Personal profile generator and recommendation engine

Abstract

Systems and methods for generating personal profiles and generating recommendations based on the generated personal profiles. The method includes: generating a plurality of clusters, each of the generated plurality of clusters including at least a set of variables of a user included in the cluster, the set of variables relating to at least one of gender, age, skin color, acne trace, acne frequency, lesion score, or body distribution score; generating a profile for the new user; associating the generated profile into a cluster of the plurality of clusters; and generating, by a machine learning (ML) model, a recommendation for the user based on the associated cluster.

Description

[0001] Cross-references to related applications

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 539,865, filed September 22, 2023, the contents of which are incorporated herein by reference in their entirety. Background Technology

[0003] Consumers often conduct personal research on specific conditions or life stages they believe they may be experiencing or about to experience (such as menopause), and based on their research, choose one or more health products that can alleviate their symptoms. This often leads to self-diagnosis, and consumers select specific health products based on advertising, word-of-mouth, or in-store or online choices. However, self-diagnosis may be incorrect and / or may not consider the root cause of one or more specific symptoms, and the chosen product may not be the most effective for addressing the root cause of the symptoms. This could be because consumers identify incorrect or incomplete information, leading to an incorrect self-diagnosis and / or the chosen product failing to resolve the symptoms the consumer is experiencing. Summary of the Invention

[0004] This summary is provided in a simplified form to introduce some concepts that will be further described in the detailed embodiments below. This summary is not intended to identify key or essential features of the subject matter protected by the claims, nor is it intended to help determine the scope of the subject matter protected by the claims.

[0005] In one example, a computer-implemented method is provided. This method includes: generating multiple clusters; generating a profile for a new user; associating the generated profile with the clusters within the multiple clusters; and generating recommendations for the user based on the associated clusters using a machine learning (ML) model.

[0006] In another example, an apparatus is provided. The apparatus includes: a user interface (UI); a memory; and a processor coupled to the memory, the processor being configured to: control the UI to present a questionnaire; receive responses to the questionnaire via the UI; generate a profile associated with the user based on the received responses to the questionnaire and a captured facial scan; associate the generated profile with clusters of multiple clusters; and execute a machine learning (ML) model to generate recommendations for the user based on the associated clusters.

[0007] In another example, a computer-readable storage medium is provided. This computer-readable storage medium stores instructions that, when executed by a processor, cause the processor to generate a plurality of clusters, each of the generated clusters including at least a menopausal stage and experienced menopausal symptoms; generate a profile for a new user, the generated profile including at least the identified user's menopausal stage and experienced menopausal symptoms; associate the generated profile with the clusters among the plurality of clusters; and generate recommendations for the user based on the associated clusters using a machine learning (ML) model.

[0008] Other technical features will be apparent to those skilled in the art from the following figures, description and claims. Attached Figure Description

[0009] Given the accompanying drawings, a better understanding of this specification will be achieved by reading the following detailed description, in which: Figure 1 An example system is shown for generating personal profiles and generating recommendations based on the generated personal profiles; Figure 2 An example system for generating multiple characters is shown; Figure 3 An example of a recommendation engine for generating interventions for a specific profile is shown; Figure 4 An example timeline showing how the profile updates over time is provided; Figure 5 This paper illustrates a method for generating one or more recommendations from a profile using example computer implementations. Figure 6 This paper illustrates a method for generating one or more recommendations from a profile using example computer implementations. Figures 7A to 7G This example UI illustrates a process that includes creating a profile, receiving data, and generating recommendations; and Figure 8 This is a block diagram illustrating an example computing environment applicable to implementing one or more of the various examples disclosed herein.

[0010] In the drawings, the corresponding reference numerals indicate the corresponding parts. Figures 1 to 8 In the diagram, the system is illustrated. The accompanying figures may not be drawn to scale. Detailed Implementation

[0011] Various specific embodiments and examples will be described in detail with reference to the accompanying drawings. Where possible, the same reference numerals will be used in all the drawings to refer to the same or similar parts. References made throughout this disclosure in connection with particular examples and embodiments are provided for illustrative purposes only and are not intended to limit all examples unless indicated otherwise.

[0012] As discussed in this article, consumers habitually perform self-diagnosis to identify specific conditions or life stages, such as menopause. This further leads to the selection of one or more health products that can alleviate the symptoms of self-diagnosed menopause. It also leads to the selection of specific health products based on advertising, word-of-mouth, or in-store or online choices, without supporting scientific data, as self-diagnosis may be inaccurate and / or may not take into account the underlying causes of one or more specific symptoms. Current solutions fail to adequately consider the co-occurrence of symptoms due to race, income, or cognition in specific conditions, and fail to provide resources to help educate women experiencing menopause and assist them in selecting products or services to treat their specific symptoms based on their profiles.

[0013] Therefore, aspects of this disclosure provide systems and methods for: generating personal profiles for consumers, identifying one or more potential causes of one or more conditions, and using artificial intelligence (AI) to generate overall product and / or lifestyle recommendations to address one or more conditions. In some examples, recommendations are updated over time to incorporate feedback on the implementation of recommendations, changes in the consumer's condition, or changes to the profile. The systems and methods described herein operate in an unconventional manner by: collecting data from various sources and in various formats to form new profiles, converting the collected data to a standardized format, generating clusters of similar profiles and associating new profiles with specific clusters, determining the expected health outcomes of profiles within the clusters and identifying associated interventions, and generating recommendations that, when applied, implement interventions to address the expected health outcomes. In some examples, this disclosure further identifies unexpected co-occurrences between different ethnicities, incomes, and menopausal perceptions to generate clusters and recommendations. Therefore, this disclosure provides a number of technical effects, including: an improved data structure that stores the transformed data initially collected from multiple sources, thereby facilitating improved information retrieval; and an improved recommendation engine that implements a trained machine learning model to optimize the content delivered to the consumer, i.e., the generated recommendations, based on specific historical user characteristics and taking into account changes in the consumer's profile over time.

[0014] For example, this disclosure provides a profile generator for generating profiles for consumers. Profiles include variables related to consumer data, including but not limited to age, ethnicity, income, geographic location, menopausal cognition, and various menopausal-related symptoms. A cluster generator generates clusters of similar profiles. Similar profiles are those whose similarity is above a similarity threshold, determined based on the values ​​of the variables. A recommendation engine generates health outcomes for specific clusters, identifies interventions for those health outcomes, and generates recommendations that include those interventions. The recommendations are then provided to the consumer. In some examples, the recommendation engine receives feedback on the provided recommendations and updates itself based on the received feedback to improve further recommendations. In some examples, the clusters associated with a particular profile change over time. For example, as a consumer's age, symptoms, geographic location, and cognition change (such as based on the implementation of recommended interventions), the profile becomes more closely associated with different clusters. The recommendation engine then generates updated recommendations based on the new clusters.

[0015] The various examples described herein provide proactive applications or suites for delivering physical interventions that will alter health outcomes. In some examples, the application is one aspect of a bundle that includes the application, content, and a gamified process points system.

[0016] In some examples, the clusters associated with a particular profile change over time. For instance, as a consumer's age changes, recommendations for changing the consumer profile are implemented, and so on, the profile shifts from one cluster to another over time. This is called a personal trajectory. Therefore, aspects of this disclosure enable consumers to experience a process of: identifying the causes of menopausal-related symptoms (such as age, geographic location, and co-occurrence of symptoms) and the reasoning behind these causes, as well as the characteristics and traits that consumers may be prone to, and providing recommendations including interventions to understand how to address and mitigate negative characteristics and traits and enhance positive characteristics and traits.

[0017] As described herein, various examples of this application provide technical solutions to the inherent technical problem of identifying similarities between digital profiles based at least in part on digital image analysis and associating newly generated profiles with a set of existing profiles, and further detecting changes in profiles over time to maintain the association between a profile and the closest set of existing profiles. This solution offers at least two technical advantages, including reduced computational resource consumption compared to existing solutions that begin a new user classification process each time a user's profile is updated, and the generation and implementation of an improved set of indexes to facilitate the retrieval of user profiles and / or cluster information for generating user recommendations.

[0018] Figure 1 It is an example system for generating personal profiles and generating recommendations based on the generated personal profiles. Figure 1 The illustrated system 100 is provided for illustrative purposes only. Other examples of system 100 may be used without departing from the scope of this disclosure. In some examples, system 100 generates personal profiles and recommendations based on whether a user is experiencing a specific life stage (such as menopause).

[0019] System 100 includes computing device 102, network 138, and user device 140. Computing device 102 refers to any device that executes computer-executable instructions 106 (e.g., as an application, operating system function, or both) to implement operations and functions associated with computing device 102. In some examples, computing device 102 includes mobile computing devices or any other portable devices. Mobile computing devices include, for example, but not limited to, mobile phones, laptop computers, tablet computers, computing boards, netbooks, gaming devices, and / or portable media players. Computing device 102 may also include less portable devices such as servers, desktop personal computers, kiosks, or desktop devices. Additionally, computing device 102 may represent a set of processing units or other computing devices.

[0020] In some examples, computing device 102 includes at least one processor 108, memory 104 (which includes computer-executable instructions 106), and user interface device 110. Processor 108 includes any number of processing units and is programmed to execute the computer-executable instructions 106. The computer-executable instructions 106 are executed by processor 108, by multiple processors within computing device 102, or by a processor external to computing device 102. In some examples, processor 108 is programmed to execute the computer-executable instructions 106, such as those described in the accompanying drawings (e.g., ...). Figure 8 The computer-executable instructions illustrated herein. In various examples, processor 108 is configured to execute one or more of profile generator 120, cluster generator 122, recommendation engine 124, co-occurrence symptom analyzer 134, and cognitive analyzer 136.

