Machine learning models for developing data and providing user interaction policies

Abstract

Systems, devices, and methods are provided for data collection and development and user interaction policies. In one embodiment, the method includes collecting context data for a first subset of a plurality of users. The method further includes generating a first set of context profiles for the first subset of the plurality of users based on the collected context data. Furthermore, the method includes training one or more imputation models to create context data for a second subset of the plurality of users. The method also includes generating the context data for the second subset of the plurality of users using the one or more imputation models. Furthermore, the method includes generating a second set of context profiles for the second subset of the plurality of users based on the generated context data for the second subset of the plurality of users.

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application claims priority to U.S. Provisional Patent Application No. 63 / 289,376, filed December 14, 2021, the disclosure of which is incorporated herein by reference.

[0002] FIELD OF THEINVENTION The present application relates generally to medical devices (eg, analyte sensors) and, more specifically, to systems, devices, and methods for data collection and development and user interaction policy provision. [Background technology]

[0003] Diabetes is a metabolic disease linked to the body's production or use of insulin, a hormone that allows the body to use glucose for energy or store it as fat.

[0004] When a human eats a meal containing carbohydrates, the food is processed by the digestive system, producing glucose in the blood. Blood glucose can be used for energy or stored as fat. The body normally maintains blood glucose levels in a range that provides enough energy to support bodily functions and avoids problems that can occur when glucose levels are too high or too low. Regulation of blood glucose levels depends on the production and use of insulin, which regulates the movement of blood glucose into cells.

[0005] When the body does not produce enough insulin or is unable to effectively use the insulin that is present, blood glucose levels can rise above the normal range. A condition in which blood glucose levels are higher than normal is called "hyperglycemia." Chronic hyperglycemia can lead to many health problems, including cardiovascular disease, cataracts and other eye diseases, nerve damage (neuropathy), and kidney damage. Hyperglycemia can also lead to acute problems, such as diabetic ketoacidosis (a condition in which the body becomes too acidic due to the presence of glucose in the blood and ketone bodies, which are produced when the body can no longer use glucose). A condition in which blood glucose levels are lower than normal is called "hypoglycemia." Severe hypoglycemia can lead to acute crises that can result in seizures or death.

[0006] Diabetics may receive insulin to manage blood glucose levels. Insulin may be received, for example, through a manual injection with a needle. Wearable insulin pumps may also be utilized to receive insulin. Diet and exercise also affect blood glucose levels.

[0007] Diabetes is sometimes referred to as "Type 1" and "Type 2". Type 1 diabetes patients are typically able to use insulin when it is present, but due to problems with the insulin-producing beta cells of the pancreas, the body is unable to produce sufficient amounts of insulin. Type 2 diabetes patients are able to produce some insulin, but due to reduced sensitivity to insulin, the patient is "insulin resistant". As a result, even though insulin is present in the body, the insulin is not used well by the patient's body and blood glucose levels are not effectively regulated.

[0008] For diabetic patients, monitoring blood glucose levels and regulating those levels within an acceptable range is important not only to mitigate long-term problems such as heart disease and vision loss, but also to avoid the effects of hyperglycemia and hypoglycemia. Because blood glucose levels are almost constantly changing over time and in response to daily events such as eating or exercise, maintaining those levels within an acceptable range can be difficult. Advances in medical technology have allowed the development of a variety of systems for monitoring blood glucose, including Continuous Glucose Monitoring (CGM) systems that measure and record glucose concentrations in virtually real time. CGM systems are an important tool for users of these systems to ensure that measured glucose levels are within an acceptable range.

[0009] This Background is provided to introduce a brief context for the Summary and Detailed Description that follow. This Background is not intended to aid in determining the scope of the claimed subject matter, nor is it intended to be considered as limiting the claimed subject matter to embodiments that solve any or all of the disadvantages or problems discussed above. Summary of the Invention [Means for solving the problem]

[0010] Various embodiments of the present systems, devices, and methods have several features, no single one of which is solely responsible for its desirable attributes. Without limiting the scope of the present embodiments, their more prominent features are set forth below. After reviewing this description, and particularly after reading the section entitled "Description of the Preferred Embodiments," one will understand how the features of the present embodiments provide the advantages described herein.

[0011] In a first aspect, a method includes dividing a plurality of users into an exploration subset of users and an utilization subset of users; randomly assigning at least one user interaction policy for each exploration subset of users; and determining at least one user interaction policy for each utilization subset of users utilizing one or more context models trained using context data corresponding to the utilization subset of users, wherein the context data corresponding to the utilization subset of users includes at least some of a first set of context profiles and a second set of context profiles.

[0012] In a second aspect, a method includes collecting context data for a first subset of a plurality of users; generating a first set of context profiles for the first subset of the plurality of users based on the collected context data; determining that context data for a second subset of the plurality of users is incomplete or unavailable; training one or more imputation models based on the context data of the first subset of the plurality of users to create context data for the second subset of the plurality of users; generating the context data for the second subset of the plurality of users using the one or more imputation models; and generating a second set of context profiles for the second subset of the plurality of users based on the generated context data for the second subset of the plurality of users.

