Skin care scheme management system and method based on AI drive

Abstract

The invention discloses a skincare scheme management system and method based on AI drive, and relates to the technical field of dynamic skincare scheme adjusting.The skincare scheme management method comprises the following steps that skincare historical data and skincare real-time data are collected; performing data preprocessing; calculating an initial skin state evaluation value by taking the data after data preprocessing as a reference and combining a preset weight of an initial skin care index; dynamically updating the contribution degree of the skin care index based on the skin state change of the skin care historical data in a plurality of periods and the corresponding initial skin state evaluation value, further updating the weight of the skin care index, and calculating the current skin state evaluation value in real time; and comparing a preset scheme to adjust a triggering condition, performing product adjustment recommendation when the triggering condition is exceeded, and adjusting the use frequency according to the variable quantity of the skin state evaluation value. According to the method, a dynamic weight updating mechanism is adopted, the influence of each index on the skin state evaluation value is continuously optimized, and personalized and iterable skin care state modeling is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic skin care plan adjustment, and specifically to an AI-driven skin care plan management system and method. Background Art

[0002] With the improvement of people's living standards, skin care has been increasingly emphasized. Traditional skin care plans are often based on experience or general skin type classification, lacking personalization and precision. However, everyone's skin condition and needs are unique and are comprehensively affected by various factors;

[0003] At the same time, the continuous progress of technology has brought new opportunities to the skin care field. Various skin care devices can measure various basic physiological indicators of the skin, and the popularization of smartphones and various APPs also enables users to conveniently record and upload data such as facial pictures and living habits; however, traditional methods lack the integration and analysis of multi-source data such as living habits and environmental factors, and it is difficult to comprehensively reflect the dynamic changes of skin conditions; existing evaluation models mostly calculate skin conditions using fixed weights and cannot dynamically adjust the importance of indicators according to individual differences or skin condition evolution, resulting in a large deviation between the evaluation results and actual needs; most systems rely on historical data to generate one-time skin care suggestions and do not dynamically adjust the plan in combination with real-time device data, making it difficult to cope with the impact of environmental or behavioral mutations on the skin. Summary of the Invention

[0004] The purpose of the present invention is to provide an AI-driven skin care plan management system and method to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: An AI-driven skin care plan management method, and the skin care plan management method specifically includes the following steps:

[0006] Collect skin care historical data and skin care real-time data;

[0007] The skin care historical data includes skin care basic data and user feedback data;

[0008] The skin care real-time data includes one or more skin care device data;

[0009] Regularly update the skin care basic data and the user feedback data;

[0010] Real-time collect one or more of the skin care device data;

[0011] Perform data preprocessing on the collected skin care historical data and skin care real-time data;

[0012] Based on the pre - processed skin care historical data and real - time skin care data, combined with the preset weights of each initial skin care index, calculate the initial skin state evaluation value;

[0013] Based on the skin state changes of the skin care historical data in several cycles and the corresponding initial skin state evaluation value, dynamically update the contribution degrees of each skin care index, and then update the weights of each skin care index, and calculate the current skin state evaluation value in real - time;

[0014] Among them, the contribution degree of each skin care index is positively correlated with the weight;

[0015] According to the currently calculated current skin state evaluation value, compare it with the preset scheme adjustment trigger condition. When it exceeds the scheme adjustment trigger condition, recommend product adjustment;

[0016] And adjust the usage frequency according to the change amount of the skin state evaluation value.

[0017] Among them, the collection of skin care historical data and real - time skin care data specifically includes the following skin care indexes:

[0018] Skin moisture content: Use a high - precision skin moisture tester to measure the moisture content of the stratum corneum of the skin through the capacitive or resistive principle;

[0019] Among them, preferably, the user can use a portable device at home, attach the probe to the skin surface, and the device automatically records the data and synchronizes it to the APP;

[0020] Sebum secretion amount: Use a sebum meter to measure the sebum secretion amount on the skin surface by using the reflection or absorption characteristics of oil on specific wavelengths of light; Measure after a certain time after cleansing, record the data of different parts, and measure periodically;

[0021] Skin pH value: Use a pH test strip or a professional pH meter to lightly touch the skin with the test strip or contact the skin with the probe of the meter to understand the acid - base balance state of the skin, and measure periodically;