[0021] Memory 104 includes any number of media associated with or accessible by computing device 102. In some examples, memory 104 is internal to computing device 102. In other examples, memory 104 is external to computing device 102, or both internal and external to computing device 102. For example, memory 104 may include both memory components internal to computing device 102 and memory components external to computing device 102. Memory 104 stores data, such as one or more applications 107. Application 107 operates when executed by processor 108 to perform various functions on computing device 102. Application 107 may communicate with peer applications or services, such as network services accessible via network 138. In examples, application 107 represents a server-side service of an application executing in the cloud, such as a cloud server. In some examples, application 107 is an application designed to assist a user through a specific life stage (such as menopause) and generate recommendations for one or more products and / or lifestyle changes to address symptoms believed to be caused by the specific life stage.

[0022] User interface device 110 includes a graphics card for displaying data to and receiving data from a user. User interface device 110 may also include computer-executable instructions (e.g., a driver) for operating the graphics card. Furthermore, user interface device 110 may include a display (e.g., a touchscreen display or a natural user interface) and / or computer-executable instructions (e.g., a driver) for operating the display. User interface device 110 may also include one or more of the following to provide data to or receive data from a user: a speaker, a sound card, a camera, a microphone, a vibration motor, one or more accelerometers, Bluetooth. ® The device includes a communication module, Global Positioning System (GPS) hardware, and a photosensitive light sensor. In a non-limiting example, a user inputs commands or manipulates data by moving the computing device 102 in one or more ways.

[0023] The computing device 102 further includes a communication interface device 112. The communication interface device 112 includes a network interface card and / or computer-executable instructions (such as a driver) for operating the network interface card. Communication between the computing device 102 and other devices (such as, but not limited to, user equipment 140) can occur using any protocol or mechanism via any wired or wireless connection.

[0024] The computing device 102 further includes a data storage device 114 for storing data such as, but not limited to, one or more profiles 116, clusters 118, and / or interventions 119. Data 114 may be data received from user device 140, and / or data received, retrieved, or obtained by one or more of profile generator 120, cluster generator 122, recommendation engine 124, co-occurrence symptom analyzer 134, and cognitive analyzer 136. Profile 116 includes at least one profile of a specific consumer, which includes data values ​​for variables related to the consumer. In some examples, the profiles, data values, and variables depend on the type of one or more applications 107 executed on the computing device 102. For example, in the case where application 107 is a menopause-related application, variables may include, but are not limited to, age, number of symptoms, symptom type, symptom severity, ethnicity, income, geographic location, menopause cognition, etc. Each profile 116 corresponds to a different consumer and is generated by profile generator 120. Cluster 118 comprises at least one cluster, which includes a selected set of profiles 116 with similarity scores above a similarity threshold. Each cluster 118 is generated by cluster generator 122. Intervention 119 comprises approved interventions for treating various health outcomes, such as products, lifestyle modifications, or combinations thereof, as described below.

[0025] In some examples, such as where system 100 is a system for generating menopausal profiles and recommendations, data storage device 114 stores an index that associates menopausal symptoms with symptom ranking values ​​and resources for addressing those symptoms. Symptoms can be ranked according to various methods, including the most common symptoms observed during menopause, the most common symptoms observed during a specific menopausal phase, the likelihood that a consumer would seek treatment or relief for the symptoms, etc. In various examples, resources include, but are not limited to, articles, videos, products, licensed medical professionals, etc. In other examples, the index includes data associated with the co-occurrence of different menopausal symptoms. For example, the index may include data on co-occurring symptoms indicating various symptoms based on ethnicity, such as fatigue, insomnia, anxiety, depression, hot flashes, night sweats, joint pain, arthritis, headaches, migraines, decreased libido, cognitive changes (such as memory problems or decreased mental acuity), irregular heartbeat, weight gain, or changes in body odor. The data included in the index can be collected from various sources, including Examples 1 and 2 below.

[0026] Example 1

[0027] The first study aimed to investigate and identify differences in menopausal symptoms across different demographic groups, particularly by race and income level. This study focused on a range of symptoms, including vasomotor, sleep, and cognitive symptoms, as well as psychological and sexual health. The first study was conducted among 4,578 female participants aged 40 to 65 years in the United States (mean = 50.2, SD = 7.6). Participants completed an online survey over a three-week period, with an average completion time of 25 minutes per person. To ensure the validity of the response, participants were excluded if anyone in their close social network worked within the health care ecosystem. In total, the first study recruited Caucasian women (n=2936), African American women (n=665), Asian women (n=147), Native American women (n=41), Hispanic women (n=665), and women of other races (n=124). The study included 118 questions about health attitudes, treatment history, understanding of HRT, and suggested solutions. To identify symptom co-occurrence, Phi correlation analysis was performed on paired symptoms at the individual level for each ethnic group. One-way ANOVA was used to assess ethnic group differences in the number of co-occurring symptoms, with group means adjusted for sample size. The role of income in symptom co-occurrence was assessed by recording significant pairs for each income group and performing chi-square tests across income levels.

[0028] Example 1 - Results

[0029] The correlations in symptom occurrence indicated that women experienced multiple symptoms simultaneously: rs = 0.3–0.6, ps < 0.05–0.005. Univariate ANOVA revealed a significant effect of race on the number of significantly co-occurring symptoms (F = 13.0, p < 0.001). Paired comparisons showed that Native American groups reported significantly more co-occurring symptoms than all other racial groups: ps < 0.05. The results are shown in Table 1 below.

[0030]

[0031] Table 1 - Number of co-occurring symptoms by race

[0032] Surprisingly, women from different racial groups experienced both common and unique menopausal symptoms. For example, hot flashes were accompanied by night sweats and mood swings in all racial groups. Furthermore, African American women experienced insomnia, white women had irritability, decreased libido, insomnia, and cognitive problems, Asian women exhibited vaginal dryness, Hispanic women noted changes in body odor, and Native Americans had up to 25 symptoms, including headaches. A chi-square test revealed a significant effect of income on the number of co-occurring symptoms: X²(3,4) = 59.2, p < 0.001. Tukey HSD pairwise analysis showed that co-occurring symptoms decreased significantly in the following order as income level increased: below $35k, $35k-$75k, $75k-$150k, and above $150k (ps < 0.05 ~ 0.001).

[0033] Therefore, these findings are consistent with the SWAN study's findings that women experience a variety of symptoms associated with menopause. These findings also demonstrate that race significantly influences symptom experience, providing information for clinical practice. These findings can aid in menopause diagnosis through symptom screening and guide healthcare providers in tailoring their consultations. Furthermore, this study highlights the importance of utilizing genetic testing to better understand racial patterns and genetic variability, and to facilitate targeted treatment for different female groups.

[0034] Example 2

[0035] A second study was conducted to provide further insights into women's perceptions and experiences with MHT, focusing on potential age-related disparities, as the topic of MHT remains controversial and polarized among patients. Previous surveys conducted through women's health initiatives have demonstrated a link between MHT and an increased risk of coronary heart disease and breast cancer. Consequently, MHT use and new prescriptions have decreased significantly. The second study was conducted in the United States with 4,578 female participants aged 40 to 65 years (mean = 50.2, SD = 7.6). Participants completed an online survey over a three-week period, with an average completion time of 25 minutes per participant. To ensure the validity of the response, participants were excluded if anyone in their close social network worked in the healthcare ecosystem. In total, the second study recruited Caucasian women (n=2936), African American women (n=665), Asian women (n=147), Native American women (n=41), Hispanic women (n=665), and women of other races (n=124). The study included 118 questions about health attitudes, treatment history, understanding of MHT, and suggested solutions. The current study assessed differences in women's self-reported understanding and attitudes towards MHT by age group.

[0036] To determine the role of MHT in informing women's attitudes toward MHT, Spearman correlation analysis was performed on paired symptoms at the individual level within each ethnic group. To examine age differences, women were categorized into four age groups: 40–45 years, 45–50 years, 50–60 years, and 60–65 years. The Kruskal-Wallis test was used to compare differences in attitudes toward MHT across the different age groups. A post-hoc Dunn test (FDR corrected) was then performed for paired comparisons to examine age-group differences in each attitude toward MHT.

[0037] Example 2 - Results

[0038] The correlation results showed a significant association between greater understanding of MHT and emotional responses to MHT (ps < 0.001). Higher levels of MHT understanding were associated with more positive feelings and fewer negative feelings. Survey responses reflecting better understanding of MHT were significantly associated with more positive attitudes toward MHT, including optimism about the helpfulness of MHT and relief upon starting MHT (rs = 0.2 ~ 0.25, ps < 0.001). Conversely, enhanced understanding of MHT was significantly associated with reduced negative attitudes, as demonstrated by survey responses reflecting feelings of inadequacy in coping with menopause independently, uncertainty about MHT, and reluctance to discuss MHT with family and friends (rs = -0.2 ~ -0.24, ps < 0.001). Regarding age-related differences, the Kruskal-Wallis test revealed significant differences in certain attitudes within the groups, with older women aged 55-65 exhibiting stronger positive attitudes toward MHT compared to younger women aged 40-50 (Hs = 15.7–256.0, ps < 0.01–0.001). In the top three survey attitude responses that showed substantial age-related differences, post-hoc analysis by Dunn Cheng indicated significant age differences across all four age groups, with the exception of two pairs: 40-45 versus 45-50 and 55-60 versus 60-65. These attitudes included feelings of inadequacy in coping with menopause independently, reluctance to discuss MHT with family and friends, and uncertainty about MHT. Older women were also more willing to consider MHT and felt a responsibility to share it with more women.

[0039] Therefore, the second study showed that older women aged 55 to 65 had a better understanding of MHT, exhibited more positive attitudes towards MHT, and were more receptive to its use compared to younger women aged 40 to 50. These findings highlight a critical opportunity to educate the next generation of women undergoing menopause. They also provide valuable insights for clinicians to better understand potential perceptions of MHT across different age groups and to foster proactive discussions about MHT.

