Process optimization method and system for implementing skin moisturizers

Abstract

The present application relates to the technical field of moisturizer process optimization, and a process optimization method and system for realizing skin moisturizers, comprising: determining a to-be-optimized direction group, designing a multi-dimension option questionnaire based on the to-be-optimized direction, conducting user experience follow-up visits by using a customer experience terminal and the multi-dimension option questionnaire, obtaining a multi-dimension experience report, constructing a user skin quality characteristic matrix by using the multi-dimension experience report and calculating user confidence, correcting the user skin quality characteristic matrix to obtain a corrected skin quality characteristic matrix, inputting the corrected skin quality characteristic matrix into a moisturizer proportioning adjustment model to obtain a moisturizer proportion, and obtaining a target moisturizer proportion by using weighted average on a user confidence set and a moisturizer proportion set. The present application can improve the user universality of moisturizer process optimization and meet the dynamic changes of user demand.

Description

Technical Field

[0001] This invention relates to the field of moisturizer process optimization technology, and in particular to a method and system for optimizing the process of skin moisturizers. Background Technology

[0002] With the increasing demand for personalized skincare and intensifying market competition, optimizing the manufacturing process of skin moisturizers has become crucial. An efficient moisturizer manufacturing process not only needs to meet the needs of different skin types but also needs to adapt quickly to market changes. Therefore, developing an intelligent moisturizer manufacturing process optimization method that comprehensively considers user skin characteristics, usage habits, and environmental factors is of great significance for enhancing product competitiveness, meeting consumer needs, and promoting sustainable development in the industry.

[0003] Traditional technologies typically rely on laboratory testing and user feedback to optimize moisturizer processes. While this approach can optimize moisturizer processes to some extent, its over-reliance on small-scale laboratory data and limited user feedback leads to insufficient data sample size, making it difficult to comprehensively cover different skin types and usage scenarios. This affects the universality of the optimization results. Furthermore, traditional methods lack real-time and dynamic capabilities, failing to respond quickly to market changes and dynamic shifts in user needs. Summary of the Invention

[0004] This invention provides a method and system for optimizing the process of skin moisturizers. Its main purpose is to improve the user applicability of moisturizer process optimization and meet the dynamic changes in user needs.

[0005] To achieve the above objectives, the present invention provides a process optimization method for realizing a skin moisturizer, comprising:

[0006] Receive process optimization instructions, and determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction;

[0007] Extract the directions to be optimized sequentially from the group of directions to be optimized, design multiple-choice questions based on the directions to be optimized, and merge the multiple-choice questions corresponding to each direction to be optimized in the group of directions to be optimized to obtain a multi-dimensional option questionnaire.

[0008] A customer experience terminal set has been identified, wherein the customer experience terminal set includes multiple customer experience terminals, and the customer experience terminals include a mobile APP;

[0009] The customer experience is extracted sequentially from the customer experience terminal. The user experience is then used to conduct a follow-up survey using the customer experience terminal and a multi-dimensional questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: user numerical option set, user habit option set, and user skin type feature set.

[0010] A user skin feature matrix is ​​constructed using the user numerical option set and user skin feature set in the multi-dimensional experience report, and user confidence is calculated based on the user skin feature set.

[0011] Based on the user's habit option group, the user's skin texture feature matrix is ​​modified according to user habits to obtain the modified skin texture feature matrix;

[0012] The modified skin texture feature matrix is ​​input into a pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio, wherein the moisturizer ratio adjustment model is a trained neural network model;

[0013] The moisturizer ratios and user confidence scores are summarized to obtain a moisturizer ratio set and a user confidence score set. The moisturizer ratio set is then weighted and averaged using the user confidence score set to obtain the target moisturizer ratio. Based on the target moisturizer ratio, the process optimization for achieving skin moisturizing is completed.

[0014] Optionally, the design of the multi-option problem based on the direction to be optimized includes:

[0015] Based on the direction to be optimized, a binary skin texture problem group is determined. The binary skin texture problem group includes multiple binary skin texture problems, and the binary skin texture problems include: options yes and options no.

[0016] Binary skin texture problems are extracted sequentially from the binary skin texture problem group. An extended numerical problem group is designed based on the binary skin texture problems. The extended numerical problem group includes multiple extended numerical problems, and each extended numerical problem includes multiple numerical options related to skin texture problems.

[0017] By summarizing the aforementioned extended numerical problem groups, we obtain the extended numerical problem group set;

[0018] Identify a user habit question group, which includes multiple user habit questions, and each user habit question includes multiple numerical options related to user usage habits;

[0019] The design includes optional input items, such as image input items.

[0020] Based on the aforementioned optional input items, user habit question group, binary skin texture question group, and extended numerical question group, the design of the multi-option question is completed.

[0021] Optionally, the user experience feedback is conducted using a customer experience terminal and a multi-dimensional questionnaire to obtain a multi-dimensional experience report, including:

[0022] A multi-dimensional options questionnaire is sent to the customer experience platform, where pre-confirmed users fill out the questionnaire.

[0023] If it is confirmed that the multi-dimensional option questionnaire has been completed, then the completed multi-dimensional option questionnaire will be recorded as a multi-dimensional option answer sheet.

[0024] A multi-dimensional experience report is generated based on the multi-dimensional options questionnaire.

[0025] Optionally, the generation of a multi-dimensional experience report based on a multi-dimensional questionnaire includes:

[0026] Binary skin texture questions are extracted sequentially from the binary skin texture question group of the multi-dimensional option questionnaire, and the user binary options of the binary skin texture questions are identified. The user binary options include: option yes and option no.

[0027] If the user's binary option is yes, then the extended numerical question group corresponding to the binary skin quality question is identified, and the user's numerical option for each extended numerical question in the extended numerical question group is obtained to obtain the user numerical option group.

[0028] If the user's binary option is no, then the preset zero option group is recorded as the user numerical option group. The zero option group consists of multiple zero elements, and the number of zero elements in the zero option group is the same as the number of extended numerical questions in the extended numerical question group.

[0029] Summarize the user value option groups to obtain the user value option group set;

[0030] Identify the user habit option group in the user habit question group of the multi-dimensional option questionnaire, and determine the user skin feature group based on the optional input items in the multi-dimensional option questionnaire;

[0031] A multi-dimensional experience report is generated based on user numerical option groups, user habit option groups, and user skin type feature groups.

[0032] Optionally, determining the user's skin texture feature group based on the optional input items in the multi-dimensional answer sheet includes:

[0033] Determine whether the optional input items in a multi-dimensional answer sheet have been filled in;

[0034] If the optional input items in the multi-dimensional answer sheet are filled in, the user skin texture image group of the optional input items is received, wherein the user skin texture image group includes: before moisturizing and after moisturizing;

[0035] Image detection is performed on the user's skin texture image group to obtain the pre-moisturization feature group and the post-moisturization feature group. The pre-moisturization feature group includes: skin saturation, skin smoothness and skin gloss.

[0036] The user's skin type feature group is obtained by calculating the ratio between the feature group before moisturizing and the feature group after moisturizing.