[0022] Facial picture data: The user takes a facial photo through a specific shooting interface of the mobile APP at a fixed light source and a fixed distance. The APP can provide shooting guidance, such as prompting the user to keep the face clean and the expression natural to ensure the photo quality;

[0023] Living habit data: Obtain data such as the user's daily sleep time, exercise duration, and diet intake by docking with a health management APP;

[0024] Environmental data: Access the public data interface of the meteorological department to obtain real - time environmental temperature, humidity, and ultraviolet intensity data;

[0025] Perform data preprocessing on the collected skin care historical data and skin care real-time data, specifically as follows:

[0026] The data preprocessing includes data cleaning and data standardization;

[0027] Preferably, the data cleaning includes outlier processing, and the specific steps are as follows:

[0028] Step S2-1-1: Perform outlier processing on the collected skin care historical data, and sort the data of the same skin care index in ascending order; calculate the upper quartile, median, and lower quartile of the ascending sequence; the upper quartile, median, and lower quartile of the ascending sequence are the data values corresponding to the upper quartile point Q1, the median point Q2, and the lower quartile point Q3 respectively; ; ; ; where m represents the number of data in the ascending sequence, and m is a positive integer; Q1 and Q2 are rounded up; Q3 is rounded down;

[0029] Step S2-1-2: Calculate the difference between the upper quartile and the lower quartile; denote the absolute value of the difference between the upper quartile and the lower quartile as IQR;

[0030] Step S2-1-3: Take the upper quartile as the upper limit and the lower quartile as the lower limit, and divide the data greater than the upper quartile plus 1.5 times IQR and the data less than the lower quartile minus 1.5 times IQR into outliers;

[0031] Step S2-1-4: Replace the outliers greater than the upper quartile plus 1.5 times IQR with the value of the upper quartile plus 1.5 times IQR, and replace the outliers less than the lower quartile minus 1.5 times IQR with the value of the lower quartile minus 1.5 times IQR;

[0032] The data standardization is specifically as follows:

[0033] For numerical data, perform data standardization by the Max-Min method, x norm = (x - x min ) / (x max - x min ), where x norm represents the data value after data standardization; x represents the original data value; x max and x min represent the maximum and minimum values of the data of the same skin care index respectively;

[0034] For categorical data, use one-hot encoding to convert it into a numerical vector;

[0035] Establish the mapping relationship between the numerical vector and the score according to the historical data;

[0036] Based on the pre-processed skin care historical data and real-time skin care data, combined with the preset weights of each initial skin care index, an initial skin condition evaluation value is calculated, specifically as follows:

[0037] ;

[0038] Among them, S represents the skin condition evaluation value; n represents the number of skin care indexes, and n is a positive integer; w i (0) represents the preset weights of each initial skin care index, and i represents the category label of the skin care index; x norm represents the data value after data standardization. Among them, for categorical data, x norm represents the mapping score corresponding to the numerical vector;

[0039] Among them, the preset weights of each initial skin care index are evenly distributed;

[0040] Based on the skin condition changes of the skin care historical data in several cycles and the corresponding initial skin condition evaluation value, dynamically update the contribution degree of each skin care index, and then update the weights of each skin care index, and calculate the current skin condition evaluation value in real time, specifically as follows:

[0041] Step S1: Adopt the method of association analysis to calculate the correlation coefficient between the change amount of the skin care index in the historical data in several cycles and the skin condition evaluation value. Preferably, use the Pearson correlation coefficient formula:

[0042] ;

[0043] Among them, r i represents the correlation coefficient between the change amount of the skin care index and the skin condition evaluation value; m represents the number of historical data, and m is a positive integer; j represents the number label of the historical data; represents the change amount of the i-th skin care index in the j-th historical data; represents the average value of the change amount of the i-th skin care index; S j represents the skin condition evaluation value in the j-th historical data; represents the average value of the skin condition evaluation value;

[0044] Step S2: According to the correlation coefficient r i , calculate the contribution degree C i of each skin care index, and the representation formula is:

[0045] ;

[0046] Among them, C i represents the contribution degree of the skin care index;