[0040] Data storage device 114 may include one or more different types of data storage devices, such as, for example, one or more spinning disk drives, one or more solid-state drives (SSDs), and / or any other type of data storage device. In some non-limiting examples, data storage device 114 includes a redundant array of independent disks (RAID). In other examples, data storage device 114 includes a database. In this example, data storage device 114 is included within, attached to, inserted into, or otherwise associated with computing device 102. In other examples, data storage device 114 includes a remote data storage device accessed by computing device 102 via network 138, such as a remote data storage device, a data storage device in a remote data center, or a cloud storage device.

[0041] The computing device 102 further includes a profile generator 120. In some examples, the profile generator 120 is an example of a dedicated processor or processing unit implemented on the processor 108. The profile generator 120 generates a profile for a consumer based on data received from the consumer, such as data input on an external device 140 and transmitted to the computing device 102. For example, the profile generator 120 receives raw data from the consumer in various data formats (i.e., text format, number format, etc.) and converts or transforms the received data into a standardized data format. The standardized data is stored as a profile 116 in the data storage device 114.

[0042] The computing device 102 further includes a cluster generator 122. In some examples, the cluster generator 122 is an example of a dedicated processor or processing unit implemented on the processor 108. The cluster generator 122 generates clusters comprising at least two profiles. To generate a cluster, the cluster generator 122 determines the similarity between a particular profile and each additional profile. In some examples, a profile is added to a cluster when the similarity threshold it has with other profiles included in the particular cluster reaches or exceeds a similarity threshold. The similarity threshold may be a threshold percentage of identical or similar data values ​​of the profiles, a threshold number of identical data values, one or more identical specific values, or any other suitable threshold used to determine similarity. In some examples, the cluster generator 122 generates a person or phenotype associated with each generated cluster. The generated person includes an artificial profile representing the data values ​​of the profiles in the cluster.

[0043] The computing device 102 further includes a recommendation engine 124. In some examples, the recommendation engine 124 is an example of a dedicated processor or processing unit implemented on the processor 108. The recommendation engine 124 generates one or more recommendations based on consumer profiles and / or the cluster to which the consumer profiles belong. The recommendation engine 124 generates recommendations based on the type of system 100 to which the recommendation engine 124 belongs. For example, in the case where system 100 is a system for generating menopausal profiles and recommendations, the recommendation engine 124 generates recommendations for addressing menopausal symptoms.

[0044] Recommendation engine 124 includes a health outcome recognizer 126, an intervention recognizer 128, a recommendation generator 130, and a feedback receiver 132. Health outcome recognizer 126 identifies health outcomes associated with a specific cluster. In some examples, health outcomes include conditions such as menopause and stages such as perimenopause, menopause, or postmenopausal. In some examples, the identified health outcomes are stored as values ​​for cluster 118 in data storage device 114.

[0045] Intervention identifyer 128 identifies specific interventions based on survey data, clinical data, etc., which are expected to mitigate or slow the identified health outcome when applied. Various examples of interventions include products, combinations of two or more products, lifestyle modifications, combinations of two or more lifestyle modifications, combinations of at least one product and at least one lifestyle modification, etc. In some examples, interventions are stored in data storage device 114 as interventions 119 cross-referenced with health outcomes. In some examples, interventions 119 are ranked according to expected outcomes. For example, interventions 119 with a higher probability of a positive outcome are ranked higher than interventions 119 with a lower probability of a positive outcome. In another example, interventions 119 with a greater potential impact are ranked higher than interventions with a lower potential impact.

[0046] Recommendation generator 130 generates recommendations for consumers associated with a specific profile. In an example where intervention identifyr 128 identifies a potential intervention, the generated recommendation includes the intervention 119 identified by intervention identifyr 128 and instructions for implementing the identified intervention 119. In an example where multiple potential interventions are identified, recommendation generator 130 generates recommendations that include either the highest-ranked intervention 119 or a set of highest-ranked interventions. In some examples, recommendation generator 130 selects one or more interventions 119 based on a weighted average of interaction with the intervention and likelihood of adherence to the intervention, menopausal stage, and an overall view of health status based on symptoms and symptom severity.

[0047] In some examples, the generated recommendations further include educational resources associated with specific life stages. These educational resources include content such as articles, videos, podcasts, etc., which include educational materials associated with specific sub-segments of the life stage, products that help treat symptoms of the life stage, and lifestyle adjustments that can help treat symptoms of the life stage.

[0048] In some examples, the intervention includes specific products and / or lifestyle modifications. The generated recommendations include instructions for implementing the intervention, such as the amount of product to be applied a set number of times per day or week. For example, the generated recommendations may include the use of two products (such as products for treating headaches and hot flashes), instructions for using these two products according to a specific protocol, and educational resources associated with the user's specific menopausal stage and specific symptoms. It should be understood that this example of a recommendation is presented for illustrative purposes only and should not be construed as limiting. Other examples of the generated recommendations are possible without departing from the scope of this disclosure. In some examples, recommendations for one or more specific products include links to websites where the recommended products are available for purchase by consumers.

[0049] Feedback receiver 132 receives feedback regarding the generated recommendations. In some examples, feedback is received via user interface device 110. In other examples, feedback is received via communication interface device 112 from an external device (such as user device 140) of the consumer implementing the generated recommendations. The received feedback includes one or more indications regarding the success, failure, and / or feasibility of the generated recommendations. In examples where the generated recommendations include one or more products for treating menopausal symptoms and associated instructions, the feedback may be that the generated recommendations produce positive results (e.g., reduced hot flashes, night sweats, or joint pain), negative results (e.g., no change in symptoms), or infeasibility (e.g., the generated recommendations are difficult to implement due to complex instructions, or that following the instructions correctly takes too much time). In some examples, feedback is received by completing a user survey, where the consumer provides ratings for various elements of the recommendations, such as a scale of one to ten, one to five, etc. In other examples, feedback is received based on user data that maps whether the recommended product was purchased via a provided link, whether the recommended product was purchased at a later time, whether the recommended product was purchased multiple times, etc. In these examples, feedback receiver 132 analyzes the received feedback and provides the analyzed feedback to recommendation generator 130.

[0050] The recommendation generator 130 updates based on the received feedback. For example, positive feedback is used to strengthen the generated recommendations, while negative feedback or feedback that the recommendations are not feasible is implemented, such that future recommendations for consumers and / or other profiles in the same cluster as the consumer include different products, different product combinations, different instructions for implementing one or more products, etc.

[0051] In some examples, the received feedback is further used by profile generator 120 to update the profile. For example, received feedback indicating a reduction in menopausal symptoms automatically updates the consumer's profile to reflect the symptom reduction. In some examples, based on the updated profile, cluster generator 122 re-clusters the updated profile based on the updated information. For example, cluster generator 122 re-clusters the updated profile into new clusters that include other profiles with similar symptoms and levels. Based on the updated clusters of the profile, recommendation generator 130 generates updated recommendations for the consumer to reflect the updated profile and clusters, including but not limited to products, combinations of two or more products, lifestyle adjustments, combinations of two or more lifestyle adjustments, combinations of at least one product and at least one lifestyle adjustment, etc., wherein the products and lifestyle adjustments may be the same as or different from the originally recommended products or lifestyle adjustments.

[0052] In some examples, as described herein, system 100 is a system for generating recommendations to address menopausal symptoms. For instance, profile generator 120 generates a profile 116 for a consumer, which includes information related to the consumer's age, type of menopausal symptoms, number of menopausal symptoms, severity of menopausal symptoms, etc. Cluster generator 122 places the consumer in a cluster based on the similarity between the consumer's profile and other profiles in cluster 118. Recommendation engine 124 generates recommendations for the consumer based on cluster 118.

[0053] In this example, health outcome identifier 126 identifies a health outcome such as menopause based on age and symptom information provided by the consumer. Intervention identifier 128 identifies the highest-ranking symptom for the consumer or for individuals within cluster 118 based on rankings stored in data storage device 114. For example, as described herein, symptoms can be ranked as the most common symptom observed during menopause, the most common symptom observed during a specific menopausal phase, the likelihood that the consumer will seek treatment or relief for the symptom, etc. In one example, symptoms are ranked based on the likelihood that the consumer will seek treatment or relief for the symptom, and intervention identifier 128 identifies resources associated with each of the highest-ranking symptoms for individuals or more specifically, those identified by the consumer. For example, intervention identifier 128 can identify the highest-ranking symptom, the top three symptoms, the top five symptoms, etc., and identify resources associated with each identified symptom. As described herein, the identified resources may include one or more of the following: articles or videos that provide instruction and / or suggest treatment for specific symptoms, products that can be used to reduce or alleviate symptoms, lists of medical professionals available for medical appointments via in-person or telemedicine appointments, etc.

[0054] Recommendation generator 130 generates recommendations that include identified resources associated with the highest-ranking identified symptom. The generated recommendations include identified resources for each of the highest-ranking identified symptoms for the consumer. The recommendations are presented on user interface device 110 and / or transmitted to user device 140, where they are presented on user interface device 148. Feedback receiver 132 receives feedback on the generated recommendations, which is used by recommendation generator 130 to update its recommendations.

[0055] User equipment 140 is another example of a computing device that is separate from and external to computing device 102. In some examples, user equipment 140 includes a mobile computing device or any other portable device. Mobile computing devices include, for example, but not limited to, mobile phones, laptop computers, tablet computers, computing boards, netbooks, gaming devices, and / or portable media players. User equipment 140 may also include less portable devices such as servers, desktop personal computers, self-service terminals, or desktop devices. Additionally, user equipment 140 may represent a set of processing units or other computing devices.

