KR20230029486A - Method and apparatus for providing a service for recommending costomized cosmetics to user based on the user's skin characteristics and cosmetic use information of other users belonging to an environment similar to the user - Google Patents

Abstract

According to various embodiments, a cosmetics recommendation server for recommending customized cosmetics in consideration of cosmetics use information of other users with similar activity areas and work environments. The server comprises: an information acquisition unit for acquiring a face image of a user and activity area information indicating an area for activity of the user from a user terminal; a skin properties determination unit for determining properties of skin of the user, based on the acquired face image; and a cosmetics determination unit for determining cosmetics based on the determined skin characteristics and the cosmetics use information of other users distinct from the user, wherein the cosmetics determination unit determines first users among the other users in the same activity area as the user, determines second users, among the determined first users, who have the same work type as the user and whose average monthly out-of-hours time difference is within a first hour, determines users whose work similarity, which indicates the similarity of the work environment between the user and the second users, is greater than or equal to a threshold among the determined second users as third users, and determines the cosmetics based on the cosmetics use information of the third users.

Description

A method and apparatus for providing a service for recommending customized cosmetics to a user based on the user's skin characteristics and other users' cosmetics use information belonging to a similar environment to the user ON THE USER'S SKIN CHARACTERISTICS AND COSMETIC USE INFORMATION OF OTHER USERS BELONGING TO AN ENVIRONMENT SIMILAR TO THE USER}

The present invention relates to a method and apparatus for recommending customized cosmetics to a user. The present invention relates to a method and apparatus for determining and recommending customized cosmetics to a user based on cosmetics use information of other users belonging thereto.

In recent years, as interest in beauty has increased, in particular, interest in facial skin care has increased. In addition, the consumption propensity of today's buyers is clearly different from that of the past. Unlike existing buyers who directly viewed and purchased products offline, today's buyers prefer to purchase a variety of products through a communication network using devices such as computers or smart phones. In this case, buyers of cosmetics can obtain cosmetics information by referring to cosmetics shopping malls, blogs, and product reviews.

However, since the skin is greatly affected by external factors as well as internal factors, the user's skin type and skin condition may be similar among people with similar living environments. Therefore, rather than simply relying on other buyers' reviews, considering the characteristics of other buyers with similar living environments (eg, working environments), the user's characteristics (eg, gender, age, skin type, type of cosmetics, cosmetics Since it is possible to provide cosmetics suitable for the purpose and effect of the product more quickly and appropriately, it is necessary to prepare a method for recommending cosmetics suitable for the user.

In addition, the skin surface is closely related to the external environment, and in particular, there may be many changes on the skin surface depending on the external temperature and humidity. Korea has distinct climate characteristics by region and four distinct seasons, so changes in the external environment are large. It may be necessary to recommend cosmetics to users by reflecting changes in the external environment during the period of use of cosmetics.

The purpose of the present invention to solve the above problems is not simply using the cosmetic use information of other users, but based on the cosmetic use information of other users who have similar activity areas and work environments to the user by utilizing big data, A method and apparatus for determining cosmetics to be recommended to a user are provided.

Another object of the present invention to solve the above problems is to identify the user's skin characteristics corresponding to the point of purchase using an artificial neural network for the user's face image, as well as skin characteristics that will occur while using cosmetics. Based on the change, a method and apparatus for determining cosmetics to be recommended to a user are provided.

According to various embodiments of the present disclosure, a cosmetics recommendation server that recommends customized cosmetics in consideration of cosmetics use information of other users with similar activity areas and work environments indicates the user's face image and the user's activity region from the user's user terminal. an information acquisition unit that obtains activity area information to be performed; a skin characteristic determination unit configured to determine skin characteristics of the user based on the obtained face image; Cosmetics are determined based on the determined skin characteristics and cosmetics use information of other users who are distinguished from the user, and among the other users, first users in the same activity area as the user are determined, and the determined first users are selected. determining second users whose work type is the same as that of the user and whose monthly average out-of-office time difference is less than a first hour; A cosmetics determining unit may be configured to determine users whose working environment similarity is equal to or greater than a threshold value as third users, and to determine the cosmetics based on cosmetics use information of the third users.