[0037] If the optional input items in the multi-dimensional answer sheet are not filled in, the preset zero feature group will be recorded as the user's skin texture feature group.

[0038] Optionally, the step of performing image detection on the user's skin texture image group to obtain a pre-moisturizing feature group and a post-moisturizing feature group includes:

[0039] Identify image feature category groups, which include: saturation, smoothness, and glossiness;

[0040] Based on image feature category groups, image detection is performed on the before and after moisturizing images in the user's skin texture image group to obtain the original before moisturizing feature group and the original after moisturizing feature group.

[0041] Based on the preset normal characteristic range group, the original pre-moisturizing characteristic group and the original post-moisturizing characteristic group are screened to obtain the effective pre-moisturizing characteristic group and the effective post-moisturizing characteristic group.

[0042] By performing the same-category screening on the characteristic groups before and after effective moisturization, the target characteristic groups before and after moisturization are obtained.

[0043] Fill the target pre-moisturization feature group and target post-moisturization feature group into the image feature category group respectively to obtain the pre-moisturization feature group and post-moisturization feature group. The empty positions in the pre-moisturization feature group and post-moisturization feature group are filled with zero values.

[0044] Optionally, the step of constructing a user skin texture feature matrix using the user numerical option set and user skin texture feature set from the multi-dimensional experience report includes:

[0045] The number of options in each user value option group in the user value option group set is determined, wherein the number of options in different user value option groups in the user value option group set is the same;

[0046] Based on the number of options, the user skin feature group is expanded to obtain an expanded skin feature group, wherein the number of expanded skin features in the expanded skin feature group is the same as the number of options, and the expansion method includes: squaring the data in the user skin feature group;

[0047] Based on the user's numerical option set and extended skin texture feature set, a user skin texture feature matrix is ​​constructed, where the user skin texture feature matrix is ​​represented as follows:

[0048]

[0049] Where R represents the user's skin texture feature matrix, (D 1,1 … D 1,n ) represents the first user value option group in the user value option group set, D 1,1 This refers to the first user value option in the first user value option group, D. 1,n This represents the nth user value option in the first user value option group, where n represents the number of options. (D m,1 … D m,n D represents the m-th user value option group in the user value option group set, where m represents the number of user value option groups in the set. m,1 D represents the first user value option in the m-th user value option group. m,n This represents the nth user value option in the mth user value option group, (K1 … K n K represents the extended skin texture feature group, K1 represents the first extended skin texture feature in the extended skin texture feature group, and K n This represents the nth extended skin texture feature.

[0050] Optionally, calculating user confidence based on user skin texture feature groups includes:

[0051] Determine the user experience terminal address and experience report date, and obtain the user environment data set for the user experience terminal address at the experience report date. The user environment data set includes: climate dryness index and ultraviolet intensity.

[0052] Determine the number of valid features in the user skin feature group and count the number of valid environments in the user environment data group. The number of valid features is the number of data in the user skin feature group that is not zero, and the number of valid environments is the number of user environment data in the user environment data group.

[0053] Based on the number of valid features and valid environments, the user confidence score is calculated using the following formula:

[0054]

[0055] Where C represents the user confidence level, α1 and α2 represent the preset feature coefficient and preset environment coefficient, respectively, S1 represents the number of effective features, S2 represents the number of effective environments, and e represents the natural logarithm.

[0056] Optionally, the step of modifying the user's skin texture feature matrix according to user habit option groups to obtain a modified skin texture feature matrix includes:

[0057] Construct the user habit vector of the user habit option group;

[0058] Select a reference environment data group from the preset standard environment data group that corresponds to the user environment data group;

[0059] Based on user habit vectors, user environment data sets, and reference environment data sets, the user skin texture feature matrix is ​​adjusted to obtain a modified skin texture feature matrix, wherein the modified skin texture feature matrix is ​​represented as follows:

[0060]

[0061] Where R' represents the modified skin texture feature matrix. This represents the preset custom weight vector. This represents the user habit vector, where p represents the number of user environment data points in the user environment data group or the number of reference environment data points in the reference environment data group. This represents the i-th user environment data in the user environment data group. This represents the i-th reference environment data in the reference environment data group.

[0062] To achieve the above objectives, the present invention also provides a process optimization system for realizing skin moisturizers, comprising:

[0063] The questionnaire design module is used to receive process optimization instructions, determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction. The optimization direction is extracted sequentially from the optimization direction group, and multiple-choice questions are designed based on the optimization direction. The multiple-choice questions corresponding to each optimization direction in the optimization direction group are merged to obtain a multi-dimensional questionnaire.

[0064] The experience report generation module is used to identify the customer experience terminal set, which includes multiple customer experience terminals, including a mobile APP. The customer experience terminals are extracted sequentially from the customer experience terminal set, and user experience feedback is conducted using the customer experience terminals and a multi-dimensional option questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: user numerical option group set, user habit option group, and user skin texture feature group.

[0065] The skin texture matrix construction module is used to construct a user skin texture feature matrix using the user numerical option set and user skin texture feature set in the multi-dimensional experience report, calculate the user confidence based on the user skin texture feature set, and perform user habit correction on the user skin texture feature matrix according to the user habit option set to obtain the corrected skin texture feature matrix.

[0066] The formula ratio optimization module is used to input the modified skin texture feature matrix into a pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio. The moisturizer ratio adjustment model is a trained neural network model. The module summarizes the moisturizer ratio and user confidence scores to obtain a moisturizer ratio set and a user confidence score set. The user confidence score set is used to perform a weighted average on the moisturizer ratio set to obtain the target moisturizer ratio.

[0067] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0068] Memory, storing at least one instruction;

[0069] The processor executes the instructions stored in the memory to implement the process optimization method for realizing the skin moisturizer described above.

[0070] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the above-described process optimization method for realizing a skin moisturizer.

[0071] To address the problems described in the background art, this invention first receives process optimization instructions and identifies the optimization direction group. This step precisely locates the functional areas that the moisturizer needs to improve, ensuring that resources are focused on the most critical issues. Next, multiple-choice questions are designed and a multi-dimensional questionnaire is constructed. This step comprehensively and accurately collects users' specific needs and feedback under different skin types, providing a structured tool for subsequent user experience follow-up. Then, the customer experience endpoint set is confirmed and user experience follow-up is conducted. Through multiple customer experience endpoints, a large amount of actual user experience data can be easily and quickly obtained. This not only increases data diversity but also improves user engagement, providing rich first-hand data for subsequent analysis. Importantly, a user skin type feature matrix is ​​constructed and user confidence scores are calculated. This step transforms the collected user experience data into a quantitative matrix, making the data more structured and easier to analyze. Simultaneously, calculating user confidence scores can assess the reliability of the data. The process of adjusting the skin texture matrix provides quality assurance for subsequent model training and formulation. Next, the user's skin texture feature matrix is ​​corrected based on user habits. This step considers the impact of user habits on skin texture features, enabling the data to more accurately reflect the user's actual situation, improving the accuracy of model predictions, and ensuring that the recommended moisturizer formulation better meets the user's actual needs. The corrected matrix is ​​then input into the moisturizer formulation adjustment model. Using a trained neural network model, the optimal moisturizer formulation can be predicted quickly and efficiently from the corrected skin texture feature matrix, significantly shortening the R&D cycle and improving the efficiency of process optimization. Finally, the data is summarized and weighted to obtain the target moisturizer formulation. By comprehensively analyzing the collected formulation data and confidence levels, a target formulation that balances user needs and data reliability can be derived. This step ensures that the final moisturizer formula meets the needs of most users, thereby achieving effective process optimization.