[0047] Step S3. According to the calculated contribution degrees C of each skin care index i , update the weights of each skin care index. The iterative formula for weight update is:

[0048] ;

[0049] where, w i (t) represents the weight of each skin care index after t iterations; w i (t - 1) represents the weight of each skin care index after t - 1 iterations; γ represents the adjustment coefficient; C i (t - 1) represents the contribution degree of the corresponding skin care index at the (t - 1)-th iteration;

[0050] When the contribution degree of a certain skin care index increases, its weight increases accordingly; on the contrary, when the contribution degree decreases, the weight decreases. By continuously updating the weight according to the new contribution degree, the calculation method of the skin state evaluation value can more accurately reflect the influence of each skin care index on the overall skin state, so as to provide more accurate skin care plan adjustment suggestions for users;

[0051] According to the currently calculated skin state evaluation value in real time, compare it with the preset plan adjustment trigger condition, and when it exceeds the plan adjustment trigger condition, recommend product adjustment. Specifically:

[0052] Step S1. Set a skin state evaluation threshold. When the change amount ΔS of the skin state evaluation value S exceeds the skin state evaluation threshold T, or the change amount of a certain single skin care index exceeds the corresponding threshold T i , trigger product adjustment recommendation, that is, when |ΔS| > T or |ΔS i | > T i , start product adjustment recommendation;

[0053] Step S2. Establish an association matrix M between the historical product efficacy and skin care indexes. The element M pi in the association matrix M represents the influence value of the p-th product on the i-th skin care index, which is represented by the change amount of the x norm value after using this product in the historical data;

[0054] Step S3. According to the change of the skin state evaluation value and the association matrix M, calculate the correlation degree between the change amount of the skin care index evaluation value and the influence value of the product on the skin care index, and conduct product adjustment recommendation. The formula is:

[0055] ;

[0056] where, P best represents the recommended historical product; ΔS irepresents the change amount of the evaluation value of the i-th skin care index; p represents the product number;

[0057] Preferably, the usage frequency is adjusted regularly according to the change amount of the skin state evaluation value, specifically:

[0058] ;

[0059] where f new represents the adjusted usage frequency; f old represents the usage frequency before adjustment; k represents the adjustment coefficient; S max represents the maximum value of the skin state evaluation value;

[0060] An AI-driven skin care plan management system, the skin care plan management system includes a data acquisition module, a data preprocessing module, a simulation calculation module, and an AI-driven module;

[0061] The data acquisition module is responsible for collecting skin care historical data and skin care real-time data. The skin care historical data includes skin care basic data and user feedback data; the skin care real-time data includes several types of skin care device data;

[0062] The skin care device data is collected and stored through the APP;

[0063] Regularly update the skin care basic data and user feedback data;

[0064] Real-time update the skin care device data;

[0065] The data preprocessing module is responsible for data preprocessing; the data preprocessing includes data cleaning and data standardization; for numerical data, data standardization is performed by the Max-Min method; for categorical data, the one-hot encoding method is used to convert it into a numerical vector;

[0066] The simulation calculation module is used to calculate the initial skin state evaluation value based on the preprocessed skin care historical data and skin care real-time data, combined with the preset weights of the initial skin care indicators; and based on the skin state changes of the skin care historical data in several cycles and the corresponding initial skin state evaluation values, dynamically update the contribution degrees of the skin care indicators, and then update the weights of the skin care indicators, and calculate the current skin state evaluation value in real time;

[0067] The AI-driven module is used to compare the preset plan adjustment trigger conditions according to the currently calculated skin state evaluation value in real time. When the plan adjustment trigger conditions are exceeded, product adjustment recommendations are made, and the usage frequency is adjusted according to the change amount of the skin state evaluation value.

[0068] The simulation calculation module includes a data storage unit, a skin condition evaluation unit, and an index weight iteration unit;

[0069] The data storage unit is used to store the data that has completed data preprocessing;

[0070] The skin condition evaluation unit is used to calculate the skin condition evaluation value based on the preprocessed skin care historical data and real-time skin care data;

[0071] The index weight iteration unit is used to dynamically update the contribution degree of skin care indexes based on the skin condition changes of skin care historical data in several cycles and the corresponding initial skin condition evaluation value, and then update the weights of skin care indexes.