In various embodiments, a cosmetics recommendation server that recommends customized cosmetics in consideration of cosmetics use information of other users with similar activity areas and work environments may indicate the user's face image and the user's activity area from the user's user terminal an information acquisition unit that acquires activity area information; a skin characteristic determination unit configured to determine skin characteristics of the user based on the obtained face image; a cosmetics determining unit which determines cosmetics including necessary ingredients and necessary effects required for the user based on the determined skin characteristics; and a recommendation list creation unit generating a recommendation list of the determined cosmetics and transmitting the recommendation list to the user terminal. The skin characteristic determination unit determines at least one skin characteristic among moisture, oil, sebum, wrinkles, pore state, and dead skin condition of the user's skin based on the face image, and the cosmetics determination unit determines the determined user's skin characteristics. The cosmetic products may be determined based on characteristics and cosmetic use information of other users distinguished from the user.

In various embodiments, the skin characteristic determination unit divides the acquired face image of the user into a plurality of face regions, determines skin characteristics for each of the plurality of divided face regions, and determines the skin characteristics for each of the plurality of face regions. Based on the characteristics, the user's skin characteristics corresponding to the point of purchase may be determined.

In various embodiments, the skin characteristic determiner may determine the skin characteristic of a first facial region having an area equal to or larger than a first area among the plurality of facial regions, and determine the skin characteristic of the first facial region through the skin characteristic of the first facial region. The skin characteristics of the second facial region exhibiting similar skin characteristics may be determined, and the skin characteristics of the user may be determined based on the determined skin characteristics of the first facial region and the skin characteristics of the second facial region.

In various embodiments, the skin characteristic determination unit determines a time point after the predetermined period from the purchase time based on the user's skin characteristics corresponding to the purchase time and a degree of change in skin characteristics for a predetermined period from the purchase time. It is possible to determine the user's skin characteristics corresponding to the duration of use. The cosmetics determination unit may determine the cosmetics based on the user's skin characteristics corresponding to the duration of use.

In various embodiments, the cosmetics recommendation server further includes a skin characteristic gradient determining unit that determines a skin characteristic gradient of the user that varies according to a climate corresponding to an active region of the user, wherein the skin characteristic gradient determining unit Based on the climate information for each region obtained from the Korea Meteorological Administration server, the climate of the activity region corresponding to the purchase time and the climate of the activity region corresponding to the use duration time are compared, and based on the comparison, the user's determining a change in skin characteristics, and determining a temperature change rate and a humidity change rate between the temperature and humidity of the user's active area corresponding to the time of purchase and the temperature and humidity of the user's active area corresponding to the duration of use; Among the skin characteristics of the user, the gradients of moisture, sebum, pores, and elasticity are determined in proportion to the determined rates of change in temperature and rate of change in humidity, and in inverse proportion to the determined rates of change in temperature and humidity, wrinkles among the characteristics of the user's skin, Changes in pH, skin brightness, and skin sensitivity can be determined.

In various embodiments, the cosmetics determining unit determines a plurality of users in the same activity area as the user among the other users, has the same work type as the user among the determined plurality of users, and has a monthly month with the user. It is possible to determine first users whose difference in average out-of-office time is within a first time.

In various embodiments, the cosmetics determining unit determines, among the determined first users, users having a working similarity representing a similarity of a working environment between the user and the first users equal to or greater than a threshold value as second users, The cosmetics are determined based on cosmetics use information of the second users, and the cosmetics use information may include at least one of purchase reviews, use period, frequency of use, and return status.

,

상기 수학식에서, ID은 상기 증감 연관도고, I_o는 상기 제1 사용자들 각각의 외근 증감도고, I_a는 상기 사용자의 기준 외근 증감도고, 상기 사용자의 기준 외근 변화도와 상기 제1 사용자들 각각의 외근 변화도 사이의 유클리디언 거리를 이하 수학식을 통해 산출하고,

상기 수학식에서, DIS는 유클리디언 거리이고, G_o는 상기 제1 사용자들 각각의 외근 변화도고, G-_a는 상기 사용자의 기준 외근 변화도이고, 이하 수학식을 통해 상기 근무 유사도를 결정하고,

상기 수학식에서, S_w는 상기 근무 유사도이고, k는 증감 연관도(ID)와 유클리디언 거리(DIS) 사이의 가중치를 결정하는 가중치 계수로서 0과 1 사이의 정수이며, DISμ는 결정된 상기 제1 사용자들에 대하여 산출된 유클리디언 거리들의 평균값이고, DISσ는 결정된 상기 제1 사용자들에 대하여 산출된 유클리디언 거리들의 표준편차일 수 있다.In various embodiments, the cosmetics recommendation server further includes a work similarity determination unit, and the work similarity determination unit determines the change in outside work and the change in outside work for each of the first users based on the outside work time for the first users. Calculate the increase and decrease relationship between the user's reference external muscle change and the external muscle increase and decrease of each of the first users through the following equation,