[0072] Therefore, the present invention can improve the user applicability of humectant process optimization and meet the dynamic changes in user needs. Attached Figure Description

[0073] Figure 1 This is a schematic flowchart of a process optimization method for realizing a skin moisturizer according to an embodiment of the present invention;

[0074] Figure 2 A functional block diagram of a process optimization system for realizing a skin moisturizer provided in an embodiment of the present invention;

[0075] Figure 3 This is a schematic diagram of an electronic device for implementing the process optimization method for realizing a skin moisturizer, according to an embodiment of the present invention.

[0076] Explanation of reference numerals in the attached figures:

[0077] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0078] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0079] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0080] This application provides a method for optimizing the process of implementing a skin moisturizer. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for optimizing the process of implementing a skin moisturizer can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0081] Reference Figure 1 The diagram shown is a flowchart illustrating a process optimization method for realizing a skin moisturizer according to an embodiment of the present invention. In this embodiment, the process optimization method for realizing a skin moisturizer includes:

[0082] S1. Receive process optimization instructions, and determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction.

[0083] Understandably, the process optimization instruction refers to a human-initiated instruction to optimize the process of the moisturizer. The optimization direction group includes multiple optimization directions, and the optimization direction refers to the function of the moisturizer that needs to be improved in this process optimization. For example, the dryness optimization direction aims to improve the moisturizing effect of the moisturizer on dry skin, the oiliness optimization direction focuses on improving the applicability of the moisturizer to oily skin, and the sensitivity optimization direction focuses on improving the mildness and tolerance of the moisturizer to sensitive skin.

[0084] S2. Extract the directions to be optimized sequentially from the group of directions to be optimized, design multiple-choice questions based on the directions to be optimized, and merge the multiple-choice questions corresponding to each direction to be optimized in the group of directions to be optimized to obtain a multi-dimensional option questionnaire.

[0085] It is clear that the multiple-choice questions refer to artificially designed questions about the area to be optimized. These questions reflect the improvement effect of the moisturizer on the user's skin quality in the area to be optimized, and the questions are designed by professionals. The merging of the multiple-choice questions refers to putting multiple multiple-choice questions into the same questionnaire. The multi-dimensional option questionnaire refers to a questionnaire that includes multiple-choice questions corresponding to different areas to be optimized.

[0086] In detail, the multi-option problem based on the direction to be optimized includes:

[0087] Based on the direction to be optimized, a binary skin texture problem group is determined. The binary skin texture problem group includes multiple binary skin texture problems, and the binary skin texture problems include: options yes and options no.

[0088] Binary skin texture problems are extracted sequentially from the binary skin texture problem group. An extended numerical problem group is designed based on the binary skin texture problems. The extended numerical problem group includes multiple extended numerical problems, and each extended numerical problem includes multiple numerical options related to skin texture problems.

[0089] By summarizing the aforementioned extended numerical problem groups, we obtain the extended numerical problem group set;

[0090] Identify a user habit question group, which includes multiple user habit questions, and each user habit question includes multiple numerical options related to user usage habits;

[0091] The design includes optional input items, such as image input items.

[0092] Based on the aforementioned optional input items, user habit question group, binary skin texture question group, and extended numerical question group, the design of the multi-option question is completed.

[0093] Understandably, the binary skin texture question refers to a question with only two options: yes and no. The extended numerical question group refers to a combination of multiple questions with specific numerical options corresponding to the binary skin texture question. When a user selects "yes" for a binary skin texture question, they will be redirected to this extended numerical question group for further data collection on the user's experience. The user habit question refers to questions reflecting the user's habitual use of moisturizers, such as the number of times a day they use moisturizer, the first time they used moisturizer, etc. These questions all have explicit numerical options. The optional input field refers to questions that users can choose to fill in, such as taking facial images before and after using moisturizer. Since images more clearly demonstrate the effect of moisturizer, but not all users are willing to provide images, this optional input field is provided.

[0094] For example, under the dryness optimization direction, a binary skin type question is designed: Does your skin often feel dry? The options are: Yes and No. This binary skin type question corresponds to the following extended numerical question group: Extended numerical question 1: How frequently does your skin feel dry? The options are: Once a day (severity set to 1), 1 to 3 times a day (severity set to 2), 3 to 5 times or more a day (severity set to 3), 5 times or more a day (severity set to 4); Extended numerical question 2: How dry is your skin? The options are: Occasionally feeling tight (severity set to 1), often feeling tight, with slight flaking (severity set to 2), obvious flaking, roughness (severity set to 3), dryness causing skin cracking (severity set to 4).

[0095] S3. Identify the customer experience terminal set, wherein the customer experience terminal set includes multiple customer experience terminals, and the customer experience terminals include mobile APP.

[0096] Understandably, the "customer experience terminal" refers to the software through which users can receive multi-dimensional questionnaires, such as mobile apps.

[0097] S4. Extract customer experience data sequentially from the customer experience platform, and conduct user experience follow-up using the customer experience platform and a multi-dimensional questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: user numerical option set, user habit option set, and user skin type feature set.

[0098] It is clear that the multi-dimensional experience report refers to the multi-dimensional option questionnaire completed by the user. Among them, the user numerical option set refers to the set of values ​​obtained after completing the binary skin texture question set and the extended numerical question set, the user habit option set refers to the set of values ​​obtained after completing the user habit question set, and the user skin texture feature set refers to the combination of values ​​obtained after analyzing the images in the optional input items.

[0099] In detail, the user experience feedback is conducted using a customer experience platform and a multi-dimensional questionnaire to obtain a multi-dimensional experience report, including:

[0100] A multi-dimensional options questionnaire is sent to the customer experience platform, where pre-confirmed users fill out the questionnaire.

[0101] If it is confirmed that the multi-dimensional option questionnaire has been completed, then the completed multi-dimensional option questionnaire will be recorded as a multi-dimensional option answer sheet.

[0102] A multi-dimensional experience report is generated based on the multi-dimensional options questionnaire.

[0103] It is understood that the multi-dimensional option questionnaire refers to the completed multi-dimensional option questionnaire. It should be understood that a submission option should be set in the multi-dimensional option questionnaire. This option is located at the end of the multi-dimensional option questionnaire. When the user selects the submission option, it is determined that the user has completed the completion of the multi-dimensional option questionnaire.

[0104] In detail, the generation of a multi-dimensional experience report based on a multi-dimensional questionnaire includes:

[0105] Binary skin texture questions are extracted sequentially from the binary skin texture question group of the multi-dimensional option questionnaire, and the user binary options of the binary skin texture questions are identified. The user binary options include: option yes and option no.