[0072] The AI-driven module includes a threshold control unit, an association matrix unit, a product recommendation unit, and a frequency update unit;

[0073] The threshold control unit is used to statistically obtain the skin condition evaluation threshold based on historical data;

[0074] The association matrix unit is used to construct an association matrix M based on the historical product efficacy and skin care indexes;

[0075] The product recommendation unit is used to make product adjustment recommendations by calculating the correlation degree between the change amount of the skin care index evaluation value and the influence value of the product on the skin care index;

[0076] The frequency update unit is used to adjust the usage frequency according to the change amount of the skin condition evaluation value.

[0077] Compared with the prior art, the beneficial effects of the present invention are as follows: By integrating skin care historical data and real-time data, combining multi-dimensional skin care indexes such as moisture, oil, pH value, image, living habits, and environment, the present invention constructs a comprehensive skin condition evaluation system. This method improves the data quality through data preprocessing, and adopts a dynamic weight update mechanism, combining Pearson correlation coefficient and contribution degree calculation, continuously optimizing the influence of each index on the skin condition evaluation value, and realizing personalized and iterative skin care state modeling. The system intelligently recommends suitable products and dynamically adjusts the usage frequency according to the skin condition changes, combined with the historical product effect matrix, so as to improve the accuracy and effectiveness of the skin care plan. At the same time, this method supports synchronous data collection by portable devices and APPs, enhances the convenience of user operation, and has the ability to dock with health management and environmental data interfaces, with strong scalability, and is suitable for constructing an intelligent and closed-loop personalized skin care management system. Description of the Drawings

[0078] Figure 1 It is a step schematic diagram of a skin care plan management method based on AI driving according to the present invention;

[0079] Figure 2 This is the component structure diagram of a skin care plan management system based on AI drive according to the present invention. Specific embodiments

[0080] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0081] Embodiment: As Figure 1 - Figure 2 shown, the present invention provides a technical solution, a skin care plan management method based on AI drive. The skin care plan management method specifically includes the following steps:

[0082] Collect skin care historical data and skin care real-time data;

[0083] The skin care historical data includes skin care basic data and user feedback data;

[0084] The skin care real-time data includes one or more skin care device data;

[0085] Regularly update the skin care basic data and user feedback data;

[0086] Real-time collect one or more skin care device data;

[0087] Perform data preprocessing on the collected skin care historical data and skin care real-time data;

[0088] Based on the skin care historical data and skin care real-time data after data preprocessing, combined with the preset weights of each initial skin care index, calculate the initial skin condition evaluation value;

[0089] Based on the skin condition changes and the corresponding initial skin condition evaluation values of the skin care historical data within several cycles, dynamically update the contribution degrees of each skin care index, and then update the weights of each skin care index, and calculate the current skin condition evaluation value in real time;

[0090] Among them, the contribution degree and weight of each skin care index show a positive correlation;

[0091] According to the current skin condition evaluation value calculated in real time, compare the preset plan adjustment trigger conditions, and perform product adjustment recommendations when the plan adjustment trigger conditions are exceeded;

[0092] And adjust the usage frequency according to the change amount of the skin condition evaluation value.

[0093] Among them, historical skin care data and real-time skin care data are collected, specifically including the following skin care indicators:

[0094] Skin moisture content: Use a high-precision skin moisture tester to measure the moisture content of the stratum corneum of the skin through the capacitive or resistive principle;

[0095] Among them, preferably, the user can use a portable device at home, attach the probe to the skin surface, and the device automatically records the data and synchronizes it to the APP;

[0096] Sebum secretion amount: Use a sebum meter to measure the sebum secretion amount on the skin surface by using the reflection or absorption characteristics of oil on specific wavelength light; Measure after a certain time (2 hours) after cleansing the face, record the data of different parts, and measure periodically;

[0097] Skin pH value: Use a pH test strip or a professional pH meter to gently touch the skin with the test strip or contact the skin with the probe of the meter to understand the acid-base balance state of the skin, and measure periodically;

[0098] Facial picture data: The user takes a facial photo through a specific shooting interface of the mobile APP at a fixed light source and a fixed distance. The APP can provide shooting guidance, such as prompting the user to keep the face clean and the expression natural to ensure the photo quality;