,

In the above equation, ID is the degree of increase/decrease correlation, I _o is the degree of increase or decrease of external muscle of each of the first users, I _a is the degree of increase or decrease of external muscle of the user based on the user, and Calculate the Euclidean distance between the external gradients through the following equation,

In the above equation, DIS is the Euclidean distance, _Go is the degree of change in external force of each of the first users, G- _a is the degree of change in external force of the user, and the work similarity is determined through the following equation, ,

In the above equation, S _w is the work similarity, k is an integer between 0 and 1 as a weight coefficient for determining the weight between the incremental association (ID) and the Euclidean distance (DIS), and DISμ is the determined first It may be an average value of Euclidean distances calculated for 1 user, and DISσ may be a standard deviation of the determined Euclidean distances calculated for the first users.

The cosmetics determining unit, in the case of cosmetics having a ratio of positive reviews among the purchase reviews of the second users equal to or greater than a first ratio, increases a weight for cosmetics having a ratio equal to or greater than the first ratio, and the ratio of positive reviews is less than the first ratio. In the case of cosmetics, weights for cosmetics that are equal to or less than the first ratio may be lowered, and cosmetics suitable for the user may be determined based on the determined weights for cosmetics.

In various embodiments, the cosmetics recommendation server further includes a usage duration determining unit, and the usage duration determining unit obtains an average usage period of the user for used cosmetics input through the user terminal, wherein the use duration determination unit Cosmetics are of the same type as the cosmetics that the user intends to purchase, and if the user's average period of use is less than or equal to the critical period, a time point after 1/2 of the average period of use from the purchase time is determined as the usage duration point, , When the average usage period of the user exceeds the threshold period, a time point after 1/4 of the average usage period from the purchase time may be determined as the usage duration time point.

According to various embodiments disclosed in this document, since characteristics of people similar to the user's living environment are referred to, cosmetics suitable for the user can be recommended more quickly and accurately.

In addition, according to various embodiments, a skin condition after a certain period of time may be predicted from the time of purchase in consideration of changes in temperature and humidity according to regional climate and seasonal changes in each region.

In addition, according to various embodiments, cosmetics are recommended by predicting the skin condition after a certain period of time from the time of purchase, so the replacement time of cosmetics can be delayed.

In addition, according to various embodiments, the amount of unused cosmetics can be reduced by delaying the replacement time of cosmetics.

In addition to this, various effects identified directly or indirectly through this document may be provided.

1 is a diagram illustrating a user cosmetics recommendation system according to an embodiment.
FIG. 2 is a diagram showing components of a cosmetics recommendation server of FIG. 1 .
3 is a diagram for explaining that a user's skin characteristics vary according to regional climate characteristics.
FIG. 4 is a diagram explaining a process of determining cosmetics suitable for skin characteristics corresponding to a user's duration of use of cosmetics.
FIG. 5 is a diagram for explaining determining work similarity in order to consider cosmetics usage information of other users.
6 is a diagram showing the hardware configuration of the server according to FIG. 1;

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. Like reference numerals have been used for like elements throughout the description of each figure.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The terms and/or include any combination of a plurality of related recited items or any of a plurality of related recited items.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a user cosmetics recommendation system 10 according to an embodiment.

Referring to FIG. 1 , the user cosmetics recommendation system 10 may include a cosmetics recommendation server 100 and a user terminal 200 .

The cosmetics recommendation server 100 may store information necessary to operate a user cosmetics recommendation service. Information necessary to operate the user cosmetics recommendation service may include information on each of cosmetics that may be provided by an operator of the service for providing the service. In addition, the cosmetics recommendation server 100 may acquire and store cosmetics use information of other existing buyers. Cosmetics use information of other buyers may include purchase reviews, use period, frequency of use, whether or not to return the product, and the like.

The cosmetics recommendation server 100 obtains a user's face image through the user terminal 200, determines the user's skin characteristics through the obtained user's face image, and recommends cosmetics to the user based on the determined skin characteristics. can decide The cosmetics recommendation server 100 may generate a recommendation list of determined cosmetics and transmit the generated recommendation list to the user terminal 200 .