[0106] If the user's binary option is yes, then the extended numerical question group corresponding to the binary skin quality question is identified, and the user's numerical option for each extended numerical question in the extended numerical question group is obtained to obtain the user numerical option group.

[0107] If the user's binary option is no, then the preset zero option group is recorded as the user numerical option group. The zero option group consists of multiple zero elements, and the number of zero elements in the zero option group is the same as the number of extended numerical questions in the extended numerical question group.

[0108] Summarize the user value option groups to obtain the user value option group set;

[0109] Identify the user habit option group in the user habit question group of the multi-dimensional option questionnaire, and determine the user skin feature group based on the optional input items in the multi-dimensional option questionnaire;

[0110] A multi-dimensional experience report is generated based on user numerical option groups, user habit option groups, and user skin type feature groups.

[0111] For example, a binary skin type question is: Does your skin often feel dry? The options are: Yes and No. If the user selects Yes, then the user is redirected to the extended numerical question group corresponding to the binary skin type question: Extended numerical question 1: How frequently does your skin feel dry? The options are: Once a day (severity set to 1), 1 to 3 times a day (severity set to 2), 3 to 5 times a day or more (severity set to 3), 5 times a day or more (severity set to 4); Extended numerical question 2: How dry is your skin? The options are: Occasionally feeling tight (severity set to 1), often feeling tight, with slight flaking (severity set to 2), obvious flaking, roughness (severity set to 3), dryness causing skin cracking (severity set to 4). The user's options for the above extended numerical question groups are: Once a day (severity set to 1), often feeling tight, with slight flaking (severity set to 2), so the user's numerical option group is (1,2); If the user selects No, then a zero option group is set: (0,0), and this zero option group is recorded as the user's numerical option group.

[0112] Specifically, determining the user's skin texture feature group based on the optional input items in the multi-dimensional answer sheet includes:

[0113] Determine whether the optional input items in a multi-dimensional answer sheet have been filled in;

[0114] If the optional input items in the multi-dimensional answer sheet are filled in, the user skin texture image group of the optional input items is received, wherein the user skin texture image group includes: before moisturizing and after moisturizing;

[0115] Image detection is performed on the user's skin texture image group to obtain the pre-moisturization feature group and the post-moisturization feature group. The pre-moisturization feature group includes: skin saturation, skin smoothness and skin gloss.

[0116] The user's skin type feature group is obtained by calculating the ratio between the feature group before moisturizing and the feature group after moisturizing.

[0117] If the optional input items in the multi-dimensional answer sheet are not filled in, the preset zero feature group will be recorded as the user's skin texture feature group.

[0118] Understandably, the user skin texture image group refers to the combination of images input by the user in the optional input field, where the before moisturizing image and the after moisturizing image refer to images taken before and after the user applied the moisturizer, respectively. The before moisturizing feature group and the after moisturizing feature group refer to the combination of features used to represent the skin moisturizing effect in the before and after moisturizing images, respectively, where: skin saturation refers to the color saturation of the user's skin in the before moisturizing image, which is used to represent the moisture content of the user's skin; the higher the saturation, the more moisture the skin contains; skin smoothness refers to the smoothness of the image texture in the before moisturizing image, which can be analyzed using the Gray-Level Co-occurrence Matrix (GLCM) and the contrast is recorded as texture smoothness; skin smoothness is used to represent the smoothness of the user's skin, and the higher the smoothness, the greater the skin smoothness; skin gloss refers to the glossiness of the skin area in the before moisturizing image, which is represented by the glossiness index (LI); the higher the skin glossiness, the smoother the user's skin. Corresponding to the explanation of each parameter in the before moisturizing feature group, the same parameters and corresponding explanations are also present in the after moisturizing feature group.

[0119] For example, the feature groups before and after moisturizing are (A1, A2, A3) and (B1, B2, B3), respectively. Here, A1, A2, and A3 represent skin saturation, skin smoothness, and skin radiance in the feature group before moisturizing, respectively. B1, B2, and B3 have the same definition and will not be described again. Taking A1 and B1 as examples, the user's skin texture feature is calculated as (B1-A1) / (1+A1), thus obtaining the user's skin texture feature group.

[0120] Furthermore, when the optional input items are not filled in, in order to ensure the correctness of the subsequent construction of the user skin texture feature matrix, it is necessary to set a zero feature group. This zero feature group includes multiple zeros, and the number of zeros is the same as the number of user skin texture features in the user skin texture feature group.

[0121] Specifically, the step of performing image detection on the user's skin texture image group to obtain a pre-moisturizing feature group and a post-moisturizing feature group includes:

[0122] Identify image feature category groups, which include: saturation, smoothness, and glossiness;

[0123] Based on image feature category groups, image detection is performed on the before and after moisturizing images in the user's skin texture image group to obtain the original before moisturizing feature group and the original after moisturizing feature group.

[0124] Based on the preset normal characteristic range group, the original pre-moisturizing characteristic group and the original post-moisturizing characteristic group are screened to obtain the effective pre-moisturizing characteristic group and the effective post-moisturizing characteristic group.

[0125] By performing the same-category screening on the characteristic groups before and after effective moisturization, the target characteristic groups before and after moisturization are obtained.

[0126] Fill the target pre-moisturization feature group and target post-moisturization feature group into the image feature category group respectively to obtain the pre-moisturization feature group and post-moisturization feature group. The empty positions in the pre-moisturization feature group and post-moisturization feature group are filled with zero values.

[0127] It needs to be explained that the image feature category group refers to the numerical types that the image detection needs to obtain, which can be set as: saturation, smoothness, and gloss. That is to say, to detect an image, it is necessary to obtain the saturation, smoothness, and gloss of the user's skin in the image. The original pre-moisturization feature group and the original post-moisturization feature group refer to the numerical combinations of each category in the above image feature category group obtained after image detection. However, since the images here are taken by the user, and the shooting equipment and shooting environment conditions of different users are different, not all features obtained after image detection are valid features. For example, if the ambient light of a user's shooting environment is low, the accuracy of the gloss features in the pre-moisturization feature group or the post-moisturization feature group will be low. Here, a manually set normal feature range group can be used for filtering: the original pre-moisturization features and original post-moisturization features that are not in the normal feature range group are removed, and the original pre-moisturization feature group and the original post-moisturization feature group after removal are recorded as the valid pre-moisturization feature group and the valid post-moisturization feature group. Since the feature categories in the feature group before and after effective moisturizing are different, in order to calculate the user's skin texture feature group in the subsequent calculation, it is necessary to perform same-category filtering: extract the intersection between the feature group before and after effective moisturizing. Here, the intersection refers to the part between them that has the same image feature category. Finally, in order to ensure that the target feature group before and after effective moisturizing for different users have the same number of parameters, the features that have been removed need to be replaced with zero values. The target feature group before and after effective moisturizing after the replacement are the feature group before and after moisturizing.

[0128] S5. Construct a user skin feature matrix using the user numerical option set and user skin feature set from the multi-dimensional experience report, and calculate the user confidence score based on the user skin feature set.