[0056] In some examples, user equipment 140 includes at least one processor 146, memory 142 (which includes computed user-executable instructions 144), and user interface device 148. Processor 146 includes any number of processing units and is programmed to execute computer-executable instructions 144. The computer-executable instructions 144 are executed by processor 146, by a plurality of processors 146 within user equipment 140, or by a processor 146 external to user equipment 140. In some examples, processor 146 is programmed to execute computer-executable instructions 144, such as those described in the accompanying drawings (e.g., ...). Figure 8 The computer-executable instructions illustrated herein. In various examples, processor 146 is configured to execute application 154, which is a client version of application 107.

[0057] Memory 142 includes any number of media associated with or accessible by user device 140. In some examples, memory 142 is located inside user device 140. In other examples, memory 142 is located outside user device 140, or both inside and outside user device 140. For example, memory 142 may include both memory components located inside and outside user device 140. Memory 142 stores data, such as one or more applications 154. Applications 154 operate when executed by processor 146 to perform various functions on user device 140. Applications may communicate with peer applications or services, such as network services accessible via network 138. In the example, an application represents a downloaded client application corresponding to a server-side service executed in the cloud, such as a cloud server.

[0058] User interface device 148 includes a graphics card for displaying data to and receiving data from a user. User interface device 148 may also include computer-executable instructions (e.g., a driver) for operating the graphics card. Furthermore, user interface device 148 may include a display (e.g., a touchscreen display or a natural user interface) and / or computer-executable instructions (e.g., a driver) for operating the display. User interface device 148 may also include one or more of the following to provide data to or receive data from a user: a speaker, a sound card, a camera, a microphone, a vibration motor, one or more accelerometers, Bluetooth. ® The device includes a communication module, Global Positioning System (GPS) hardware, and a photosensitive light sensor. In a non-limiting example, a user inputs commands or manipulates data by moving the user equipment 140 in one or more ways.

[0059] In some examples, user interface device 148 is configured to launch application 154 and display a visualization of that application, such as Figures 7A to 7G As illustrated. For example, processor 146 can execute computer-executable instructions 144 stored in memory 142 to execute application program 154, and Figures 7A to 7G The visualization of the illustrated application 154 is presented via user interface device 148.

[0060] User equipment 140 further includes a communication interface device 150. The communication interface device 150 includes a network interface card and / or computer-executable instructions (such as a driver) for operating the network interface card. Communication between user equipment 140 and other devices (such as, but not limited to, computing device 102) can occur using any protocol or mechanism via any wired or wireless connection.

[0061] User equipment 140 further includes a data storage device 152 for storing data, such as, but not limited to, user-provided data associated with a consumer profile 116, the consumer being a user of user equipment 140. The data may be received via user interface device 148 and / or received, retrieved, or obtained from computing device 102.

[0062] Data storage device 152 may include one or more different types of data storage devices, such as, for example, one or more spinning disk drives, one or more solid-state drives (SSDs), and / or any other type of data storage device. In some non-limiting examples, data storage device 152 includes a redundant array of independent disks (RAID). In other examples, data storage device 152 includes a database. In this example, data storage device 152 is included within, attached to, inserted into, or otherwise associated with user equipment 140. In other examples, data storage device 152 includes a remote data storage device accessed by user equipment 140 via network 138, such as a remote data storage device, a data storage device in a remote data center, or a cloud storage device.

[0063] User device 140 further includes an image capture device 156. In some examples, the image capture device 156 is a camera operable to capture still images and / or videos of a consumer's skin. In some examples, the captured images and / or videos are used by recommendation engine 124 to enhance generated recommendations, as described herein. For example, the captured images and / or videos are included in a user profile 116 stored in data storage device 114 and used to provide objective details about the consumer's skin, compared to subjective details provided by the user, for example, in response to a questionnaire.

[0064] Figure 2 An example system for generating multiple characters is shown. Figure 2 The illustrated example systems are for illustrative purposes only and should not be construed as limiting. Various examples of system 200 may be implemented without departing from the scope of this disclosure. In various examples, system 200 is implemented by one or more elements of system 100, such as cluster generator 122.

[0065] System 200 includes cluster generator 202. In some examples, cluster generator 202 is... Figure 1 An example of cluster generator 122 is illustrated. Cluster generator 202 receives one or more factors 204 and one or more profiles 206 as input, and generates one or more clusters 208 based on combinations of factors 204 and profiles 206. Factors or variables 204 include different factors related to a specific condition such as menopause, including symptoms associated with different menopausal stages. For example, various factors may include age, ethnicity, health history, symptoms, medications, exercise programs, environmental factors, etc. Each profile 206 includes data related to each factor 204 for a specific consumer associated with the profile 206. For example, each profile 206 includes consumer data related to the consumer's age, ethnicity, historical symptoms, current symptoms, etc.

[0066] Cluster generator 202 implements a clustering algorithm to determine the similarity between a specific profile and each additional profile. In some examples, a profile is added to a cluster or a new cluster is generated when the similarity threshold between the profile and other profiles included in a specific cluster is met or exceeded. The similarity threshold can be a threshold percentage of identical or similar data values ​​in the profiles, a threshold number of identical data values, one or more identical specific values, or any other suitable threshold used to determine similarity.

[0067] In the example where cluster generator 202 initially generates clusters, it identifies one or more identifying features of individuals within the cluster. An individual is a human profile representing data values ​​from a profile within the cluster. For example, the individuals in the first cluster include women aged 50 to 54 who are in menopause. Cluster generator 202 identifies profiles that match various aspects of the individuals and then groups these profiles into one or more clusters based on additional factors. In this example, a first cluster 208a is generated for profiles of women aged 50 to 54 who are in menopause and have self-reported "mild" symptoms; a second cluster 208b is generated for profiles of women aged 50 to 54 who are in menopause and have self-reported "moderate" symptoms; and a third cluster 208n is generated for profiles of women aged 50 to 54 who are in menopause and have self-reported "severe" symptoms. First person 210 is associated with first cluster 208a, second person 212 is associated with second cluster 208b, and third person 214 is associated with third cluster 208n. It should be understood that this example is presented for illustrative purposes only. In various examples, various clusters are generated based on various different factors without departing from the scope of this disclosure.

[0068] Therefore, the architecture of clusters 208 and characters 210 to 214 provides an improved set of indexes stored in data storage device 114, which facilitates improved retrieval by one or more elements of recommendation engine 124. Improved retrieval of data associated with one or more clusters 208 and / or one or more characters 210 to 214 enables more efficient association of new profiles with existing clusters and more efficient re-association of existing profiles with different clusters based on newly received data. Data retrieval is improved by storing cluster 208 (i.e., cluster 118) together with associated characters 210 to 214 embedded in cluster 208 in data storage device 114 (which reduces the computational resources required to associate updated profiles with new clusters). Cluster 208 can be stored in data storage device 114 along with tags indicating similarity between different clusters 208, allowing cluster generator 202 to identify new clusters or sets of new clusters with which updated profiles have a high probability of being associated. This, in turn, enables the cluster generator 122 to learn over time how specific changes to profiles in a particular cluster 208 historically led to the profiles being associated with specific new clusters. This reduces the computational resources consumed in associating existing profiles with new clusters based on the new data being received.

[0069] In the example where cluster generator 202 receives a new or updated profile to include in an existing cluster, cluster generator 202 performs a clustering algorithm to compare the received profile with other profiles in different clusters. In some examples, the profile is added to the cluster with which the received profile has the highest similarity score. In other examples, the highest similarity score is compared to a similarity threshold to determine whether the similarity between the profile and the profile in the nearest cluster is sufficient to include the profile in that cluster. If the similarity score is equal to or greater than the similarity threshold, the profile is added to the cluster. If the similarity score is less than the similarity threshold, the profile is not similar enough to the profiles in the cluster to include the profile in that cluster, and cluster generator 202 generates a new cluster for that profile.

[0070] In some examples, given that the data may include a wide variety of features, some of which are highly correlated with a specific condition (such as menopause) while others are less correlated, the cluster generator 202 implements a two-stage process to strategically reduce the feature space of biometrics. The two-stage process includes i) hierarchical feature collapse and ii) selecting the highest-ranking features by performing principal component analysis (PCA). In hierarchical feature collapse, the cluster generator 202 groups features that are more similar to each other together, thus reflecting the strength of similarity between features along the vertical axis. The number of resulting feature clusters is determined based on where a threshold is cut along the vertical axis. In scenarios where there are multiple elements in a feature cluster, the feature with the largest variance is selected. In some examples, a more aggressive threshold is used to further collapse the feature space. In other examples, a more conservative threshold is used to eliminate only highly redundant features.

[0071] PCA is an unsupervised machine learning method used to partition the variability in data along non-overlapping axes composed of linear combinations of features, such that the first few axes explain most of the variability. The informativeness of PCA is increased by normalizing the data through centering and scaling. Otherwise, some features of higher rank might dominate the axes. In some examples, a first number of features (such as the top ten features from the first three principal components) are selected to identify the highest-ranking features.

[0072] In some examples, clusters generated by cluster generator 202 are generated and output as visualizations (such as heatmaps). To visualize the clusters, the biological dataset is scaled according to the maximum feature importance of each feature among the features in the first three principal components. This effectively weights the original data by the amount of variability explained by the features in the biological dataset and thus provides a better visual representation. In some examples, given the continuous nature of biological data, clusters are visualized as clouds of similar individuals rather than discrete clusters. In some examples, additional models are used to further cluster the data. For example, Partition Around Centroids (PAM) can be chosen, which is robust, stable, fast, and capable of handling different data types. Various performance metrics, such as the Silhouette index, the Calinski-Harabasz index, or any other suitable clustering performance metric, can be used to determine cluster size. Since the biology behind menopause is continuous, it is inappropriate to interpret these discrete clustering metrics literally. Instead, using them as guidelines for the range of k-values ​​can provide more informative clustering and analysis. The selection of the k value is based on a visual examination of the balance between the curves showing how these metrics change with the k value and the personalized marketing bandwidth for menopause. The resulting heatmap can then display the features by clusters, where color intensity corresponds to more extreme values, with the first color corresponding to higher values ​​and the second color corresponding to lower values.