The cosmetics recommendation server 100 may perform supervised learning or unsupervised learning on a cosmetics recommendation model according to the determined skin characteristics, and the cosmetics recommendation model may be implemented as an artificial neural network (ANN). An artificial neural network is a predictive model implemented in software or hardware that mimics the computational power of a biological system using a large number of artificial neurons (or nodes). The determined skin characteristics may include the user's skin characteristics corresponding to the point of purchase and/or the user's skin characteristics corresponding to the duration of use.

The cosmetics recommendation model can be supervised and learned by using the 'face image' and 'cosmetics corresponding to the face image' by the cosmetics recommendation model learning unit. At this time, supervised learning means learning to find an output value according to a given input value by using data with input values and corresponding output values as learning data, and means learning that takes place in a state where the correct answer is known. The set of inputs and outputs given to supervised learning is called training data. That is, the aforementioned 'face image' and 'cosmetics corresponding to the facial image' are input values and output values, respectively, and may be used as training data for supervised learning of a cosmetic recommendation model.

For example, the cosmetics recommendation model learning unit generates an input value by converting a season into a unique first one-hot vector, and converts skin characteristics according to the season into a unique second one-hot vector. After generating the output value, the cosmetic recommendation model can be supervised using the generated input value and output value. Here, the first one-hot vector and the second one-hot vector may be vectors in which one of the component values constituting the vector is '1' and the other component values are '0'.

The user terminal 200 includes a communicable desktop computer, a laptop computer, a notebook, a smart phone, a tablet PC, a mobile phone, and a smart watch. (smart watch), smart glass, e-book reader, PMP (portable multimedia player), portable game console, navigation device, digital camera, DMB (digital multimedia broadcasting) player, digital voice It may be a digital audio recorder, a digital audio player, a digital video recorder, a digital video player, and a personal digital assistant (PDA).

Throughout the specification, products recommended through the user cosmetics recommendation service may be cosmetics, but are not limited thereto and may include all of a plurality of beauty-related products that are not normally classified as cosmetics.

FIG. 2 is a diagram showing components of the cosmetics recommendation server 100 of FIG. 1 .

Referring to FIG. 2, the cosmetics recommendation server 100 includes an information acquisition unit 101, a skin characteristic determination unit 102, a skin characteristic change degree determination unit 104, a usage duration determination unit 103, and a work similarity determination unit. It may include a unit 105, a cosmetics determination unit 106 and a recommendation list generation unit 107, and the like.

The information obtaining unit 101 may determine the user's characteristic information by using the user's face image obtained from the user terminal 200 . For example, the characteristic information may include age, gender, and skin characteristics of the user. Also, the user may photograph his or her face using the user terminal 200, and the user terminal 200 may transmit the photographed face image to the cosmetics recommendation server 100. The user terminal 200 may include a camera capable of photographing the user's face and a depth camera capable of measuring distances to each point on the user's face.

The information acquisition unit 101 may obtain at least one of the user's age, gender, skin type, skin condition, and allergy information as user information. In addition, the information acquisition unit 101 may obtain activity region information of the user as user information. The information obtaining unit 101 acquires the user's location information sensed using the Global Positioning System (GPS) built in the user terminal 200, and indicates the user's activity area based on the obtained user's location information. Activity area information can be obtained.

The information acquisition unit 101 may acquire information about existing cosmetics purchase history of an account matching user's personal information from an internal storage device or an external cosmetics sales server connected to the cosmetics recommendation server 100 . Alternatively, the information acquisition unit 101 may acquire existing cosmetics purchase information input from the user terminal 200 from the user terminal 200 .

The information acquisition unit 101 may acquire the user's interest list as well as the user's purchase history for purchased cosmetics.

The skin characteristic determining unit 102 may determine the user's skin characteristic corresponding to the point of purchase based on the acquired facial image of the user. Skin characteristics may include skin type and skin condition. For example, skin characteristics may include skin moisture, oil content, sebum, pH level, sensitivity, elasticity, wrinkles, skin tone, pore condition, pigmentation, and keratin condition. The skin characteristic determining unit 102 may determine the user's skin characteristic through a separate measuring device (eg, a measuring mask) in addition to the user terminal 200 .

The skin characteristic determination unit 102 may divide the obtained user's face image into a plurality of facial regions and determine skin characteristics for each of the plurality of divided facial regions. The plurality of face regions may include eye rims, forehead, cheeks, mouth rims, chin, neck, and the like.

The skin characteristic determiner 102 may determine the skin characteristic of a first face region having an area greater than or equal to a first area among the plurality of divided face regions. For example, the skin characteristic determination unit 102 may determine skin characteristics of the forehead and cheeks, which are large areas among the plurality of face regions.