[0129] Understandably, the user skin texture feature matrix refers to a matrix used to represent the moisturizing effect of the moisturizer on the user's skin, and the user confidence score refers to a numerical value characterizing the reliability of the user's skin texture features. Because the completeness of data collected from different users varies—for example, a user may not have the aforementioned image—data with lower user confidence scores will have a lower weight in the subsequent adjustment of the moisturizer formulation.

[0130] In detail, the construction of a user skin texture feature matrix using the user numerical option set and user skin texture feature set from the multi-dimensional experience report includes:

[0131] The number of options in each user value option group in the user value option group set is determined, wherein the number of options in different user value option groups in the user value option group set is the same;

[0132] Based on the number of options, the user skin feature group is expanded to obtain an expanded skin feature group, wherein the number of expanded skin features in the expanded skin feature group is the same as the number of options, and the expansion method includes: squaring the data in the user skin feature group;

[0133] Based on the user's numerical option set and extended skin texture feature set, a user skin texture feature matrix is ​​constructed, where the user skin texture feature matrix is ​​represented as follows:

[0134]

[0135] Where R represents the user's skin texture feature matrix, (D 1,1 … D 1,n ) represents the first user value option group in the user value option group set, D 1,1 This refers to the first user value option in the first user value option group, D. 1,n This represents the nth user value option in the first user value option group, where n represents the number of options. (D m,1 … D m,n D represents the m-th user value option group in the user value option group set, where m represents the number of user value option groups in the set. m,1 D represents the first user value option in the m-th user value option group. m,n This represents the nth user value option in the mth user value option group, (K1 … K n K represents the extended skin texture feature group, K1 represents the first extended skin texture feature in the extended skin texture feature group, and K n This represents the nth extended skin texture feature.

[0136] Understandably, the number of options refers to the number of user value options in the user value option group. The extended skin texture feature group refers to the extended user skin texture feature group. Since the number of user skin texture features in the user skin texture feature group is less than the number of options, the user skin texture feature group needs to be extended so that the number of features in the extended skin texture feature group is the same as the number of options. The extension method is as follows: extract the user skin texture features from the user skin texture feature group sequentially, and process the user skin texture features by squaring or cubed or other preset functions to obtain the extension number. The extension number is then added to the user skin texture feature group. The above steps of sequentially extracting user skin texture features from the user skin texture feature group are repeated until the number of features in the extended user skin texture feature group is equal to the number of options. The above sequential extraction steps are then stopped, and the user skin texture feature group at this time is recorded as the extended skin texture feature group.

[0137] Specifically, the calculation of user confidence based on user skin texture feature groups includes:

[0138] Determine the user experience terminal address and experience report date, and obtain the user environment data set for the user experience terminal address at the experience report date. The user environment data set includes: climate dryness index and ultraviolet intensity.

[0139] Determine the number of valid features in the user skin feature group and count the number of valid environments in the user environment data group. The number of valid features is the number of data in the user skin feature group that is not zero, and the number of valid environments is the number of user environment data in the user environment data group.

[0140] Based on the number of valid features and valid environments, the user confidence score is calculated using the following formula:

[0141]

[0142] Where C represents the user confidence level, α1 and α2 represent the preset feature coefficient and preset environment coefficient, respectively, S1 represents the number of effective features, S2 represents the number of effective environments, and e represents the natural logarithm.

[0143] It is clear that the "experience terminal address" refers to the IP address of the customer's experience terminal, the "experience report date" refers to the date the user completed the multi-dimensional option questionnaire, and the "user environment data group" is a combination of indicators for the user's IP address at the time of the experience report date. The climate dryness index represents the dryness of the experience terminal address. It should be noted that the specific data included in the user environment data group can be manually adjusted. Users' skin types will differ in different regions and climates; therefore, it is necessary to obtain the environmental parameters when the user completed the multi-dimensional option questionnaire. The more user environment data in the user environment data group, the higher the completeness of the environmental parameters. The "effective feature count" refers to the number of features with non-zero values ​​in the user skin type feature group. Due to the steps mentioned above of filling missing positions in the pre-moisturizing and post-moisturizing feature groups with zero values, there are features with zero values ​​in the user skin type feature group. These features will not represent any information about the user's skin type; therefore, counting the effective feature count can reflect the data completeness of the user skin type features. The "feature coefficient" and "environment coefficient" refer to the constants manually set to represent the weights of the effective feature count and the effective environment count, respectively.

[0144] S6. Based on the user habit option group, the user skin feature matrix is ​​modified according to user habits to obtain the modified skin feature matrix.

[0145] Understandably, due to differences in daily habits and environmental conditions among users—for example, the frequency of moisturizing application varies, and the dryness of the climate in different regions differs—the information represented by the aforementioned user skin feature matrix can be affected by these differences. To mitigate this interference, user habit options (such as how many times a day to spray moisturizer) can be included in the questionnaire to obtain data on users' moisturizing habits and environmental data (i.e., user environment data). Based on this habit and environmental data, the user skin feature matrix can be adjusted, resulting in a more representative modified skin feature matrix.

[0146] In detail, the step of adjusting the user's skin texture feature matrix according to user habit option groups to obtain a corrected skin texture feature matrix includes:

[0147] Construct the user habit vector of the user habit option group;

[0148] Select a reference environment data group from the preset standard environment data group that corresponds to the user environment data group;

[0149] Based on user habit vectors, user environment data sets, and reference environment data sets, the user skin texture feature matrix is ​​adjusted to obtain a modified skin texture feature matrix, wherein the modified skin texture feature matrix is ​​represented as follows:

[0150]

[0151] Where R' represents the modified skin texture feature matrix. This represents the preset custom weight vector. This represents the user habit vector, where p represents the number of user environment data points in the user environment data group or the number of reference environment data points in the reference environment data group. This represents the i-th user environment data in the user environment data group. This represents the i-th reference environment data in the reference environment data group.

[0152] It is clear that the user habit vector refers to the vector composed of user habit options in the user habit option group. The standard environmental data group refers to an array of manually set standard environmental data. This standard environmental data group is used to train the neural network model. The number of environmental data in the standard environmental data group must be greater than or equal to the number of environmental data in the user environmental data group. Therefore, a reference environmental data group with the same environmental data category as the user environmental data group needs to be selected from the standard environmental data group. For example, if the standard environmental data group is: (climate dryness index Q1, ultraviolet intensity Q2, relative humidity Q3, sunshine duration Q4), and the user environmental data group is: (climate dryness index E1, ultraviolet intensity E2, relative humidity E3), then the data corresponding to the user environmental data group in the standard environmental data group are: climate dryness index Q1, ultraviolet intensity Q2, relative humidity Q3. Therefore, the reference environmental data group is: (climate dryness index Q1, ultraviolet intensity Q2, relative humidity Q3). The habit weight vector refers to the vector composed of the weights of each element in the manually preset user habit vector.

