the first fragrance information, a step of collecting second usage environment information of a second fragrance diffusion device from a second user, a step of generating second fragrance information corresponding to a recommended fragrance associated with the second usage environment information based on the artificial intelligence learning model, and a step of providing the second fragrance information to the second user.

[0021] According to the present invention, a method and device can be provided that can recommend a user-customized fragrance by taking into account the user's individual preferences, usage environment, such as the location and place of use.

[0022] In addition, according to the present invention, a method and device for providing a TTS (Text To Scent) function that recommends a scent based on text input by a user can be provided.

[0023] FIG. 1 is a diagram showing entities for user-customized fragrance recommendation according to one embodiment of the present invention.

[0024] FIG. 2 is a flowchart illustrating a user-customized fragrance recommendation method according to one embodiment of the present invention.

[0025] FIG. 3 is a flowchart illustrating a user-customized fragrance recommendation method according to another embodiment of the present invention.

[0026] FIG. 4 is a flowchart illustrating a user-tailored fragrance recommendation method according to one embodiment of the present invention for user keyword-based fragrance recommendation.

[0027] FIG. 5 is a flowchart illustrating a user-tailored fragrance recommendation method according to another embodiment of the present invention for user keyword-based fragrance recommendation.

[0028] FIG. 6 is a flowchart illustrating a user-tailored fragrance recommendation method according to another embodiment of the present invention for user keyword-based fragrance recommendation.

[0029] FIG. 7 is a diagram illustrating a computer system according to one embodiment of the present invention.

[0030] The present invention will be described in detail with reference to the attached drawings. Herein, repetitive descriptions, well-known functions that may unnecessarily obscure the gist of the present invention, and detailed descriptions of configurations are omitted. The embodiments of the present invention are provided to more fully explain the present invention to those of ordinary skill in the art. Accordingly, the shapes and sizes of elements in the drawings may be exaggerated for clarity.

[0031] Although "first" or "second" are used to describe various components, these components are not limited by such terms. Such terms may only be used to distinguish one component from another. Accordingly, a first component referred to below may also be a second component within the technical scope of the present invention.

[0032] Throughout the specification, whenever a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise stated.

[0033] Hereinafter, a preferred embodiment according to the present invention will be described in detail with reference to the attached drawings.

[0034] FIG. 1 is a diagram showing entities for user-customized fragrance recommendation according to one embodiment of the present invention.

[0035] Referring to FIG. 1, user-customized fragrance recommendation according to one embodiment of the present invention can be performed by a fragrance recommendation device (110), a fragrance diffusion device (120), a user terminal (130), and a fragrance database (140).

[0036] The fragrance recommendation device (110) may be a device that acquires various data from a fragrance emitting device (120) and a user terminal (130), stores them in a fragrance database (140), performs learning on an artificial intelligence learning model for fragrance recommendation, and provides fragrance recommendation information to the user terminal (130).

[0037] The fragrance recommendation device (110) may be a server that performs the function of a digital fragrance platform to control the fragrance dissipation device (120) using a user terminal (130) and to enable use of various fragrance-related information.

[0038] The fragrance emitting device (120) may be a device that is placed in an indoor space and capable of emitting fragrance. The fragrance emitting device (120) may have a plurality of fragrance capsules that can be detached to emit fragrance substances of different fragrances, and may adjust the relative intensity of the fragrance substances emitted by each of the fragrance capsules to create a predetermined fragrance.

[0039] The user terminal (130) may be a terminal corresponding to a user who wants to connect to the fragrance recommendation device (110) or control the fragrance emitting device (120).

[0040] A user terminal (130) refers to a communication terminal capable of utilizing a terminal application in a wired or wireless communication environment. For example, the user terminal (130) may be a portable terminal such as a smart phone or a personal computer. However, the spirit of the present invention is not limited thereto, and any terminal equipped with a terminal application may be utilized without limitation.

[0041] Here, the terminal application may be a dedicated application for controlling the launching device (120) or a web browser capable of controlling the launching device on a web page.