The skin characteristic determination unit 102 may determine skin characteristics of the second facial region exhibiting similar skin characteristics to those of the first facial region through skin characteristics of the first facial region. For example, since the forehead and the chin generally have similar moisture and oil content, and the cheeks, nose, and eye area have similar moisture and oil content, the skin characteristic determination unit 102 determines the skin characteristics of the chin through the skin characteristics of the forehead. , and the skin characteristics of the nose and eyes can be determined through the skin characteristics of the cheeks.

The skin characteristic determination unit 102 determines the user's skin characteristics based on moisture, oil, sebum, pH, sensitivity, wrinkles, skin tone, pore condition, pigmentation, dead skin condition, etc. of the skin for each of a plurality of facial areas. can

Determining cosmetics to be recommended based on the user's skin characteristics corresponding to the point of purchase through the current user's face image is the skin characteristics that are greatly affected by the external environment and those of us with high variability in the external environment due to seasonal changes. It may be inappropriate due to the climate characteristics of the country. In other words, if you purchase cosmetics that reduce oiliness in the summer when oiliness increases, and the season when oiliness decreases while using cosmetics approaches, it may be unsuitable to continuously use the cosmetics that reduce oiliness.

Accordingly, the skin characteristic determining unit 102 may determine the user's skin characteristic corresponding to a specific time point by additionally considering the skin characteristic variation determined through the skin characteristic variation determining unit 104 below.

The use continuation time determining unit 103 may determine a use continuation time point, which is a specific time point after a certain period from the user's purchase time point. A detailed description of this will be described in detail with reference to FIG. 4 below.

The skin characteristic change degree determination unit 104 may determine the skin characteristic change degree corresponding to the duration of use. In other words, a degree of change in skin characteristics indicating how much the user's skin characteristics corresponding to the purchase point determined through the skin characteristic determining unit 102 will change after a certain period of time may be determined. For example, when the user purchases cosmetics in June, the skin characteristic change determination unit 104 may determine the user's skin characteristic change corresponding to September after a predetermined period from June. The usage duration point may refer to a specific point in time at which the use period of the cosmetic product may be extended if the cosmetic product is determined based on skin characteristics at the specific time point.

The skin characteristic gradient determination unit 104 may perform supervised learning or unsupervised learning on a skin characteristic estimation model according to the skin characteristic gradient, and the skin characteristic estimation model is an artificial neural network (ANN). can be implemented

The skin characteristic estimation model may be supervised and learned by using the 'season (date)' and 'skin characteristics according to the season (date)' by the skin characteristic estimation model learning unit. That is, the above-described 'season (date)' and 'skin characteristics according to the season (date)' are input and output values, respectively, and may be used as training data for supervised learning of a cosmetics recommendation model.

For example, the skin characteristic estimation model learning unit converts 'season (date)' into a unique third one-hot vector to generate an input value, and generates an input value, 'skin characteristic according to season (date)' After generating an output value by converting to a unique fourth one-hot vector, a skin characteristic estimation model may be supervised using the generated input value and output value. Here, the third one-hot vector and the fourth one-hot vector may be vectors in which one of the component values constituting the vector is '1' and the other component values are '0'.

The cosmetics determining unit 106 may determine necessary components and effects necessary for the user based on the user's skin characteristics, and may determine cosmetics including the determined necessary components and necessary effects. Specifically, the cosmetics determining unit 106 determines the necessary ingredients and needs of the user based on the skin characteristics of the user corresponding to the point of purchase and the skin characteristics of the user corresponding to the time of continued use, which are determined based on the skin characteristics of the user corresponding to the point of purchase and the change in skin characteristics. The effect can be determined, and cosmetics including the determined necessary component and necessary effect can be determined.

The cosmetics determining unit 106 may determine the cosmetics based on cosmetics use information of other users distinguished from the user, as well as the determined necessary ingredients and necessary effects. The cosmetics determining unit 106 may determine the cosmetics based on cosmetics use information of other users whose work similarity determined by the work similarity determining unit 105 is greater than or equal to a threshold value.

The recommendation list generator 107 may generate a recommendation list of cosmetics for the user based on the determined cosmetics. The recommendation list generator 107 may transmit the generated recommendation list and information on cosmetics included in the recommendation list to the user terminal 200 .