[0073] In summary, the cluster generator 202 performs clustering by: performing hierarchical feature collapse to remove redundant features, selecting the highest-ranking PCA feature as the chosen feature, scaling the PCA visualization by the weight of the highest-ranking feature, applying additional models such as PAM to cluster the scaled data of the chosen features, selecting a value k by using various performance metrics such as the contour index, the Calinski-Harabasz index, or other marketing resource constraints, and re-clustering with the optimal k value and analyzing using cluster labels.

[0074] In some examples, certain biometrics (also referred to herein as included features) are included in the feature space utilized by cluster generator 202 because the healthcare provider or practitioner believes that certain biometrics are particularly relevant to a particular condition, or are removed from the feature space utilized by cluster generator 202 because the healthcare provider or practitioner believes that certain biometrics are not particularly relevant to a particular condition. In some examples, the biometric model includes unweighted features to emphasize one or more specific features.

[0075] In some examples, cluster generator 202 further clusters behaviors identified as specifically related to a particular condition in question (such as menopause). To this end, cluster generator 202 i) correlates biometrics and behavioral features, ii) for each mapped biometric, identifies behavioral features mapped to biometrics above a set threshold, iii) uses the behavior with the highest mapping degree for the cluster, and iv) applies the clustering process described herein to various factors, except for behavioral feature collapse, since feature filtering has already been performed. For example, cluster generator 202 correlates biometrics and behavioral features and identifies which behavioral features have the highest correlation with biometrics, which are then used as plausible proxies for potential menopausal symptoms. The heatmap of the correlation between biometrics and behavioral features indicates that behavioral features are meaningfully clustered biologically. In other words, there is clear grouping of behavioral issues / topics, which provides positive feedback to the clustering approach. Although the absolute correlation between biometrics and behavioral features may be relatively low in some examples, this signal is meaningful because real-world correlations (especially with humans) tend to be much weaker and more difficult to measure than laboratory-controlled experiments.

[0076] In some examples, behavioral models are combined with biological models to develop a single model that models the dependencies between biological and behavioral features. In this example, a lighter hierarchical feature collapse is performed because the feature space expands significantly with feature union. Using this reduced feature set, the same set of clustering processes as discussed in this paper is performed, except for the PCA feature selection phase, which is performed separately in each model.

[0077] Figure 3 An example of a recommendation engine for generating interventions for a specific profile is shown. Figure 3 The example recommendation engine 124 illustrated is for illustrative purposes only and should not be construed as limiting. Other examples of recommendation engine 124 may be used without departing from the scope of this disclosure.

[0078] As described in this article, recommendation engine 124 generates one or more recommendations based on consumer profiles and / or the clusters to which the consumer profiles belong. Figure 3Examples of inputs received by the recommendation engine 124, in addition to factor 204 for a specific profile 206, are illustrated. These factors are used to generate one or more recommended interventions for a specific profile. Examples of various inputs received and used by the recommendation engine 124 include, but are not limited to, user goals, user engagement, and user motivation. User goals are examples of things a user wants to accomplish. For example, user goals could include receiving as much instruction as possible, receiving a set amount of instruction, alleviating symptoms, etc. User engagement refers to the user's historical and / or expected interaction with the system 100, such as the likelihood that the consumer will implement the generated recommendations, provide feedback on the implemented recommendations, etc. User motivation refers to the reasons why a consumer initially interacted with the system 100 and / or continues to interact with the system. User motivation stems from an individual's psychological profile. For example, a goal may be provided, but depending on the user's motivation, the goal may or may not be achieved. With an understanding of the user's underlying motivations, different products can be recommended based on the user's capabilities or expected capabilities.

[0079] Based on the received input, recommendation engine 124 generates recommendations that include at least one intervention, such as intervention A, intervention B, intervention C, intervention D, etc., which addresses the consumer's expected health outcomes when implemented. As described herein, interventions include products, combinations of two or more products, lifestyle modifications, combinations of two or more lifestyle modifications, combinations of at least one product and at least one lifestyle modification, etc.

[0080] Figure 4 An example timeline showing how the profile is updated over time is shown. Figure 4 The example timeline 400 illustrated is provided for illustrative purposes only and should not be construed as limiting. Other examples of timeline 400 may be used without departing from the scope of this disclosure.

[0081] The timeline of 400 examples illustrates health trajectories, user goals, engagement and motivation, rankings, and combinations of models over time. As time progresses, consumer goals, engagement and motivation, and rankings may change. Furthermore, the cluster containing consumer profiles can change based on factors such as the consumer's changing age, updates to the consumer's profile (e.g., symptom increases or decreases over time), and whether one or more recommended interventions have been implemented.

[0082] like Figure 4As shown, at the initial time T = 0, the consumer has a ranking C and implements a specific recommendation model ABC to generate recommendations. At the second time T + 1, as the consumer's goal changes, the consumer has an updated ranking B, but still implements recommendation model ABC to generate recommendations. Additional changes are shown at additional times T = 2 and T = Xi.

[0083] Figure 5 Examples of computer-implemented methods for generating one or more recommended templates are illustrated. The computer-implemented method 500 is presented for illustrative purposes only and should not be construed as limiting. Other examples of the computer-implemented method 500 may be used without departing from the scope of this disclosure. The computer-implemented method 500 may be implemented by one or more electronic devices (such as computing device 102) described herein.

[0084] The computer-implemented method 500 begins in operation 502 with profile generator 120 generating multiple profiles 116. As described herein, profile generator 120 generates a separate profile 116 for each corresponding consumer based on data received from consumers, such as data input on external device 140 and transmitted to computing device 102. Each newly generated profile 116 is stored in data storage device 114.

[0085] In operation 504, cluster generator 122 identifies multiple generated profiles 116. In some examples, cluster generator 122 automatically identifies multiple generated profiles 116 when generating profiles. In other examples, cluster generator 122 identifies multiple generated profiles 116 in response to triggering events (such as the time of day). For example, cluster generator 122 identifies multiple generated profiles 116 at regular intervals (such as the same time of day, once a week, once every two weeks, etc.).

[0086] In operation 506, cluster generator 122 identifies the value of each variable in each profile 116. As described herein, consumer variables include, but are not limited to, age, ethnicity, income, geographic location, menopause awareness, and various menopause-related symptoms. Values ​​are the values ​​of specific variables. For example, the value of the age variable includes the consumer's numerical age.

[0087] In operation 508, cluster generator 122 determines that the two profiles 116 are similar. In some examples, cluster generator 202 implements a clustering algorithm to determine the similarity between a particular profile and each supplementary profile. For example, cluster generator 202 identifies one or more identifying features of the people in the cluster. Then, cluster generator 202 identifies profiles that match aspects of the people and then groups these profiles into one or more clusters based on supplementary factors. In operation 510, a cluster 118 comprising the two determined profiles 116 is generated. The generated cluster 118 is stored in data storage device 114.

[0088] In operation 512, profile generator 120 generates a new profile for the new user. For example, profile generator 120 generates a new profile for the consumer based on information indicating personal variables received from the consumer. The newly generated profile 116 is stored in data storage device 114.

[0089] In operation 514, cluster generator 122 associates the newly generated profile 116 with existing cluster 118. For example, cluster generator 202 performs a clustering algorithm to compare the received profile with other profiles in different clusters. In some examples, the profile is added to the cluster with which the received profile has the highest similarity score. In other examples, the highest similarity score is compared to a similarity threshold to determine whether the similarity between the profile and the profile in the nearest cluster is sufficient to include the profile in that cluster. If the similarity score is equal to or greater than the similarity threshold, the profile is added to the cluster. If the similarity score is less than the similarity threshold, the profile is not similar enough to the profiles in the cluster to include the profile in that cluster, and cluster generator 202 generates a new cluster for that profile. Cluster 118 in data storage device 114 is updated to include the newly generated profile 116.

[0090] In operation 516, health outcome recognizer 126 identifies health outcomes associated with cluster 118, which includes newly generated profile 116. In some examples, health outcomes include the user's menopausal stage associated with the profile based on the received information. In some examples, the identified health outcomes are stored as values ​​for cluster 118 in data storage device 114.

[0091] In operation 518, intervention identifier 128 identifies interventions that are likely to address the identified health outcome. For example, intervention identifier 128 identifies specific interventions based on survey data, clinical data, etc., that, when applied, are expected to reduce or alleviate symptoms associated with the identified menopausal stage and the most severe symptoms identified by the user. Various examples of interventions include products, combinations of two or more products, lifestyle modifications, combinations of two or more lifestyle modifications, combinations of at least one product and at least one lifestyle modification, etc. In some examples, the identified intervention includes teaching content associated with at least one of the identified menopausal stage or identified user symptoms. In some examples, the intervention is stored in data storage device 114 as an intervention 119 cross-referenced with the health outcome.

[0092] In operation 520, recommendation generator 130 generates a recommendation for profile 116 that includes the identified intervention 119. The generated recommendation includes the intervention 119 identified by intervention identifier 128 and instructions for implementing the identified intervention 119. In some examples, the intervention includes specific products, teaching content, and lifestyle modifications. The generated recommendation includes instructions for implementing the intervention.

[0093] In operation 522, feedback receiver 132 determines whether feedback has been received regarding the generated and presented recommendation. The received feedback may include one or more indications of the success, failure, and / or feasibility of the generated recommendation, and / or the value of the teaching content conveyed. In some examples, feedback is received by completing a user survey, where consumers provide ratings for various elements of the recommendation, such as one to ten, one to five, etc. In other examples, feedback is received based on user data mapping whether the recommended product was purchased via a provided link, whether the recommended product was purchased later, whether the recommended product was purchased multiple times, etc. In examples where no feedback is received, the computer-implemented method 500 terminates.