[0042] The fragrance database (140) may be a device that stores various fragrance information, user usage information, and operation information of a fragrance device, and provides this to a fragrance recommendation device (110).

[0043] The fragrance recommendation device (110), the fragrance emitting device (120), the user terminal (130), and the fragrance database (140) can be interconnected through a communication network.

[0044] The communication network refers to a connection path for transmitting and receiving data between the fragrance recommendation device (110), the fragrance diffusion device (120), the user terminal (130), and the fragrance database (140). For example, the communication network may include wired networks such as LANs (Local Area Networks), WANs (Wide Area Networks), MANs (Metropolitan Area Networks), and ISDNs (Integrated Service Digital Networks), or wireless networks such as wireless LANs, CDMA, Bluetooth, and satellite communication, but the scope of the communication network applicable to the present invention is not limited thereto.

[0045] FIG. 2 is a flowchart illustrating a user-customized fragrance recommendation method according to one embodiment of the present invention.

[0046] Referring to FIG. 2, a user-customized fragrance recommendation method according to one embodiment of the present invention first allows a fragrance recommendation device (110) to collect, from a first user, usage environment information of a first fragrance emitting device and fragrance information corresponding to a first fragrance emitting from the first fragrance emitting device (S210).

[0047] Here, the usage environment information may include at least one of the user's age, the user's gender, the user's preference survey information, the location information of the induction device, the purpose of the space where the induction device is placed, the usage time zone, and the weather.

[0048] As an optional embodiment, the usage environment information may include information about fragrance capsules mounted on the scenting device.

[0049] The scent information may include at least one of information on the relative intensity of the scent material emitted by each of the first to nth scent capsules mounted on the scenting device to emit a predetermined scent, scent emission duration, and scent emission intensity.

[0050] Next, the fragrance recommendation device (110) can perform learning on an artificial intelligence learning model based on the first user's usage environment information and the first fragrance information (S220).

[0051] Here, the fragrance recommendation device (110) can perform learning on the correlation between the usage environment and the fragrance.

[0052] Next, the fragrance recommendation device (110) can collect usage environment information of the second fragrance device from the second user (S230).

[0053] Next, the fragrance recommendation device (110) can generate second fragrance information corresponding to the recommended fragrance related to the usage environment information of the second user based on the artificial intelligence learning model (S240).

[0054] Next, the fragrance recommendation device (110) can provide second fragrance information to a second user (S250).

[0055] FIG. 3 is a flowchart illustrating a user-customized fragrance recommendation method according to another embodiment of the present invention.

[0056] Referring to FIG. 3, in a method for recommending a user-tailored fragrance according to another embodiment of the present invention, the fragrance recommendation device (110) may, in the step of generating second fragrance information, select a first recommended fragrance based on a fragrance capsule mounted on the fragrance diffusion device (S241), select a second recommended fragrance based on a fragrance capsule possessed by the second user (S242), or select a third recommended fragrance based on the entire fragrance capsule including a fragrance capsule not possessed by the second user (S243).

[0057] Specifically, first, the fragrance recommendation device (110) can select a first recommended fragrance that can be diffused only by a combination of fragrance capsules mounted on the fragrance diffusion device among the recommended fragrance candidates related to the usage environment information of the second user (S241).

[0058] In this case, there is an advantage in that the second user can easily change the scent by simply selecting the recommended scent.

[0059] Alternatively, the fragrance recommendation device (110) may further utilize information about fragrance capsules held by the second user included in the usage environment information of the second user, and may select a second recommended fragrance that can be diffused by combining the fragrance capsules mounted on the second fragrance diffusion device and the fragrance capsules held by the second user among the recommended fragrance candidates determined to be related using other usage environment information of the second user (S242).

[0060] In this case, in addition to selecting the recommended scent, the second user needs to replace the capsule mounted on the second scenting device with a suitable second recommended scent.

[0061] To this end, the fragrance recommendation device (110) can provide guidance to the second user to indicate a replacement target among the fragrance capsules mounted on the second fragrance emitting device for emitting the second recommended fragrance.