3 is a diagram for explaining that a user's skin characteristics vary according to regional climate characteristics. Referring to FIG. 3 , it is possible to know the change trend of climate characteristics and skin characteristics for each region during spring, summer, fall and winter.

Referring to FIG. 3 , a graph of climate characteristics (eg, temperature and humidity) according to seasonal changes is shown. Korea has clear four seasons, and it varies by region, but in general, the highest temperature is recorded in July and August, which is summer, and then the temperature gradually drops and becomes the lowest temperature in January and February. Similarly, precipitation and humidity may be high in summer and may decrease as winter progresses.

Referring to FIG. 3, among the skin characteristics greatly affected by the external environment, the moisture content, sebum amount, pores, and elasticity of the skin are generally proportional to the climate characteristics graph, and among the skin characteristics, wrinkles, pH, skin brightness, and skin sensitivity are climate characteristics. It can be seen that it is generally inversely proportional to the characteristic graph.

FIG. 4 is a diagram explaining a process of determining cosmetics suitable for skin characteristics corresponding to a user's duration of use of cosmetics.

The use continuation time determining unit 103 may determine a use continuation time, which is after a certain period of time from the user's purchase time. The use duration determining unit 103 may determine the duration of use based on the user's use of cosmetics. Recommending cosmetics suitable for the user's skin characteristics corresponding to the point of purchase of cosmetics may not be suitable for continuous use, considering the use period of the cosmetics and the skin characteristics that change during the use period. In other words, since the user's skin characteristics change after purchasing the cosmetics in FIG. Cosmetics suitable for the user's skin characteristics corresponding to the duration of use determined by ) may be determined.

However, if a point too close to the point of purchase is determined as the point of continuation of use, it may not be suitable for the user for a long period after the point of continuation of use, and if a point of time too far from the point of purchase is determined as the point of continuation of use, May not be suitable for long term users. Accordingly, the usage duration determination unit 103 may determine the usage duration duration based on the user's cosmetic use pattern.

The usage duration determining unit 103 may obtain the user's average usage period for the user's used cosmetics received through the user terminal 200 . The cosmetics used by the user may be the same type of cosmetics as the cosmetics the user intends to purchase.

When the user's average usage period is less than or equal to the threshold period, the usage duration determination unit 103 determines a time point after 1/2 of the average usage period from the purchase time as the usage duration time point, and the user's When the average usage period exceeds the critical period, a time point after 1/4 of the average usage period from the purchase time may be determined as the usage continuation time point.

The skin characteristic gradient determining unit 104 may determine the user's skin characteristic gradient by comparing the climate corresponding to the purchase time and the climate corresponding to the continuous use time. The skin characteristic gradient determining unit 104 determines the climate corresponding to the time of purchase and the time of continued use based on climate information such as monthly average temperature, monthly maximum temperature, monthly minimum temperature, and monthly average humidity for each region obtained from the Korea Meteorological Administration server. corresponding climates can be compared.

The skin characteristic gradient determining unit 104 determines a temperature change rate between the temperature of the user's activity area corresponding to the purchase time and the user's activity area temperature corresponding to the duration of use, and determines the temperature change rate of the user's activity area at the purchase time. A humidity change rate between the corresponding humidity and the humidity corresponding to the duration of use may be determined.

The skin characteristic gradient determining unit 104 determines the gradient of moisture content, sebum amount, pores, and elasticity among the user's skin characteristics in proportion to the temperature change rate and humidity change rate, and determines the change rate of the user's skin characteristics in inverse proportion to the temperature change rate and humidity change rate. Changes in wrinkles, pH, skin brightness, and skin sensitivity can be determined.

The skin characteristic determination unit 102 may determine the user's skin characteristics corresponding to the duration of use based on the determined change in skin characteristics, and the cosmetics determination unit 106 may determine the user's skin characteristics corresponding to the duration of use. Based on this, cosmetics to be recommended to the user may be determined.

FIG. 5 is a diagram for explaining determining work similarity in order to consider cosmetics usage information of other users. Cosmetics use information of other users can be stored and collected as big data.

The cosmetics determining unit 106 may determine cosmetics suitable for the user based on the user's skin characteristics determined by the skin characteristic determining unit 102 as well as cosmetics usage information of other users.

Since the skin characteristics of people are different according to the climate of each region and there is a high probability that the skin characteristics are similar in the same region, the cosmetics determining unit 106 may determine a plurality of users in the same activity area as the user among other users. .

The cosmetics determining unit 106 may determine users with similar working environments among the plurality of determined users, since the probability that the skin characteristics are similar is high as the working environment is similar.