[0153] Furthermore, in the formula for calculating the corrected skin texture feature matrix mentioned above, through... Elements in the user habit vector can be mapped to the (0,1) interval, thereby dynamically scaling the values ​​of each element in the user skin texture feature matrix. As an environmental compensation term, by comparing the data in the user's environmental data set with the data in the reference environmental data set, the user's skin texture feature matrix can be further scaled so that the environment in which the user's skin texture feature matrix is ​​located is close to that of the reference environmental data set.

[0154] S7. Input the modified skin texture feature matrix into the pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio, wherein the moisturizer ratio adjustment model is a trained neural network model.

[0155] It should be explained that the moisturizer ratio adjustment model refers to a model obtained by pre-training a neural network. This model can obtain a moisturizer ratio that conforms to the skin information contained in the modified skin feature matrix by correcting the skin feature matrix. The training of the neural network model is as follows: First, multiple test subjects are acquired, all located at the same address, and standard environmental data sets are measured at that address. Then, the multiple test subjects experience the moisturizer, and multiple training skin feature matrices are obtained. The acquisition method of these matrices is the same as that of the user skin feature matrices mentioned above. Professional personnel communicate with the multiple test subjects to understand their usage requirements and ideas. Then, the professional personnel improve the original moisturizer ratio to obtain a training moisturizer ratio. The above steps are repeated to obtain a large amount of training data, which consists of a large number of training skin feature matrices and a large number of training moisturizer ratios (the specific amount of data is determined manually and is not specifically limited here). The large amount of training data is used to train the neural network to obtain a moisturizer ratio adjustment model. The number of neurons in the input layer of this neural network is the same as the number of matrix elements in the training skin feature matrices, and the number of neurons in the output layer is the same as the number of each material in the training moisturizer ratio.

[0156] Furthermore, the humectant ratio refers to the output value of the humectant ratio adjustment model.

[0157] For example, after inputting a certain skin texture feature matrix into the moisturizer ratio adjustment model, the model outputs: hyaluronic acid: 1.5%, glycerin: 7%, butylene glycol: 3%, allantoin: 0.7%, plant extract: 2%, deionized water: 82.5%, which yields a moisturizer ratio of: 1.5%: 7%: 3%: 0.7%: 2%: 82.5%.

[0158] S8. Summarize the moisturizer ratios and user confidence scores to obtain a moisturizer ratio set and a user confidence score set. Use the user confidence score set to perform a weighted average on the moisturizer ratio set to obtain the target moisturizer ratio. Based on the target moisturizer ratio, complete the process optimization for achieving skin moisturizer.

[0159] It should be explained that the target moisturizer ratio refers to a value obtained after weighted averaging, and the formula involved in the weighted averaging is as follows:

[0160]

[0161] Among them, F m The target moisturizer ratio is represented by z, which represents the number of user confidence scores or the number of moisturizer ratios. j F represents the confidence level of the j-th user in the user confidence set.j C represents the j-th humectant formulation in the set of humectant formulations. ALL Represents the sum of all user confidence scores in the user confidence set, where in item C... j ×F j In the middle, it is C j With F j The weights of each material in the formula are multiplied together. For example, in a formula of 1.5%:7%:3%:0.7%:2%:82.5%, 1.5% is the weight of one material.

[0162] To address the problems described in the background art, this invention first receives process optimization instructions and identifies the optimization direction group. This step precisely locates the functional areas that the moisturizer needs to improve, ensuring that resources are focused on the most critical issues. Next, multiple-choice questions are designed and a multi-dimensional questionnaire is constructed. This step comprehensively and accurately collects users' specific needs and feedback under different skin types, providing a structured tool for subsequent user experience follow-ups. Then, the customer experience endpoint set is confirmed and user experience follow-ups are conducted. Through multiple customer experience endpoints, a large amount of actual user experience data can be easily and quickly obtained. This not only increases data diversity but also improves user engagement, providing rich first-hand data for subsequent analysis. Importantly, a user skin type feature matrix is ​​constructed and user confidence scores are calculated. This step transforms the collected user experience data into a quantitative matrix, making the data more structured and easier to analyze. Simultaneously, calculating user confidence scores can assess the reliability of the data. The process of adjusting the skin texture matrix provides quality assurance for subsequent model training and formulation. Next, the user's skin texture feature matrix is ​​corrected based on user habits. This step considers the impact of user habits on skin texture features, enabling the data to more accurately reflect the user's actual situation, improving the accuracy of model predictions, and ensuring that the recommended moisturizer formulation better meets the user's actual needs. The corrected matrix is ​​then input into the moisturizer formulation adjustment model. Using a trained neural network model, the optimal moisturizer formulation can be predicted quickly and efficiently from the corrected skin texture feature matrix, significantly shortening the R&D cycle and improving the efficiency of process optimization. Finally, the data is summarized and weighted to obtain the target moisturizer formulation. By comprehensively analyzing the collected formulation data and confidence levels, a target formulation that balances user needs and data reliability can be derived. This step ensures that the final moisturizer formula meets the needs of most users, thereby achieving effective process optimization.

[0163] like Figure 2 The diagram shown is a functional block diagram of a process optimization system for realizing skin moisturizers provided in an embodiment of the present invention.

[0164] The process optimization system 100 for realizing skin moisturizers described in this invention can be installed in an electronic device. Depending on the functions implemented, the process optimization system 100 for realizing skin moisturizers may include an option questionnaire design module 101, an experience report generation module 102, a skin type matrix construction module 103, and a formula ratio optimization module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0165] The questionnaire design module 101 is used to receive process optimization instructions, determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction, extract the optimization direction in turn from the optimization direction group, design multi-option questions based on the optimization direction, and merge the multi-option questions corresponding to each optimization direction in the optimization direction group to obtain a multi-dimensional questionnaire.

[0166] The experience report generation module 102 is used to identify a customer experience terminal set, wherein the customer experience terminal set includes multiple customer experience terminals, and the customer experience terminals include a mobile APP. The customer experience terminals are extracted sequentially from the customer experience terminal set, and user experience feedback is conducted using the customer experience terminals and a multi-dimensional option questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: a user numerical option group set, a user habit option group, and a user skin texture feature group.

[0167] The skin texture matrix construction module 103 is used to construct a user skin texture feature matrix using the user numerical option set and user skin texture feature set in the multi-dimensional experience report, calculate the user confidence based on the user skin texture feature set, and perform user habit correction on the user skin texture feature matrix according to the user habit option set to obtain the corrected skin texture feature matrix.

[0168] The formula ratio optimization module 104 is used to input the modified skin texture feature matrix into a pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio. The moisturizer ratio adjustment model is a trained neural network model. The moisturizer ratio and user confidence are summarized to obtain a moisturizer ratio set and a user confidence set. The user confidence set is used to perform a weighted average on the moisturizer ratio set to obtain the target moisturizer ratio.

[0169] In detail, the modules in the process optimization system 100 for realizing skin moisturizers described in this embodiment of the invention employ the same methods as described above during use. Figure 1 The same technical means are used to optimize the process of skin moisturizers as described in the article, and can produce the same technical effect, so they will not be repeated here.