[0062] Here, the fragrance recommendation device (110) can provide replacement instructions for the remaining fragrance capsules, excluding capsules that overlap with the fragrance capsules used for emitting the second recommended fragrance among the fragrance capsules mounted on the second fragrance emitting device.

[0063] For example, in a case where the first to fourth fragrance capsules are mounted on the second fragrance emitting device and the first to third and fifth fragrance capsules are required to emit the second recommended fragrance, when the second user selects the second recommended fragrance, an instruction may be provided to replace the fourth fragrance capsule, excluding the first to third fragrance capsules used to emit the second recommended fragrance, with the fifth fragrance capsule among the fragrance capsules mounted on the existing second fragrance emitting device.

[0064] In addition, by providing a visual representation of the location of the fragrance capsule mounted on the second fragrance device in the user terminal, the replacement target of the fragrance capsule can be visually distinguished, thereby guiding the second user to easily replace the fragrance capsule.

[0065] Alternatively, the fragrance recommendation device (110) can select a third recommended fragrance that can be diffused by combining fragrance capsules that the second user does not have among the recommended fragrance candidates associated with the second usage environment information (S243).

[0066] In this case, the second user would need to purchase a fragrance capsule that is needed to emit the third recommended fragrance, but which he or she does not have.

[0067] To this end, the fragrance recommendation device (110) can provide the second user with guidance on purchasing fragrance capsules that the second user does not have in order to diffuse the third recommended fragrance.

[0068] The fragrance recommendation device (110) can recommend fragrances by differentiating the recommendation priorities in the order of first recommended fragrance, second recommended fragrance, and third recommended fragrance.

[0069] Each of the first recommended scent, the second recommended scent, and the third recommended scent may be at least one recommended scent.

[0070] In addition, the fragrance recommendation device (110) can recommend fragrances by changing the recommendation priority as the number of fragrance capsules that must be purchased for fragrance diffusion decreases, from a case where the second user only needs to purchase one more fragrance capsule that the second user does not have to a case where all fragrance capsules must be purchased, when recommending the third recommended fragrance.

[0071] As an optional embodiment, the fragrance recommendation device (110) may receive a user keyword from a second user and recommend a fragrance based on the user keyword.

[0072] FIG. 4 is a flowchart illustrating a user-tailored fragrance recommendation method according to one embodiment of the present invention for user keyword-based fragrance recommendation.

[0073] Referring to FIG. 4, in order to recommend a fragrance based on a user keyword of the present invention, first, in the step of collecting second usage environment information, the fragrance recommendation device (110) can collect at least one user keyword indicating a fragrance or mood from a second user (S231).

[0074] Here, user keywords can be words that describe the scent or keywords that include the feeling of the scent.

[0075] For example, a user keyword can be multiple keywords, such as "vanilla coffee scent, sweet enough to make your teeth ache," and a single keyword can be composed of multiple words, including spaces.

[0076] And, the fragrance recommendation device (110) can select a fourth recommended fragrance generated based on the user keyword in the step of generating the second fragrance information (S244).

[0077] FIG. 5 is a flowchart illustrating a user-tailored fragrance recommendation method according to another embodiment of the present invention for user keyword-based fragrance recommendation.

[0078] In particular, Fig. 5 is a diagram showing a process of selecting a fourth recommended fragrance based on user keywords.

[0079] Referring to FIG. 5, the fragrance recommendation device (110) can determine a fragrance style parameter set including at least one fragrance style parameter based on a user keyword (S510).

[0080] The fragrance recommendation device (110) performs natural language processing on user keywords to convert them into a set of tokens, which are the smallest units having meaning, and can determine multiple sets of fragrance style parameters based on the correlation and weight between each token.

[0081] For example, if the user keyword is "vanilla coffee scent, sweet enough to make your teeth ache", the fragrance style parameters may be determined as main ingredient parameters with coffee and vanilla as main ingredients to form the overall background, and caramel, white chocolate, cream note, and tonka bean as auxiliary ingredient parameters so that sweet notes can be emphasized, and a fragrance style parameter set including six fragrance style parameters may be determined in this way.