Specifically, the cosmetics determining unit 106 may first determine first users among the plurality of users who have the same working type as the user and have a difference between the user and the average monthly out-of-office time within a first hour. Work types may include office work, field work, and rotational work.

The cosmetics determining unit 106 may determine users whose work similarity is greater than or equal to a threshold value among the determined first users as the second users. The work similarity indicates the similarity of the working environment between the user and other users. The higher the working similarity, the more similar the user and the working environment are, and the lower the working similarity, the less similar the user and the working environment are.

The work similarity determination unit 105 may calculate the degree of change in outside work and the degree of change in outside work for each of the first users based on the collected outside work time for the first users. At this time, the out-of-office hours serving as basic data may be determined by data within the last year or the last 6 months. Out-of-office time may be expressed as hours (days) according to each month. At this time, the degree of increase in outside work may be calculated using the 'slope code' of the outside work time for each month. External muscle sensitization can be expressed as a vector.

For example, if there is a decrease (-), increase (+), decrease (-), increase (+), decrease (-), increase (+) of out-of-work hours from March to August, then from March to August The extrinsic muscle sensitivity corresponding to may be (-1, +1, -1, +1, -1, +1). As another example, if the out-of-office hours were the same (0), decreased (-), increased (+), increased (+), decreased (-), or increased (+) from March to August, then from March to August The corresponding extensor sensitivity may be (0, -1, +1, +1, -1, +1).

In other words, if the amount of out-of-office hours decreased in the current month compared to the previous month, the component value of the increase/decrease in outer work corresponding to the current month is '-1'. The component value of is '0', and if the out-of-office time increases in the current month compared to the previous month, the component value of the increase in out-of-work corresponding to the current month may be '+1'.

On the other hand, the degree of change in outside work may be calculated using the 'size of the slope' of the outside work time for each month. For example, if the out-of-office hours fluctuate from -2 days, +4 days, -3 days, +2 days, -4 days, +1 days from March to August, the degree of change in out-of-office hours is (2, 4, 3 , 2, 4, 1). That is, the component value of the degree of change in outside work corresponding to the current month may indicate a date of change in outside work time.

The work similarity determiner 105 may calculate an increase/decrease correlation (ID) between the user's reference increase in external work and the increase in external work of each of the first users based on Equation 1 below.

In Equation 1, I _a is the user's standard external muscle sensitization, I _o means the external muscle sensitization of each of the first users, and the numerator is the user's standard external muscle sensitization and the external muscle sensitization of each of the first users. It can be a dot product operation between degrees.

The work similarity determiner 105 may calculate the Euclidean distance (DIS) between the user's standard external shift and the external shift of each of the first users based on Equation 2 below.

In Equation 2, _Ga is the user's standard external shift, _Go is the external shift of each of the first users, and the work similarity determiner 105 determines the increase and decrease correlation (ID) The work similarity (S _w ) according to Equation 3 below can be calculated using the Dian distance (DIS).

Referring to Equation 3, k is an integer between 0 and 1 as a weight coefficient for determining the weight between the incremental association (ID) and the Euclidean distance (DIS), and DISμ is the determined first users. It may be an average value of the calculated Euclidean distances, and DISσ may be a standard deviation of the calculated Euclidean distances for the determined first users.

The cosmetics determining unit 106 may determine users whose work similarity (S _w ) is greater than or equal to a threshold value among the first users as the second users.

The cosmetics determination unit 106 may determine cosmetics suitable for the user based on the cosmetics usage information of the second users acquired through the information acquisition unit 101 . Cosmetics use information may include purchase reviews, use period, use frequency, return status, and the like.

The cosmetics determining unit 106 may increase the weight of the cosmetics in the case of cosmetics having a positive review rate equal to or greater than the first rate. In the case of cosmetics having a positive review rate less than the first rate, the cosmetics determining unit 106 may lower the weight of the cosmetics. The cosmetics determining unit 106 may determine cosmetics suitable for the user based on weights of the cosmetics.

FIG. 6 is a diagram showing the hardware configuration of the cosmetics recommendation server 100 according to FIG. 1 .

Referring to FIG. 6, the cosmetics recommendation server 100 stores at least one processor 110 and instructions instructing the at least one processor 110 to perform at least one operation. May contain memory.

The at least one operation may include at least some of the operations or functions of the cosmetics recommendation server 100 and may be implemented in the form of instructions and performed by the processor 110 .