[0170] like Figure 3 The diagram shown is a schematic representation of an electronic device for implementing a process optimization method for producing a skin moisturizer, according to an embodiment of the present invention.

[0171] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a process optimization method program for implementing a skin moisturizer.

[0172] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as code for optimizing the process of a skin moisturizer, but also to temporarily store data that has been output or will be output.

[0173] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device via various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a process optimization method program for implementing skin moisturizers) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0174] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0175] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0176] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management system, thereby enabling functions such as charging management, discharging management, and power consumption management through the power management system. The power supply may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0177] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0178] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0179] The memory 11 in the electronic device 1 stores a process optimization method program for implementing a skin moisturizer, which is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0180] Receive process optimization instructions, and determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction;

[0181] Extract the directions to be optimized sequentially from the group of directions to be optimized, design multiple-choice questions based on the directions to be optimized, and merge the multiple-choice questions corresponding to each direction to be optimized in the group of directions to be optimized to obtain a multi-dimensional option questionnaire.

[0182] A customer experience terminal set has been identified, wherein the customer experience terminal set includes multiple customer experience terminals, and the customer experience terminals include a mobile APP;

[0183] The customer experience is extracted sequentially from the customer experience terminal. The user experience is then used to conduct a follow-up survey using the customer experience terminal and a multi-dimensional questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: user numerical option set, user habit option set, and user skin type feature set.

[0184] A user skin feature matrix is ​​constructed using the user numerical option set and user skin feature set in the multi-dimensional experience report, and user confidence is calculated based on the user skin feature set.

[0185] Based on the user's habit option group, the user's skin texture feature matrix is ​​modified according to user habits to obtain the modified skin texture feature matrix;

[0186] The modified skin texture feature matrix is ​​input into a pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio, wherein the moisturizer ratio adjustment model is a trained neural network model;

[0187] The moisturizer ratios and user confidence scores are summarized to obtain a moisturizer ratio set and a user confidence score set. The moisturizer ratio set is then weighted and averaged using the user confidence score set to obtain the target moisturizer ratio. Based on the target moisturizer ratio, the process optimization for achieving skin moisturizing is completed.

[0188] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0189] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or system capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0190] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0191] Receive process optimization instructions, and determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction;

[0192] Extract the directions to be optimized sequentially from the group of directions to be optimized, design multiple-choice questions based on the directions to be optimized, and merge the multiple-choice questions corresponding to each direction to be optimized in the group of directions to be optimized to obtain a multi-dimensional option questionnaire.

[0193] A customer experience terminal set has been identified, wherein the customer experience terminal set includes multiple customer experience terminals, and the customer experience terminals include a mobile APP;

[0194] The customer experience is extracted sequentially from the customer experience terminal. The user experience is then used to conduct a follow-up survey using the customer experience terminal and a multi-dimensional questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: user numerical option set, user habit option set, and user skin type feature set.

[0195] A user skin feature matrix is ​​constructed using the user numerical option set and user skin feature set in the multi-dimensional experience report, and user confidence is calculated based on the user skin feature set.

[0196] Based on the user's habit option group, the user's skin texture feature matrix is ​​modified according to user habits to obtain the modified skin texture feature matrix;

[0197] The modified skin texture feature matrix is ​​input into a pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio, wherein the moisturizer ratio adjustment model is a trained neural network model;

[0198] The moisturizer ratios and user confidence scores are summarized to obtain a moisturizer ratio set and a user confidence score set. The moisturizer ratio set is then weighted and averaged using the user confidence score set to obtain the target moisturizer ratio. Based on the target moisturizer ratio, the process optimization for achieving skin moisturizing is completed.

[0199] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0200] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0201] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0202] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0203] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A process optimization method for realizing a skin moisturizer, characterized in that, The method includes: Receive process optimization instructions, and determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction; Extract the directions to be optimized sequentially from the group of directions to be optimized, design multiple-choice questions based on the directions to be optimized, and merge the multiple-choice questions corresponding to each direction to be optimized in the group of directions to be optimized to obtain a multi-dimensional option questionnaire. A customer experience terminal set has been identified, wherein the customer experience terminal set includes multiple customer experience terminals, and the customer experience terminals include a mobile APP; The customer experience is extracted sequentially from the customer experience terminal. The user experience is then used to conduct a follow-up survey using the customer experience terminal and a multi-dimensional questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: user numerical option set, user habit option set, and user skin type feature set. A user skin feature matrix is ​​constructed using the user numerical option set and user skin feature set in the multi-dimensional experience report, and user confidence is calculated based on the user skin feature set. Based on the user's habit option group, the user's skin texture feature matrix is ​​modified according to user habits to obtain the modified skin texture feature matrix; The modified skin texture feature matrix is ​​input into a pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio, wherein the moisturizer ratio adjustment model is a trained neural network model; The moisturizer ratios and user confidence scores are summarized to obtain a moisturizer ratio set and a user confidence score set. The moisturizer ratio set is then weighted and averaged using the user confidence score set to obtain the target moisturizer ratio. Based on the target moisturizer ratio, the process optimization for achieving skin moisturizing is completed.

2. The process optimization method for realizing a skin moisturizer as described in claim 1, characterized in that, The multi-option problem based on the direction to be optimized includes: Based on the direction to be optimized, a binary skin texture problem group is determined. The binary skin texture problem group includes multiple binary skin texture problems, and the binary skin texture problems include: options yes and options no. Binary skin texture problems are extracted sequentially from the binary skin texture problem group. An extended numerical problem group is designed based on the binary skin texture problems. The extended numerical problem group includes multiple extended numerical problems, and each extended numerical problem includes multiple numerical options related to skin texture problems. By summarizing the aforementioned extended numerical problem groups, we obtain the extended numerical problem group set; Identify a user habit question group, which includes multiple user habit questions, and each user habit question includes multiple numerical options related to user usage habits; The design includes optional input items, such as image input items. Based on the aforementioned optional input items, user habit question group, binary skin texture question group, and extended numerical question group, the design of the multi-option question is completed.

3. The process optimization method for realizing a skin moisturizer as described in claim 2, characterized in that, The user experience feedback was conducted using a customer experience platform and a multi-dimensional questionnaire, resulting in a multi-dimensional experience report, including: A multi-dimensional options questionnaire is sent to the customer experience platform, where pre-confirmed users fill out the questionnaire. If it is confirmed that the multi-dimensional option questionnaire has been completed, then the completed multi-dimensional option questionnaire will be recorded as a multi-dimensional option answer sheet. A multi-dimensional experience report is generated based on the multi-dimensional options questionnaire.