[0099] Lifestyle data: By docking with a health management APP (such as a sports bracelet, a sleep monitoring APP), obtain data such as the user's daily sleep time, exercise duration, and dietary intake (such as sugar, oil, and vitamin intake);

[0100] Environmental data: Connect to the public data interface of the meteorological department to obtain real-time environmental temperature, humidity, and ultraviolet intensity data;

[0101] Perform data preprocessing on the collected historical skin care data and real-time skin care data, specifically:

[0102] Data preprocessing includes data cleaning and data standardization;

[0103] Data cleaning includes outlier processing, and the specific steps are as follows:

[0104] Step S2-1-1: Perform outlier processing on the collected historical skin care data, and sort the data of the same skin care indicator in ascending order; Calculate the upper quartile, median, and lower quartile of the ascending sequence; The upper quartile, median, and lower quartile of the ascending sequence are the data values corresponding to the upper quartile point Q1, the median point Q2, and the lower quartile point Q3 respectively; ; ; ; where m represents the number of data in the ascending sequence, and m is a positive integer; Q1 and Q2 are rounded up; Q3 is rounded down;

[0105] Step S2-1-2: Calculate the difference between the upper quartile and the lower quartile; Denote the absolute value of the difference between the upper quartile and the lower quartile as IQR.

[0106] Step S2-1-3: Take the upper quartile as the upper limit and the lower quartile as the lower limit, and divide the data greater than the upper quartile plus 1.5 times IQR and the data less than the lower quartile minus 1.5 times IQR into outliers.

[0107] Step S2-1-4: Replace the outliers greater than the upper quartile plus 1.5 times IQR with the value of the upper quartile plus 1.5 times IQR, and replace the outliers less than the lower quartile minus 1.5 times IQR with the value of the lower quartile minus 1.5 times IQR.

[0108] Data standardization is specifically as follows:

[0109] For numerical data, data standardization is performed by the Max-Min method, x norm = (x - x min ) / (x max - x min ), where x norm represents the data value after data standardization; x represents the original data value; x max and x min respectively represent the maximum and minimum values of the same skin care index data;

[0110] For categorical data, the one-hot encoding method is used to convert it into a numerical vector;

[0111] According to historical data, establish a mapping relationship between the numerical vector and the score;

[0112] One-hot encoding of categorical data converts non-numerical categorical data (skin type) into a numerical vector for easy processing by machine learning models;

[0113] The original categorical data is: ["dry", "oily", "combination", "sensitive"];

[0114] After one-hot encoding, it is converted into a vector:

[0115] Dry: [1, 0, 0, 0]; Oily: [0, 1, 0, 0];

[0116] Combination: [0, 0, 1, 0]; Sensitive: [0, 0, 0, 1];

[0117] Establishment of the mapping relationship between the numerical vector and the score:

[0118] Analyze the influence of category features on the skin condition evaluation value through historical data analysis, and assign scoring weights corresponding to different categories:

[0119] Based on the statistical analysis of user feedback data, establish the mapping relationship between the numerical vector and the score, and normalize the score to the interval [0, 1];

[0120]

[0121] Table 1

[0122] Solve the regression coefficients corresponding to different skin types through machine learning:

[0123]

[0124] Table 2

[0125] Among them, in Table 2, if the user has dry skin, the weight of the skin moisture index is increased to 0.72 times the currently detected moisture value;

[0126] For users with oily skin, the weight of the sebum secretion amount is 0.55 times the current sebum value;

[0127] Based on the preprocessed historical skin care data and real-time skin care data, combined with the preset initial weights of each skin care index, calculate the initial skin condition evaluation value, specifically:

[0128] ;

[0129] Among them, S represents the skin condition evaluation value; n represents the number of skin care indexes, and n is a positive integer; w i (0) represents the preset initial weights of each skin care index, i represents the category label of the skin care index; x norm represents the data value after data standardization. Among them, for categorical data, x norm represents the mapping score corresponding to the numerical vector;

[0130] Among them, the preset initial weights of each skin care index are evenly distributed;