[0082] Through this, the scent recommendation device can perform TTS (Text To Scent) which receives text input from the user and outputs a scent.

[0083] Next, the fragrance recommendation device can set a default value for each of the fragrance style parameters included in the fragrance style parameter set (S520).

[0084] For example, if the user keyword is "vanilla coffee aroma, tooth-numbingly sweet," coffee and vanilla are the main components and core of the aroma, so their default value would be set to 0.5; caramel and white chocolate emphasize the tooth-numbingly sweet feeling, so their default value would be set to a relatively high 0.7; and cream notes and tonka bean are secondary components that add softness and depth, so their default value could be set to 0.2.

[0085] FIG. 6 is a flowchart illustrating a user-tailored fragrance recommendation method according to another embodiment of the present invention for user keyword-based fragrance recommendation.

[0086] Referring to FIG. 6, the fragrance recommendation device (110) can provide a user interface for adjusting the values ​​of each fragrance style parameter defining the fourth recommended fragrance (S610).

[0087] For example, if the user keyword is "vanilla coffee scent, tooth-numbingly sweet," default values ​​set for the scent style parameters of coffee, vanilla, caramel, white chocolate, cream note, and tonka bean, as well as a user interface for adjusting these values, may be provided.

[0088] For example, in the user terminal of the second user, a user interface is provided in which the values ​​of the fragrance style parameters constituting the fourth recommended fragrance corresponding to the user keyword can be adjusted from 0 to 1.0 in a slide format, and the user can adjust the values ​​by moving the slide for each of the fragrance style parameters.

[0089] The fragrance recommendation device (110) can provide second fragrance information by converting a fragrance style parameter set and a fragrance style parameter value into information on the type of fragrance capsule and the relative intensity of the fragrant substance emitted by each fragrance capsule (S620).

[0090] In a similar manner to how red, green, and blue inks are used to output an image, a result value converted into information on the type of fragrance capsule and the relative intensity of fragrant substance emission of each fragrance capsule for outputting a scented fragrance using a user keyword is generated and transmitted to a second scenting device, thereby allowing the fourth recommended fragrance to be reproduced.

[0091] FIG. 7 is a diagram illustrating a computer system according to one embodiment of the present invention.

[0092] A user-customized fragrance recommendation device according to one embodiment of the present invention can be implemented in a computer system (1000) such as a computer-readable recording medium.

[0093] Referring to FIG. 7, a computer system (1000) may include one or more processors (1010), memory (1030), a user interface input device (1040), a user interface output device (1050), and storage (1060) that communicate with each other via a bus (1020). In addition, the computer system (1000) may further include a network interface (1070) connected to a network (1080). The processor (1010) may be a central processing unit or a semiconductor device that executes processing instructions stored in the memory (1030) or the storage (1060). The memory (1030) and the storage (1060) may be various forms of volatile or non-volatile storage media. For example, the memory may include ROM (1031) or RAM (1032).

[0094] The specific implementations described in the present invention are merely exemplary embodiments and do not limit the scope of the present invention in any way. For the sake of brevity, descriptions of conventional electronic components, control systems, software, and other functional aspects of the systems may be omitted. In addition, the lines connecting or connecting members between components depicted in the drawings are merely exemplary functional connections and / or physical or circuit connections, and may be replaced or represented as various additional functional connections, physical connections, or circuit connections in an actual device. In addition, unless specifically mentioned as “essential,” “important,” etc., a component may not be absolutely necessary for the application of the present invention.

[0095] Therefore, the idea of ​​the present invention should not be limited to the embodiments described above, and not only the scope of the patent claims described below but also all scopes equivalent to or equivalently modified from the scope of the patent claims are considered to fall within the scope of the idea of ​​the present invention.

[0096] According to the user-customized fragrance recommendation method and device according to embodiments of the present invention, a user-customized fragrance can be recommended by considering the user's individual preferences or the usage environment such as the location and place of use, and a TTS (Text To Scent) function that recommends a fragrance based on text input by the user can be provided, so that there is industrial applicability.