[0094] In the example where feedback is received, in operation 524, cluster generator 122 analyzes the received feedback and determines whether changes to profile 116 result in profile 116 remaining in cluster 118. For example, cluster generator 122 repeats the process described in operation 514 and then determines whether profile 116 is still most similar to the initial cluster 118 or most similar to the new cluster 118. In the example where changes to profile 116 do not result in a sufficiently significant change to move the profile to a different cluster—that is, profile 116 remains in the same cluster 118—computer-implemented method 500 terminates. In the example where changes to profile 116 do result in a sufficiently significant change to move the profile to a different cluster, computer-implemented method 500 proceeds to operation 526, and cluster generator 122 associates the profile with the new cluster 118. Then, in operation 528, recommendation generator 130 generates updated recommendations for profile 116 based on the association of profile 116 with the new cluster 118. Updated recommendations are presented to the consumer via user device 140, and computer-implemented method 500 returns to operation 522 to determine whether feedback regarding the updated recommendations has been received.

[0095] Figure 6 Examples of computer-implemented methods for generating one or more recommended templates are illustrated. The computer-implemented method 600 is presented for illustrative purposes only and should not be construed as limiting. Other examples of the computer-implemented method 600 may be used without departing from the scope of this disclosure. The computer-implemented method 600 may be implemented by one or more electronic devices (such as computing device 102) described herein.

[0096] The computer-implemented method 600 begins in operation 602 by detecting user navigation on application 107 or application 154, respectively, on user interface device 110 or user interface device 148. The user navigation on application 107 or application 154 may be a mobile application, a web-based application, or any other suitable type of hosted application. In some examples, the detected user navigation includes detecting input to user interface device 110 or user interface device 148, responding to prompts presented on user interface device 110 or user interface device 148, etc.

[0097] In operation 604, profile generator 120 determines whether the user who navigated to the site in operation 602 is an existing user. In some examples, profile generator 120 determines whether a user is an existing user based on data obtained from the user's device being used to navigate to the site. For example, application 107 or application 154 may determine whether a user is logged into application 107 or application 154, capture Internet Protocol (IP) data from the user's device, and compare the IP data with the IP data of an existing user or any other suitable data to determine whether the user is an existing user. In an example where the user is not determined to be an existing user, method 600 proceeds to operation 606, where profile generator 120 generates a new profile for the new user. In an example where the user is not determined to be an existing user, method 600 proceeds to operation 608, where profile generator 120 retrieves a user profile. For example, the user profile and associated data are stored as data 116 in data storage device 114, and profile generator 120 retrieves the user profile and associated data.

[0098] In operation 610, after generating a user profile in operation 606 or retrieving a user profile in operation 608, application 107 or application 154 captures user navigation and input context via user interface device 110 or user interface device 148, respectively. The user navigation and input context captured by application 107 or application 154 includes, but is not limited to, items selected by the user, items viewed by the user, and data entered by the user into application 107 or application 154.

[0099] In operation 612, the recommendation engine 124 performs one or more processing actions. In some examples, this includes one or more of the following: a health outcome recognizer 126 identifies or predicts a user's menopausal stage based on received input, an intervention recognizer 128 identifies interventions, and a recommendation generator 130 generates recommendations for the user. In some examples, concurrent processing is performed for current needs (i.e., recommendations to be presented to the user while the user is browsing the site) and for future needs (i.e., improving future recommendations for the user or other users).

[0100] In operation 614, recommendation engine 124 determines whether there is sufficient data available for recommendation generator 130 to generate a sufficiently robust recommendation that includes one or more identified interventions. In some examples, recommendation engine 124 requires a threshold amount of data to be available in order to generate recommendations that are expected to be accurate for the user. If insufficient data is available, in operation 614, recommendation engine 124 determines that there is not enough data available, identifies what additional data would be beneficial for the current recommendation and future simulations, and method 600 returns to operation 610 to capture additional user guidance and input context in order to collect the required data. For example, insufficient data may be unavailable in cases such as in operation 604 where the user is not an existing user and the user has not yet provided enough input to collect sufficient data.

[0101] With sufficient data available, in operation 616, recommendation generator 130 analyzes available promotions and optimized consumer journeys to optimize data inputs within application 107 or application 154. For example, promotions can be presented when needed to incentivize engagement and drive additional user responses to prompts. Examples of available promotions and optimized consumer journeys include, but are not limited to, discounted purchases, free samples, early access to new products, etc. Optimizing data inputs includes including those that facilitate additional data for training one or more aspects of recommendation engine 124, additional data needed to generate improved recommendations, and additional data required to present the improved recommendations to the user, etc.

[0102] In operation 618, the recommendation generator 130 determines whether a promotion or incentive offer is available. If a promotion is identified and determined to be available, method 600 proceeds to operation 620, where user interface device 110 or user interface device 148 displays a prompt that selects the identified promotion when chosen by the user. If a promotion is not determined to be available, or after the prompt was displayed in operation 620, method 600 proceeds to operation 622, and user interface device 110 or user interface device 148 displays a prompt that responds to one or more data inputs when chosen by the user. For example, if no promotion is available, user interface device 110 or user interface device 148 may present a prompt that, when chosen, indicates whether the user is willing to share data.

[0103] In operation 624, recommendation generator 130 determines whether the user has provided data in response to a prompt displayed on user interface device 110 or user interface device 148. If data has been provided, method 600 proceeds to operation 626, and recommendation generator 130 determines whether the provided data indicates that the user has chosen to continue and accept the recommendation. If the user has chosen to continue and allowed the use of their data, method 600 returns to operation 612, where recommendation engine 124 continues processing, now including the user data. If the user has not chosen to continue and allowed the use of their data in operation 626, or if no data was provided in operation 624, then in operation 628, recommendation generator 130 generates recommendations for the user as described herein. Then, in operation 630, the generated recommendations are output and displayed on user interface device 110 or user interface device 148. After outputting the generated recommendations, method 600 returns to operation 612 and continues processing.

[0104] like Figure 6 The illustrated computer-implemented method 600 illustrates that achieving meaningful recommendations / guidance requires sufficient data quality and quantity. In some examples, the dataset needed to generate actions that can drive the expected results may not yet exist. The computer-implemented method 600 operates to create feedback loops, in which one or more basic models are implemented in the case of proactive consumer interaction via simple prompts to the consumer. In some examples, promotions are used to incentivize engagement when needed. The computer-implemented method 600 considers the data elements that have the highest impact on accuracy and prioritizes those data elements to strengthen data collection in those areas that a) have the greatest impact on the model and b) produce the most accurate results for the individual consumer. In cases where no input is provided or no data exists to drive the optimal results of the model, a default profile is used based on dataset limitations. Many interactions of this type rely on the consumer's initial follow-up interaction with the digital platform. This approach proactively interacts with the consumer via very few and quick prompts based on events within that progression during a natural navigation of the site. The combination of these two approaches expands the scope of interaction and lowers the barrier to interaction.

[0105] Furthermore, if prior consent is obtained and available, this method also allows for the collection of both anonymized and named consumer data, and has the ability to segment these two datasets. Traditionally, these types of methods rely on named consumer data, which creates an additional barrier to reaching the full target audience and increases the exposure and collection of important data to accelerate early discovery and patterns, which can then inform and drive different approaches to further research or analysis of these patterns. Therefore, this disclosure recognizes and takes into account these challenges and provides segmented datasets, thereby improving the ability of one or more models to be trained and executed based on the captured data. In some examples, improvements in recommendations provided directly to users and predictions of future progression of symptoms or needs have significantly improved consumer satisfaction and increased consumer loyalty. The collected data informs models for individuals, resulting in more accurate results, but often also improves models for larger groups in a given space.

[0106] Figures 7A to 7G An example user interface (UI) is shown, illustrating the process of creating a profile, receiving data, and generating recommendations. Figures 7A to 7G The example UI shown is provided for illustrative purposes only and should not be construed as restrictive. Various examples may be used without departing from the scope of this disclosure. In some examples, Figures 7A to 7G The example UI shown is Figure 1 The example shown is of a user interface device 148 that presents a client-side application 154. In some examples, Figures 7A to 7G The example UI shown is that of application 154, which focuses on providing resources and support for women going through menopause.

[0107] Figure 7A An example of a first UI 701 is shown. The first UI 701 is an example of a login page for application 154 that introduces application 154 and provides information about application 154 to the consumer. For example, the first UI 701 includes information about a first step in which UI 701 receives one or more inputs specifying the consumer's goal in using application 154.

[0108] Figure 7B Example of second UI 702. Second UI 702 is an example of a part of the consumer's overall rating and illustrates selectable icons that provide a response to a portion of the questionnaire when selected. In particular, second UI 702 illustrates example selections for different age ranges of consumers, including "under 40 years old", "40 to 44 years old", "45 to 49 years old", etc.

[0109] Figure 7CThe third UI 703 is illustrated as another aspect of the questionnaire. For example, the third UI 703 illustrates sample questions from a questionnaire about a consumer's energy and sleep, where multiple selectable options represent different menopausal symptoms related to energy and sleep. Figure 7D Example of the fourth UI 704 is shown, which includes another aspect of the questionnaire, such as example questions about the consumer's hair, skin, and nails, with multiple selectable options representing different menopausal symptoms related to hair, skin, and nails. Although Figure 7C and Figure 7D Two examples of questions about menopausal symptoms are given, but various examples are possible. Various other questions may include, but are not limited to, questions about symptoms related to age, race, energy and sleep, mood and thoughts, hot flashes, sexual relationships, weight changes, digestion, bone health, headaches, muscle and joint pain, etc.