Here, the at least one processor 110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor for performing methods according to embodiments of the present invention. can Each of the memory 120 and the storage device 160 may include at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory 120 may be one of a read only memory (ROM) and a random access memory (RAM), and the storage device 160 may be a flash-memory. , a hard disk drive (HDD), a solid state drive (SSD), or various memory cards (eg, a micro SD card).

In addition, the cosmetics recommendation server 100 may include a transceiver 130 that performs communication through a wireless network. In addition, the cosmetics recommendation server 100 may further include an input interface device 140, an output interface device 150, a storage device 160, and the like. Each component included in the cosmetics recommendation server 100 is connected by a bus 170 to communicate with each other.

The methods according to the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on a computer readable medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in the art of computer software.

Examples of computer readable media may include hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions may include not only machine language codes generated by a compiler but also high-level language codes that can be executed by a computer using an interpreter and the like. The hardware device described above may be configured to operate with at least one software module to perform the operations of the present invention, and vice versa.

In addition, the above-described method or device may be implemented by combining all or some of its components or functions, or may be implemented separately.

Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art will variously modify and change the present invention within the scope not departing from the spirit and scope of the present invention described in the claims below. You will understand that it can be done.

100: Cosmetics recommendation server 200: user terminal

Claims (1)

A cosmetics recommendation server that provides a service for recommending customized cosmetics to the user based on the user's skin characteristics and cosmetics use information of other users belonging to a similar environment to the user,
an information acquisition unit that obtains a face image of the user and activity area information indicating an activity area of the user from the user terminal of the user;
a skin characteristic determination unit configured to determine skin characteristics of the user based on the obtained face image;
a cosmetics determining unit which determines cosmetics including necessary ingredients and necessary effects required for the user based on the determined skin characteristics; and
A recommendation list generating unit generating a recommendation list of the determined cosmetics and transmitting the recommendation list to the user terminal;
The skin characteristic determining unit,
Based on the face image, at least one skin characteristic among moisture, oil, sebum, wrinkles, pore state, and dead skin condition of the user's skin is determined according to the purchase time point, and the user's skin characteristics corresponding to the purchase time point; and Based on the change in skin characteristics for a predetermined period of time from the purchase time, determining the user's skin characteristics corresponding to the duration of use, which is a time point after the predetermined period from the purchase time;
The cosmetics determination unit determines the cosmetics based on the user's skin characteristics corresponding to the duration of use and cosmetics use information of other users distinguished from the user;
The cosmetics recommendation server,
Further comprising a skin characteristic gradient determination unit for determining a skin characteristic gradient of the user that varies according to a climate corresponding to an active region of the user;
The skin characteristic change degree determining unit,
Based on climate information for each region obtained from the Korea Meteorological Administration server, comparing the climate of the active region corresponding to the time of purchase and the climate of the active region corresponding to the time of continuing use,
Determine the degree of change in the skin characteristics of the user based on the comparison,
determining a rate of change in temperature and humidity between the temperature and humidity of the user's active area corresponding to the purchase time and the temperature and humidity of the user's active area corresponding to the duration of use;
Determine the degree of change in moisture content, sebum amount, pores, and elasticity among the user's skin characteristics in proportion to the determined temperature change rate and humidity change rate;
Determining changes in wrinkles, pH, skin brightness, and skin sensitivity among skin characteristics of the user so as to be inversely proportional to the determined rate of change in temperature and rate of change in humidity;
The cosmetic decision unit,
determine first users among the other users in the same active area as the user;
Among the determined first users, determining second users having the same work type as the user and having a difference between the user and the monthly average outgoing time within a first hour;
Among the determined second users, determining users whose work similarity, which indicates a similarity of working environments between the user and the second users, is equal to or greater than a threshold value as third users;
Determining the cosmetics based on cosmetics use information of the third users, wherein the cosmetics use information includes at least one of purchase reviews, use period, use frequency, and return status;
The cosmetics recommendation server further includes a work similarity determination unit,
The work similarity determining unit,
For the second users, based on the time spent outside, calculating the degree of change in outside work and the degree of change in outside work for each of the second users;
Calculating an increase/decrease correlation between the user's standard external muscle sensitivity and the external muscle sensitivity of each of the second users;
Calculate a Euclidean distance between the user's standard external muscle gradient and the external muscle gradient of each of the second users;
Cosmetics recommendation server for determining the work similarity based on the Euclidean distance and the increasing/decreasing correlation.

Images