[0131] Based on the skin condition changes and the corresponding initial skin condition evaluation values of the skin care historical data in several cycles, dynamically update the contribution degrees of each skin care index, and then update the weights of each skin care index, and calculate the current skin condition evaluation value in real time, specifically:

[0132] Step S1: Adopt the method of association analysis to calculate the correlation coefficient between the change amount of the skin care index in the historical data in several cycles and the skin condition evaluation value. Preferably, use the Pearson correlation coefficient formula:

[0133] ;

[0134] Among them, r i represents the correlation coefficient between the change amount of the skin care index and the skin state evaluation value; m represents the number of historical data, and m is a positive integer; j represents the label of the number of historical data; represents the change amount of the i-th skin care index in the j-th historical data; represents the average value of the change amounts of the i-th skin care index; S j represents the skin state evaluation value in the j-th historical data; represents the average value of the skin state evaluation values;

[0135] Step S2: According to the correlation coefficient r i , calculate the contribution degree C i of each skin care index, and the representation formula is:

[0136] ;

[0137] Among them, C i represents the contribution degree of the skin care index;

[0138] Step S3: According to the calculated contribution degrees C i of each skin care index, update the weights of each skin care index, and the iterative formula for weight update is:

[0139] ;

[0140] Among them, w i (t) represents the weight of each skin care index after t iterations; w i (t - 1) represents the weight of each skin care index after t - 1 iterations; γ represents the adjustment coefficient; C i (t - 1) represents the contribution degree of the corresponding skin care index at the (t - 1)-th iteration;

[0141] When the contribution degree of a certain skin care index increases, its weight increases accordingly; conversely, when the contribution degree decreases, the weight decreases; by continuously updating the weight according to the new contribution degree, the calculation method of the skin state evaluation value can more accurately reflect the influence of each skin care index on the overall skin state, so as to provide more accurate skin care plan adjustment suggestions for users;

[0142] According to the currently calculated current skin state evaluation value, compare it with the preset scheme adjustment trigger condition, and when it exceeds the scheme adjustment trigger condition, recommend product adjustment, specifically:

[0143] Step S1: Set the skin state evaluation threshold. When the change amount ΔS of the skin state evaluation value S exceeds the skin state evaluation threshold T, or the change amount of a certain single skin care index exceeds the corresponding threshold T iWhen ∣ΔS∣>T or ∣ΔS i ∣>T i When this happens, initiate the product adjustment recommendation;

[0144] Historical data statistics: Collect the change data of the comprehensive skin condition score of a large number of users during the normal skin care process over a period of time (one month); Analyze these data and calculate their mean μ and standard deviation σ;

[0145] Determine the threshold range: According to the historical statistical law, 95% of the data falls within the range of the mean plus or minus 2σ, and approximately 99.7% of the data falls within the range of the mean plus or minus 3σ;

[0146] Set the threshold T range to μ + 3σ, and if it exceeds this value, initiate the scheme adjustment process;

[0147] The threshold for a single skin care index is obtained using the same method.

[0148] Step S2: Establish the association matrix M between the historical product efficacy and skin care indexes. The element M in the association matrix M pi represents the influence value of the p-th product on the i-th skin care index, which is represented by the change amount of the x norm value after using this product in the historical data;

[0149] Step S3: According to the change of the skin condition evaluation value and the association matrix M, by calculating the correlation degree between the change amount of the skin care index evaluation value and the influence value of the product on the skin care index, perform the product adjustment recommendation. The formula is:

[0150] ;

[0151] Among them, P best represents the recommended historical product; ΔS i represents the change amount of the i-th skin care index evaluation value; p represents the product number;

[0152] Preferably, regularly adjust the usage frequency according to the change amount of the skin condition evaluation value. Specifically:

[0153] ;

[0154] Among them, f new represents the adjusted usage frequency; f old represents the usage frequency before adjustment; k represents the adjustment coefficient; S max represents the maximum value of the skin condition evaluation value;

[0155] An AI-driven skin care scheme management system. The skin care scheme management system includes a data acquisition module, a data preprocessing module, a simulation calculation module, and an AI-driven module;

[0156] The data acquisition module is responsible for collecting historical skin care data and real-time skin care data. The historical skin care data includes basic skin care data and user feedback data; the real-time skin care data includes several types of skin care device data.