Claims

1. A step of collecting first usage environment information of a first induction device from a first user and first fragrance information corresponding to a first fragrance emitted from the first induction device; A step of performing learning for an artificial intelligence learning model based on the first usage environment information and the first fragrance information; A step of collecting second usage environment information of a second triggering device from a second user; A step of generating second fragrance information corresponding to a recommended fragrance associated with the second usage environment information based on the artificial intelligence learning model; and A step of providing the second fragrance information to the second user A method for recommending a customized fragrance, comprising:

2. In paragraph 1, The above first usage environment information is A method for recommending a user-tailored fragrance, comprising at least one of the age of the first user, the gender of the first user, preference survey information of the first user, location information of the first fragrance emitting device, purpose of the space in which the first fragrance emitting device is placed, time zone of use, and weather.

3. In paragraph 1, The above first scent information is A user-customized fragrance recommendation method, comprising at least one of information on the relative intensity of fragrance material discharged from each of the first to nth fragrance capsules mounted on the first fragrance emitting device to emit the first fragrance, the duration of emission of the first fragrance, and the intensity of emission of the first fragrance.

4. In paragraph 1, The above second usage environment information is Includes information about the fragrance capsules mounted on the second fragrance device, The step of generating the above second scent information is A step of selecting a first recommended fragrance that can be diffused only by a combination of fragrance capsules mounted on the second fragrance device among the recommended fragrance candidates related to the second usage environment information. A method for recommending a customized fragrance, comprising:

5. In paragraph 4, The above second usage environment information is Further including information about the fragrance capsules held by the second user; The step of generating the above second scent information is A step of selecting a second recommended scent that can be diffused by combining the scent capsules mounted on the second diffuser and the scent capsules possessed by the second user among the recommended scent candidates related to the second usage environment information. Including, The step of providing the above second fragrance information is A step of providing guidance to the second user to indicate a replacement target among the fragrance capsules mounted on the second fragrance dispersing device for dispersing the second recommended fragrance. A method for recommending a customized fragrance, comprising:

6. In paragraph 5, The step of generating the above second scent information is A step of selecting a third recommended scent that can be diffused by combining scent capsules that the second user does not have among the recommended scent candidates related to the second usage environment information. Including, The step of providing the above second fragrance information is A step of providing the second user with guidance on purchasing fragrance capsules that the second user does not have in order to diffuse the third recommended fragrance. A method for recommending a customized fragrance, comprising:

7. In paragraph 6, The step of collecting the above second usage environment information is A step of collecting at least one user keyword indicating a scent or mood from the second user. Including, The step of generating the above second scent information is Step of selecting the fourth recommended fragrance generated based on the above user keywords A method for recommending a customized fragrance, comprising:

8. In paragraph 7, The step of selecting the fourth recommended fragrance is determining a set of fragrance style parameters including at least one fragrance style parameter based on the user keyword; and A step of setting the default value of each of the fragrance style parameters included in the above fragrance style parameter set. A method for recommending a customized fragrance, comprising:

9. In paragraph 7, The step of providing the above second fragrance information is A step of providing a user interface for adjusting the value of each fragrance style parameter defining the fourth recommended fragrance; and A step of providing the second fragrance information by converting the above fragrance style parameter set and fragrance style parameter values ​​into information on the type of fragrance capsule and the relative intensity of the fragrance substance emitted from each fragrance capsule. A method for recommending a customized fragrance, comprising:

10. Memory in which at least one program is recorded; and Processor that runs the above program Including, The above program is, A step of collecting first usage environment information of a first induction device from a first user and first fragrance information corresponding to a first fragrance emitted from the first induction device; A step of performing learning for an artificial intelligence learning model based on the first usage environment information and the first fragrance information; A step of collecting second usage environment information of a second triggering device from a second user; A step of generating second fragrance information corresponding to a recommended fragrance associated with the second usage environment information based on the artificial intelligence learning model; and A step of providing the second fragrance information to the second user A user-customized fragrance recommendation device comprising commands for performing the following:

Images