[0110] Figure 7E Example 705 illustrates a fifth UI, which includes a section for consumers to input identifying information such as their name and email address. The provided email address is used to provide recommendations to the consumer based on their responses to questions in a questionnaire, and to identify the consumer upon logging into application 154 to monitor and track the progression of symptoms over time in order to provide updated recommendations. Figure 7F Example 6, UI 706, includes a description of menopausal stages based on provided responses to questions in the questionnaire. In some examples, the determined stage is based on a provided age or age range. In other examples, the determined stage is based on responses to one or more questions in the questionnaire.

[0111] Figure 7G Example 7 illustrates UI 707, which presents recommendations based on the consumer's identified menopausal stage and the identified symptoms provided in their responses to questionnaire questions. In some examples, the generated recommendations are based on the highest-ranking identified symptoms, such as those the consumer considers most severe or those the consumer is most likely to seek treatment for or alleviate. The generated recommendations include one or more of the following: resources providing education and / or suggested treatment for specific symptoms (such as articles or videos), products that can be used to reduce or alleviate symptoms, and a list of medical professionals available for medical appointments via in-person or telemedicine appointments.

[0112] Example operating environment

[0113] Figure 8This is a block diagram of an example computing device 800 for implementing the aspects disclosed herein, and the example computing device is generally referred to as computing device 800. Computing device 800 is an example of a suitable computing environment and is not intended to impose any limitation on the scope or functionality of the examples disclosed herein. Computing device 800 should also not be construed as having any dependencies or requirements associated with any of the illustrated components / modules or combinations thereof. The examples disclosed herein can be described in the general context of computer code or machine-usable instructions, including computer-executable instructions (such as program components), executed by a computer or other machine (such as a personal data assistant or other handheld device). Generally, program components (including routines, programs, objects, parts, data structures, etc.) refer to code that performs a specific task or implements a specific abstract data type. The disclosed examples can be practiced in a variety of system configurations, including personal computers, laptops, smartphones, mobile tablets, handheld devices, consumer electronics, dedicated computing devices, etc. The disclosed examples can also be practiced in distributed computing environments when the task is performed by a remote processing device linked via a communication network.

[0114] Computing device 800 includes a bus 820 that directly or indirectly couples to the following devices: computer storage memory 802, one or more processors 808, one or more presentation units 810, I / O ports 814, I / O components 816, power supply 818, and network components 812. Although computing device 800 is depicted as a single device, multiple computing devices 800 can work together and share the depicted device resources. For example, memory 802 can be distributed across multiple devices, and processors 808 can be housed with different devices.

[0115] Bus 820 indicates that it can be one or more buses (such as an address bus, a data bus, or a combination thereof). Although for clarity... Figure 8 The individual boxes are shown with lines, but alternative representations can be used to depict the individual components. For example, in some examples, presentation components (such as display devices) are I / O components, and some examples of processors have their own memory. No distinction is made between categories such as "workstation," "server," "laptop," and "handheld device," because all of these are envisioned as... Figure 8Within the scope of this document and the reference to "computing device," memory 802 may take the form of the computer storage medium referenced below and is operable to provide storage for computer-readable instructions, data structures, program modules, and other data for computing device 800. In some examples, memory 802 stores one or more of an operating system, a general-purpose application platform, or other program modules and program data. Thus, memory 802 is capable of storing and accessing data 804 and instructions 806, which can be executed by processor 808 and configured to perform the various operations disclosed herein.

[0116] In some examples, memory 802 includes computer storage media in the form of volatile and / or non-volatile memory, removable or non-removable memory, a data disk in a virtual environment, or a combination thereof. Memory 802 may include any number of memories associated with or accessible by computing device 800. Memory 802 may be located within computing device 800 (e.g., Figure 8 The memory 802 may be located outside the computing device 800, or both. Examples of memory 802 include, but are not limited to, random access memory (RAM); read-only memory (ROM); electrically erasable programmable read-only memory (EEPROM); flash memory or other memory technologies; CD-ROM, digital universal disc (DVD) or other optical or holographic media; magnetic tape cassettes, magnetic tape, disk storage devices or other magnetic storage devices; memory wired to an analog computing device; or any other medium used to encode desired information and for access by the computing device 800. Alternatively or additionally, memory 802 may be distributed across multiple computing devices 800, for example, in a virtualized environment where instruction processing is performed on multiple computing devices 800. For the purposes of this disclosure, “computer storage medium,” “computer storage memory,” “memory,” and “memory device” are synonymous terms for computer storage memory 802, and none of these terms include a carrier wave or propagation signaling.

[0117] Processor 808 may include any number of processing units that read data from various entities, such as memory 802 or I / O components 816, and may include a CPU and / or a GPU. Specifically, processor 808 is programmed to execute computer-executable instructions for implementing aspects of this disclosure. These instructions may be executed by a processor, by multiple processors within computing device 800, or by a processor external to client computing device 800. In some examples, processor 808 is programmed to execute instructions such as those illustrated in the accompanying drawings. Furthermore, in some examples, processor 808 represents a specific implementation of analog technology for performing the operations described herein. For example, the operations may be performed by analog client computing device 800 and / or digital client computing device 800. Presentation component 810 presents data indications to a user or other device. Exemplary presentation components include display devices, speakers, printing components, vibrating components, etc. Those skilled in the art will understand and recognize that computer data can be presented in a variety of ways, such as visually in a graphical user interface (GUI), audibly through speakers, wirelessly between computing devices 800, via a wired connection, or otherwise. I / O port 814 allows computing device 800 to be logically coupled to other devices including I / O components 816, some of which may be built-in. Example I / O components 816 include, for example, but not limited to, microphones, joysticks, game controllers, satellite antennas, scanners, printers, wireless devices, etc.

[0118] Computing device 800 can operate in a networked environment via a network component 812 through a logical connection to one or more remote computers. In some examples, network component 812 includes a network interface card and / or computer-executable instructions (e.g., a driver) for operating the network interface card. Communication between computing device 800 and other devices can occur via any wired or wireless connection using any protocol or mechanism. In some examples, network component 812 is operable to transmit data via public, private, or hybrid (public and private) transport protocols, using short-range communication technologies (e.g., Near Field Communication (NFC), Bluetooth). ™ (e.g., brand communication) wirelessly transmits data between devices, or a combination thereof. Network component 812 communicates with cloud resource 824 across network 826 via wireless communication link 822 and / or wired communication link 822a. Various examples of communication links 822 and 822a include wireless connections, wired connections, and / or dedicated links, and in some examples, at least a portion is routed over the Internet.

[0119] Although described in conjunction with example computing device 800, the examples of this disclosure can be implemented with many other general-purpose or special-purpose computing system environments, configurations, or devices. Examples of well-known computing systems, environments, and / or configurations suitable for use with aspects of this disclosure include, but are not limited to, smartphones, mobile tablets, mobile computing devices, personal computers, server computers, handheld or laptop devices, multiprocessor systems, game consoles, microprocessor-based systems, set-top boxes, programmable consumer electronics, mobile phones, wearable or accessory form factor mobile computing and / or communication devices (e.g., watches, glasses, headsets, or headphones), network PCs, minicomputers, mainframes, distributed computing environments including any of the above systems or devices, virtual reality (VR) devices, augmented reality (AR) devices, mixed reality devices, holographic devices, and the like. Such systems or devices can accept input from a user in any manner, including from input devices (such as keyboards or pointer devices), via gesture input, proximity input (such as by hover), and / or via voice input.

[0120] Examples of this disclosure can be described in the general context of computer-executable instructions (such as program modules) that are executed by one or more computers or other devices as software, firmware, hardware, or a combination thereof. Computer-executable instructions can be organized into one or more computer-executable parts or modules. Typically, program modules include, but are not limited to, routines, programs, objects, components, and data structures that perform a particular task or implement a particular abstract data type. Aspects of this disclosure can be implemented with any number and organization of such parts or modules. For example, aspects of this disclosure are not limited to specific computer-executable instructions, or specific parts or modules illustrated in the figures and described herein. Other examples of this disclosure may include different computer-executable instructions or parts having more or fewer functions than those shown and described herein. In examples involving general-purpose computers, aspects of this disclosure transform a general-purpose computer into a special-purpose computing device when configured to execute the instructions described herein.

[0121] By way of example and not limitation, computer-readable media include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable memory implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, etc.). Computer storage media are tangible and mutually exclusive with communication media. Computer storage media are implemented in hardware and are non-transitory, i.e., they do not include carrier waves and propagating signals. Computer storage media used for the purposes of this disclosure are not signals themselves. Exemplary computer storage media include hard disks, flash drives, solid-state memory, phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, optical disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage devices, magnetic tape cassettes, magnetic tape, disk storage devices or other magnetic storage devices, or any other non-transfer medium that can be used to store information for access by a computing device. In contrast, communication media typically embody computer-readable instructions, data structures, program modules, etc., in the form of modulated data signals (such as carrier waves or other transmission mechanisms), and include any information transmission medium.

[0122] In some examples, a computer-implemented method includes: generating multiple clusters; generating a profile for a new user; associating the generated profile with the clusters in the multiple clusters; and generating recommendations for the user based on the associated clusters using a machine learning (ML) model.

[0123] In some examples, an apparatus includes: a user interface (UI); a memory; and a processor coupled to the memory, the processor being configured to: control the UI to present a questionnaire; receive responses to the questionnaire via the UI; generate a profile associated with the user based on the received responses to the questionnaire and a captured facial scan; associate the generated profile with clusters in a plurality of clusters; and execute a machine learning (ML) model to generate recommendations for the user based on the associated clusters.

[0124] In some examples, a computer-readable storage medium stores instructions that, when executed by a processor, cause the processor to generate multiple clusters, each of the multiple clusters including at least a menopausal stage and experienced menopausal symptoms; generate a profile for a new user, the generated profile including at least the identified user's menopausal stage and experienced menopausal symptoms; associate the generated profile with the clusters in the multiple clusters; and generate recommendations for the user based on the associated clusters using a machine learning (ML) model.

[0125] This article describes another example of how to generate recommendations for users.