[0157] The skin care device data is collected and stored through the APP.

[0158] The basic skin care data and user feedback data are updated regularly.

[0159] The skin care device data is updated in real time.

[0160] The data preprocessing module is responsible for data preprocessing; data preprocessing includes data cleaning and data standardization; for numerical data, data standardization is performed by the Max-Min method; for categorical data, one-hot encoding is used to convert it into a numerical vector.

[0161] The simulation calculation module is used to calculate the initial skin state evaluation value based on the preprocessed historical skin care data and real-time skin care data, combined with the preset weights of the initial skin care indicators; and based on the skin state changes and the corresponding initial skin state evaluation values of the historical skin care data in several cycles, dynamically update the contribution degree of the skin care indicators, and then update the weights of the skin care indicators to calculate the current skin state evaluation value in real time.

[0162] The AI-driven module is used to compare the preset scheme adjustment trigger conditions according to the currently calculated skin state evaluation value in real time. When the scheme adjustment trigger conditions are exceeded, product adjustment recommendations are made, and the usage frequency is adjusted according to the change amount of the skin state evaluation value.

[0163] The simulation calculation module includes a data storage unit, a skin state evaluation unit, and an index weight iteration unit;

[0164] The data storage unit is used to store the data that has completed data preprocessing;

[0165] The skin state evaluation unit is used to calculate the skin state evaluation value based on the preprocessed historical skin care data and real-time skin care data;

[0166] The index weight iteration unit is used to dynamically update the contribution degree of the skin care indicators based on the skin state changes and the corresponding initial skin state evaluation values of the historical skin care data in several cycles, and then update the weights of the skin care indicators.

[0167] The AI-driven module includes a threshold control unit, an association matrix unit, a product recommendation unit, and a frequency update unit;

[0168] The threshold control unit is used to statistically obtain the skin state evaluation threshold based on historical data;

[0169] The association matrix unit is used to construct an association matrix M based on the historical product efficacy and skin care indicators of use;

[0170] The product recommendation unit is used to recommend product adjustments by calculating the correlation degree between the change amount of the skin care indicator evaluation value and the influence value of the product on the skin care indicator;

[0171] The frequency update unit is used to adjust the usage frequency according to the change amount of the skin condition evaluation value.

[0172] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be construed as limiting the claimed invention.

Claims

1. A skin care program management method based on AI, characterized by: The skin care regimen management method specifically comprises the following steps: Collect historical skin care data and real-time skin care data; Performing data preprocessing on the collected skin care history data and skin care real-time data; The initial skin condition evaluation value is calculated based on the historical skin care data and real-time skin care data after data preprocessing and the preset weights of the initial skin care indicators; Based on the skin condition changes in the skin care history data within several cycles and the corresponding initial skin condition evaluation value, the contribution of the skin care index is dynamically updated, and then the weight of the skin care index is updated, and the current skin condition evaluation value is calculated in real time; Based on the skin condition evaluation value calculated in real time, the preset scheme adjustment trigger conditions are compared, and product adjustment recommendations are made when the scheme adjustment trigger conditions are exceeded, and the frequency of use is adjusted according to the change in the skin condition evaluation value.

2. The AI-driven skin care program management method according to claim 1, characterized in that: Collect skin care historical data and real-time skin care data, specifically: The skin care history data includes skin care basic data and user feedback data; The real-time skin care data includes skin care equipment data; Regularly update the skin care basic data and the user feedback data; The skin care device data is collected in real time.

3. The AI-driven skin care program management method according to claim 2, characterized in that: Specifically: The data preprocessing includes data cleaning and data standardization.

4. The AI-driven skin care program management method according to claim 3, characterized in that: The data standardization is specifically as follows: For numerical data, the Max-Min method is used to standardize the data. norm =(xx min ) / (x max -x min ), where x norm represents the data value after data standardization; x represents the original data value; x max and x min Respectively represent the maximum and minimum values ​​of the same skin care index data; For categorical data, one-hot encoding is used to convert it into a numerical vector.

5. The AI-driven skin care program management method according to claim 4, characterized in that: Specifically: The initial skin condition evaluation value is calculated based on the historical skin care data and real-time skin care data after data preprocessing and the preset weights of the initial skin care indicators; The weights of the preset initial skin care indicators are evenly distributed.