4. The process optimization method for realizing a skin moisturizer as described in claim 3, characterized in that, The multi-dimensional experience report generated from the multi-dimensional questionnaire includes: Binary skin texture questions are extracted sequentially from the binary skin texture question group of the multi-dimensional option questionnaire, and the user binary options of the binary skin texture questions are identified. The user binary options include: option yes and option no. If the user's binary option is yes, then the extended numerical question group corresponding to the binary skin quality question is identified, and the user's numerical option for each extended numerical question in the extended numerical question group is obtained to obtain the user numerical option group. If the user's binary option is no, then the preset zero option group is recorded as the user numerical option group. The zero option group consists of multiple zero elements, and the number of zero elements in the zero option group is the same as the number of extended numerical questions in the extended numerical question group. Summarize the user value option groups to obtain the user value option group set; Identify the user habit option group in the user habit question group of the multi-dimensional option questionnaire, and determine the user skin feature group based on the optional input items in the multi-dimensional option questionnaire; A multi-dimensional experience report is generated based on user numerical option groups, user habit option groups, and user skin type feature groups.

5. The process optimization method for realizing a skin moisturizer as described in claim 4, characterized in that, The method for determining a user's skin texture feature group based on the optional input items in a multi-dimensional answer sheet includes: Determine whether the optional input items in a multi-dimensional answer sheet have been filled in; If the optional input items in the multi-dimensional answer sheet are filled in, the user skin texture image group of the optional input items is received, wherein the user skin texture image group includes: before moisturizing and after moisturizing; Image detection is performed on the user's skin texture image group to obtain the pre-moisturization feature group and the post-moisturization feature group. The pre-moisturization feature group includes: skin saturation, skin smoothness and skin gloss. The user's skin type feature group is obtained by calculating the ratio between the feature group before moisturizing and the feature group after moisturizing. If the optional input items in the multi-dimensional answer sheet are not filled in, the preset zero feature group will be recorded as the user's skin texture feature group.

6. The process optimization method for realizing a skin moisturizer as described in claim 5, characterized in that, The step of performing image detection on the user's skin texture image group to obtain the pre-moisturizing feature group and the post-moisturizing feature group includes: Identify image feature category groups, which include: saturation, smoothness, and glossiness; Based on image feature category groups, image detection is performed on the before and after moisturizing images in the user's skin texture image group to obtain the original before moisturizing feature group and the original after moisturizing feature group. Based on the preset normal characteristic range group, the original pre-moisturizing characteristic group and the original post-moisturizing characteristic group are screened to obtain the effective pre-moisturizing characteristic group and the effective post-moisturizing characteristic group. By performing the same-category screening on the characteristic groups before and after effective moisturization, the target characteristic groups before and after moisturization are obtained. Fill the target pre-moisturization feature group and target post-moisturization feature group into the image feature category group respectively to obtain the pre-moisturization feature group and post-moisturization feature group. The empty positions in the pre-moisturization feature group and post-moisturization feature group are filled with zero values.

7. The process optimization method for realizing a skin moisturizer as described in claim 6, characterized in that, The construction of a user skin texture feature matrix using the user numerical option set and user skin texture feature set from the multi-dimensional experience report includes: The number of options in each user value option group in the user value option group set is determined, wherein the number of options in different user value option groups in the user value option group set is the same; Based on the number of options, the user skin feature group is expanded to obtain an expanded skin feature group, wherein the number of expanded skin features in the expanded skin feature group is the same as the number of options, and the expansion method includes: squaring the data in the user skin feature group; Based on the user's numerical option set and extended skin texture feature set, a user skin texture feature matrix is ​​constructed, where the user skin texture feature matrix is ​​represented as follows: Where R represents the user's skin texture feature matrix, (D 1,1 …D 1,n ) represents the first user value option group in the user value option group set, D 1,1 This refers to the first user value option in the first user value option group, D. 1,n This represents the nth user value option in the first user value option group, where n represents the number of options. (D m,1 … D m,n D represents the m-th user value option group in the user value option group set, where m represents the number of user value option groups in the set. m,1 D represents the first user value option in the m-th user value option group. m,n This represents the nth user value option in the mth user value option group, (K1 … K n K represents the extended skin texture feature group, K1 represents the first extended skin texture feature in the extended skin texture feature group, and K n This represents the nth extended skin texture feature.

8. The process optimization method for realizing a skin moisturizer as described in claim 7, characterized in that, The calculation of user confidence based on user skin texture feature groups includes: Determine the user experience terminal address and experience report date, and obtain the user environment data set for the user experience terminal address at the experience report date. The user environment data set includes: climate dryness index and ultraviolet intensity. Determine the number of valid features in the user skin feature group and count the number of valid environments in the user environment data group. The number of valid features is the number of data in the user skin feature group that is not zero, and the number of valid environments is the number of user environment data in the user environment data group. Based on the number of valid features and valid environments, the user confidence score is calculated using the following formula: Where C represents the user confidence level, α1 and α2 represent the preset feature coefficient and preset environment coefficient, respectively, S1 represents the number of effective features, S2 represents the number of effective environments, and e represents the natural logarithm.

9. The process optimization method for realizing a skin moisturizer as described in claim 8, characterized in that, The step of adjusting the user's skin texture feature matrix according to user habit option groups to obtain a corrected skin texture feature matrix includes: Construct the user habit vector of the user habit option group; Select a reference environment data group from the preset standard environment data group that corresponds to the user environment data group; Based on user habit vectors, user environment data sets, and reference environment data sets, the user skin texture feature matrix is ​​adjusted to obtain a modified skin texture feature matrix, wherein the modified skin texture feature matrix is ​​represented as follows: Where R' represents the modified skin texture feature matrix. This represents the preset custom weight vector. This represents the user habit vector, where p represents the number of user environment data points in the user environment data group or the number of reference environment data points in the reference environment data group. This represents the i-th user environment data in the user environment data group. This represents the i-th reference environment data in the reference environment data group.

10. A process optimization system for realizing skin moisturizers, characterized in that, The system includes: The questionnaire design module is used to receive process optimization instructions, determine the optimization direction group based on the process optimization instructions, wherein the optimization direction group includes: dryness optimization direction, oiliness optimization direction and sensitivity optimization direction. The optimization direction is extracted sequentially from the optimization direction group, and multiple-choice questions are designed based on the optimization direction. The multiple-choice questions corresponding to each optimization direction in the optimization direction group are merged to obtain a multi-dimensional questionnaire. The experience report generation module is used to identify the customer experience terminal set, which includes multiple customer experience terminals, including a mobile APP. The customer experience terminals are extracted sequentially from the customer experience terminal set, and user experience feedback is conducted using the customer experience terminals and a multi-dimensional option questionnaire to obtain a multi-dimensional experience report. The multi-dimensional experience report includes: user numerical option group set, user habit option group, and user skin texture feature group. The skin texture matrix construction module is used to construct a user skin texture feature matrix using the user numerical option set and user skin texture feature set in the multi-dimensional experience report, calculate the user confidence based on the user skin texture feature set, and perform user habit correction on the user skin texture feature matrix according to the user habit option set to obtain the corrected skin texture feature matrix. The formula ratio optimization module is used to input the modified skin texture feature matrix into a pre-trained moisturizer ratio adjustment model to obtain the moisturizer ratio. The moisturizer ratio adjustment model is a trained neural network model. The module summarizes the moisturizer ratio and user confidence scores to obtain a moisturizer ratio set and a user confidence score set. The user confidence score set is used to perform a weighted average on the moisturizer ratio set to obtain the target moisturizer ratio.

Images