[0126] The various examples also include one or more of the following: The generation of the multiple clusters further includes: identifying multiple existing profiles, each of which includes a set of variables; for each of these existing profiles, identifying the value of each variable in the set of variables; determining, using a clustering algorithm, that the similarity between a first existing profile and a second existing profile among the multiple existing profiles is higher than a similarity threshold; and generating the cluster including the first existing profile and the second existing profile using the clustering algorithm. Associating the generated profile with the clusters in the plurality of clusters further includes: identifying the set of variables in the generated profile; identifying the value of each variable in the set of variables in the generated profile; determining the cluster most similar to the generated profile among the plurality of clusters using the clustering algorithm based on the identified values ​​of each variable in the set of variables in the generated profile; and associating the generated profile with the determined clusters. The set of variables includes variables that are related to at least one of age, number of symptoms, type of symptoms, severity of symptoms, race, income, geographic location, or menopausal cognition; The process of generating the recommendation for the user further includes: identifying a health outcome associated with the cluster among the plurality of clusters; identifying an intervention that, when applied, could potentially resolve the identified health outcome; and generating the recommendation for the user using the ML model, the recommendation including the intervention. Wherein: the identified health outcome is menopausal stage; and the identified interventions include at least one of treatment for symptoms of menopausal stage or educational content associated with menopausal stage; The method also includes: receiving feedback indicating the result of the recommendation; and updating the ML model based on the received feedback; The method further includes: receiving updated information from the new user; associating the generated profile with a second cluster among the plurality of clusters based on the received updated information, the second cluster being different from the first cluster; and generating a second recommendation for the user based on the associated second cluster using the ML model. The process involves: identifying a health outcome associated with one of the plurality of clusters, the identified health outcome including the identified menopausal stage of the user; identifying an intervention associated with the associated cluster, which, when applied, could potentially resolve the identified health outcome, the identified intervention including at least one of treatment for symptoms of the menopausal stage or educational content associated with the menopausal stage; and generating a recommendation for the user using the ML model, the recommendation including the intervention.

[0127] The order in which operations are performed or carried out in the examples of this disclosure illustrated and described herein is not required and may be performed in different orders in various examples. For example, it is contemplated that a particular operation is performed or carried out before, simultaneously with, or after another operation within the scope of aspects of this disclosure. When introducing elements of aspects of this disclosure or examples thereof, the articles “a,” “an,” “the,” and “described” are intended to mean the presence of one or more of that element. The terms “comprising,” “including,” and “having” are intended to be inclusive and mean that additional elements may be present in addition to the listed elements. The term “exemplary” is intended to mean “an example of…”. The phrase “one or more of the following: A, B, and C” means “at least one A and / or at least one B and / or at least one C.”

[0128] Having described in detail various aspects of this disclosure, it will be apparent that modifications and variations are possible without departing from the scope of the aspects of this disclosure as defined in the appended claims. Since various changes can be made to the above structures, products, and methods without departing from the scope of each aspect of this disclosure, everything contained in the foregoing description or shown in the accompanying drawings should be interpreted as illustrative and not restrictive.

Claims

1. A computer-implemented method, the computer-implemented method comprising: Generate multiple clusters; Generate profiles for new users; The generated profile is associated with one of the multiple clusters; as well as Recommendations are generated for the user based on the associated clusters using machine learning (ML) models.

2. The computer-implemented method according to claim 1, wherein generating the plurality of clusters further comprises: Identify multiple existing profiles, each of which includes a set of variables; For each existing profile in the existing profile, identify the value of each variable in the variable set; as well as Clustering algorithms were used to determine that the first existing profile and the second existing profile among the plurality of existing profiles had a similarity higher than a similarity threshold. as well as The clustering algorithm generates the cluster that includes the first existing profile and the second existing profile.

3. The computer-implemented method of claim 2, wherein associating the generated profile with one of the plurality of clusters further comprises: Identify the set of variables in the generated profile; Identify the value of each variable in the set of variables in the generated profile; Based on the identified value of each variable in the variable set of the generated profile, the clustering algorithm determines the cluster among the plurality of clusters that is most similar to the generated profile. as well as The generated profile is associated with the identified cluster.

4. The computer-implemented method of claim 3, wherein the set of variables includes variables related to at least one of age, number of symptoms, type of symptoms, severity of symptoms, race, income, geographic location, or menopausal cognition.

5. The computer-implemented method of claim 1, wherein generating the recommendation for the user further comprises: Determine the health outcome associated with the cluster among the plurality of clusters; Identify interventions that, when applied, may resolve the identified health outcomes; as well as The recommendation is generated for the user using the ML model, and the recommendation includes the intervention.

6. The computer-implemented method according to claim 5, wherein: The identified health outcome is the menopausal stage; and The identified interventions include at least one of treatment for symptoms of the menopausal stage or educational content associated with the menopausal stage.

7. The computer-implemented method according to claim 1, further comprising: Receive feedback indicating the results of the recommendations; as well as The ML model is updated based on the received feedback.

8. The computer-implemented method according to claim 1, further comprising: Receive updated information from the new user; Based on the received updated information, the generated profile is associated with a second cluster among the plurality of clusters, the second cluster being different from the clusters; as well as The ML model generates a second recommendation for the user based on the associated second cluster.

9. An apparatus comprising: User interface (UI); Memory; and A processor, coupled to the memory, is configured to: Control the UI to present the questionnaire; Receive responses to the questionnaire via the UI; Generate a profile associated with the user based on the received responses to the questionnaire; The generated profile is associated with multiple clusters in the cluster; as well as Execute a machine learning (ML) model to generate recommendations for the user based on the associated clusters.

10. The apparatus of claim 9, wherein the processor is further configured to: Identify multiple existing profiles, each of which includes a set of variables; For each existing profile in the existing profile, identify the value of each variable in the variable set; as well as A clustering algorithm is performed to determine that the first existing profile and the second existing profile among the plurality of existing profiles have a similarity higher than a similarity threshold; as well as The clustering algorithm generates the cluster that includes the first existing profile and the second existing profile.

11. The apparatus of claim 10, wherein, in order to associate the generated profile with one of the plurality of clusters, the processor is further configured to: Identify the set of variables in the generated profile; Identify the value of each variable in the set of variables in the generated profile; Based on the identified values ​​of each variable in the variable set of the generated profile, the clustering algorithm is executed to determine the cluster among the plurality of clusters that is most similar to the generated profile; and The generated profile is associated with the identified cluster.

12. The apparatus of claim 11, wherein the set of variables includes variables related to at least one of age, number of symptoms, type of symptoms, severity of symptoms, race, income, geographic location, or menopausal cognition.

13. The apparatus of claim 9, wherein, in order to generate the recommendation for the user, the processor is further configured to: Determine the health outcome associated with the cluster among the plurality of clusters; Identify interventions that, when applied, may resolve the identified health outcomes; as well as The ML model is executed to generate the recommendation for the user, the recommendation including the intervention.

14. The apparatus according to claim 13, wherein: The identified health outcome is the menopausal stage; and The identified interventions include at least one of treatment for symptoms of the menopausal stage or educational content associated with the menopausal stage.

15. The apparatus of claim 9, wherein the processor is further configured to: Receive updated information from the new user; Based on the received updated information, the generated profile is associated with a second cluster among the plurality of clusters, the second cluster being different from the first cluster; and The ML model is executed to generate a second recommendation for the user based on the associated second cluster.

16. One or more non-transitory computer-readable media, said one or more non-transitory computer-readable media storing instructions, said instructions causing the processor, when executed by a processor, to: Multiple clusters are generated, and each of the generated clusters includes at least the menopausal stage and the menopausal symptoms experienced; Generate profiles for new users, including at least the identified menopausal stage and menopausal-related symptoms experienced by the user. The generated profile is associated with one of the multiple clusters; as well as Recommendations are generated for the user based on the associated clusters using machine learning (ML) models.

17. The one or more non-transitory computer-readable media of claim 16, wherein the one or more non-transitory computer-readable media further stores instructions for generating the recommendation for the user, the instructions causing the processor, when executed by the processor, to: Determine health outcomes associated with the clusters in the plurality of clusters, including the identified menopausal stage of the user; Identify interventions associated with the associated clusters, which, when applied, are likely to address the determined health outcomes. The identified interventions include at least one of treatment for symptoms of the menopausal phase or teaching content associated with the menopausal phase; and The recommendation is generated for the user using the ML model, and the recommendation includes the intervention.

18. The one or more non-transitory computer-readable media of claim 16, wherein the one or more non-transitory computer-readable media further stores instructions for generating the plurality of clusters, the instructions causing the processor, when executed by the processor, to: Identify multiple existing profiles, each of which includes a set of variables; For each existing profile in the existing profile, identify the value of each variable in the variable set; as well as Clustering algorithms were used to determine that the first existing profile and the second existing profile among the plurality of existing profiles had a similarity higher than a similarity threshold. as well as The clustering algorithm generates the cluster that includes the first existing profile and the second existing profile.

19. The one or more non-transitory computer-readable media of claim 18, wherein the one or more non-transitory computer-readable media further stores instructions for associating the generated profile with the clusters in the plurality of clusters, the instructions causing the processor, when executed by the processor, to: Identify the set of variables in the generated profile; Identify the value of each variable in the set of variables in the generated profile; Based on the identified values ​​of each variable in the variable set of the generated profile, the clustering algorithm determines the cluster among the plurality of clusters that is most similar to the generated profile; and The generated profile is associated with the identified cluster.

20. The one or more non-transitory computer-readable media of claim 16, wherein the one or more non-transitory computer-readable media further stores instructions that, when executed by the processor, cause the processor to: Receive updated information from the new user; Based on the received updated information, the generated profile is associated with a second cluster among the plurality of clusters, the second cluster being different from the clusters; as well as The ML model generates a second recommendation for the user based on the associated second cluster.

Images