6. The AI-driven skin care program management method according to claim 5, characterized in that: Based on the skin condition changes in the skin care history data within several cycles and the corresponding initial skin condition evaluation value, the contribution of the skin care index is dynamically updated, and then the weight of the skin care index is updated, and the current skin condition evaluation value is calculated in real time, which specifically includes the following steps: Step S6-1, using a correlation analysis method to calculate the correlation coefficient between the skin care index change amount and the skin condition evaluation value in the historical data within a certain period; Step S6-2, calculating the contribution of the skin care index according to the correlation coefficient; Step S6-3: iteratively update the weight of the skin care index according to the calculated contribution of the skin care index.

7. The AI-driven skin care program management method according to claim 6, characterized in that: According to the current skin condition evaluation value calculated in real time, compared with the preset solution adjustment trigger conditions, product adjustment recommendations are made when the solution adjustment trigger conditions are exceeded, specifically: Step S7-1, setting a skin condition evaluation threshold, when the change in the skin condition evaluation value exceeds the skin condition evaluation threshold, triggering a product adjustment recommendation; Step S7-2: Establish a correlation matrix M between the historical product efficacy and skin care indicators. The elements M in the correlation matrix M are pi represents the impact value of the p-th product on the i-th skin care index; Step S7-3: According to the change of the skin condition evaluation value and the correlation matrix M, the correlation between the change of the skin care index evaluation value and the impact value of the product on the skin care index is calculated to make a product adjustment recommendation.

8. An AI-driven skin care program management system, using an AI-driven skin care program management method as described in any one of claims 1 to 7, characterized in that: The skin care program management system includes a data acquisition module, a data preprocessing module, a simulation calculation module and an AI driving module; The data acquisition module is responsible for collecting skin care history data and skin care real-time data, wherein the skin care history data includes skin care basic data and user feedback data; the skin care real-time data includes several types of skin care equipment data; The skin care device data is collected and stored via the APP; Regularly update basic skin care data and user feedback data; Update skin care equipment data in real time; The data preprocessing module is responsible for data preprocessing; the data preprocessing includes data cleaning and data standardization; For numerical data, the Max-Min method is used to standardize the data; for categorical data, the one-hot encoding method is used to convert it into a numerical vector; The simulation calculation module is used to calculate the initial skin condition evaluation value based on the skin care history data and the skin care real-time data after data preprocessing, combined with the preset weight of the initial skin care index; and based on the skin condition changes in the skin care history data within several cycles and the corresponding initial skin condition evaluation value, dynamically update the contribution of the skin care index, and then update the weight of the skin care index, and calculate the current skin condition evaluation value in real time; The AI ​​driving module is used to compare the current skin condition evaluation value calculated in real time with the preset scheme adjustment trigger conditions, make product adjustment recommendations when the scheme adjustment trigger conditions are exceeded, and adjust the frequency of use according to the change in the skin condition evaluation value.

9. The AI-driven skin care program management system according to claim 8, characterized in that: The simulation calculation module includes a data storage unit, a skin condition evaluation unit and an index weight iteration unit; The data storage unit is used to store the data after the data preprocessing; The skin condition evaluation unit is used to calculate the skin condition evaluation value based on the skin care history data and the skin care real-time data after data preprocessing; The indicator weight iteration unit is used to dynamically update the contribution of the skin care indicator based on the skin condition changes in the skin care history data within several cycles and the corresponding initial skin condition evaluation value, and then update the weight of the skin care indicator.

10. The AI-driven skin care program management system according to claim 9, characterized in that: The AI ​​driving module includes a threshold control unit, an association matrix unit, a product recommendation unit and a frequency update unit; The threshold control unit is used to obtain a skin condition assessment threshold based on historical data statistics; The association matrix unit is used to construct an association matrix M based on the historical product efficacy and skin care indicators; The product recommendation unit is used to make product adjustment recommendations by calculating the correlation between the change in the skin care index evaluation value and the impact value of the product on the skin care index; The frequency updating unit is used to adjust the use frequency according to the change amount of the skin condition evaluation value.

Images