[0013] Also described herein are embodiments of non-transitory computer readable media that include instructions that execute on a computer system, which when executed on the computer system perform the methods described above.

[0014] Also described herein are computer system embodiments, the software of which is programmed to carry out the methods described above.

[0015] Also described herein are embodiments of computer systems comprising means for carrying out the above-described methods. [Brief description of the drawings]

[0016] [Figure 1] 1 illustrates an environment in an exemplary implementation operable to employ the techniques described herein, in accordance with certain embodiments of the present disclosure. [Diagram 2] 2 illustrates an example analyte monitoring device of FIG. 1 in greater detail, according to one embodiment of the present disclosure. [Diagram 3] 1 is a flow diagram illustrating a process for determining a user interaction policy based on psychographics according to an embodiment of the present disclosure. [Figure 4] 4 is an exemplary operational process for conducting the psychographic data collection and development and exploration-exploitation stages of the process of FIG. 3 according to an embodiment of the present disclosure. [Diagram 5] 1 illustrates an example of a displayed User Interface ("UI") according to one embodiment of the present disclosure. [Figure 6] 13 illustrates another example of a displayed Achievement UI according to an embodiment of the present disclosure. [Figure 7] 13 illustrates another example of a displayed Achievement UI according to an embodiment of the present disclosure. [Figure 8] 1 illustrates an example of badges displayed via a user's Achievement UI and via an additional user's Achievement UI, according to an embodiment of the present disclosure. [Figure 9] 1 illustrates an example of a displayed family collaboration UI according to an embodiment of the present disclosure. [Figure 10] 13 illustrates another example of a displayed family collaboration UI, according to an embodiment of the present disclosure. [Figure 11] 13 illustrates another example of a displayed family collaboration UI, according to an embodiment of the present disclosure. [Figure 12] 13 illustrates another example of a displayed family collaboration UI, according to an embodiment of the present disclosure. [Figure 13]1 illustrates an example of a process sharing UI and process displayed according to an embodiment of the present disclosure. [Figure 14] 13 illustrates another example of a displayed process sharing UI according to an embodiment of the present disclosure. [Figure 15] 13 illustrates another example of a displayed process sharing UI according to an embodiment of the present disclosure. [Figure 16] 13 shows an example of a displayed glucose intake UI according to an embodiment of the present disclosure. [Figure 17] 13 shows another example of a displayed glucose intake UI according to an embodiment of the present disclosure. [Figure 18] 13 shows another example of a displayed glucose intake UI according to an embodiment of the present disclosure. [Figure 19] 1 illustrates an example of a displayed follower highlight UI according to an embodiment of the present disclosure. [Figure 20] 13 illustrates another example of a displayed follower highlight UI according to an embodiment of the present disclosure. [Figure 21] 13 illustrates another example of a displayed follower highlight UI according to an embodiment of the present disclosure. [Figure 22] 1 illustrates an example of a displayed expert portal UI according to an embodiment of the present disclosure. [Figure 23] 13 illustrates another example of a displayed expert portal UI, according to an embodiment of the present disclosure. [Figure 24] 13 illustrates another example of a displayed expert portal UI, according to an embodiment of the present disclosure. [Diagram 25] 1 illustrates an example of a displayed personalization UI according to an embodiment of the present disclosure. [Figure 26] 1 illustrates an example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. [Figure 27] 13 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. [Figure 28] 13 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. [Figure 29]13 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. [Diagram 30] 13 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. [Diagram 31] 13 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. [Diagram 32] 1 illustrates an example of a displayed video story UI according to an embodiment of the present disclosure. [Diagram 33] 13 illustrates another example of a displayed video story UI according to an embodiment of the present disclosure. [Diagram 34] 13 illustrates another example of a displayed video story UI according to an embodiment of the present disclosure. [Diagram 35] 13 illustrates another example of a displayed video story UI according to an embodiment of the present disclosure. [Diagram 36] 1 illustrates an example of a displayed Health Hub UI according to an embodiment of the present disclosure. [Figure 37] 13 illustrates another example of a displayed health hub UI, according to an embodiment of the present disclosure. [Figure 38] 13 illustrates another example of a displayed health hub UI, according to an embodiment of the present disclosure. [Figure 39] An example of a system including various components of an exemplary device that can be implemented as any type of computing device described and / or utilized with reference to Figures 1-38 for implementing embodiments of the technology described herein is shown. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Portable and / or wearable health monitoring devices (also referred to herein as "health monitoring devices") and mobile health applications (also referred to herein as "applications") are rapidly becoming renowned for their ability to support user-centered care. For example, managing diabetes can pose complex challenges to patients, clinicians, and caregivers as a collection of many factors can affect a patient's glucose values ​​and glucose trends. To assist patients in better managing this condition, various health monitoring devices (e.g., sensors and other types of monitoring and diagnostic devices) and mobile health applications, such as, but not limited to, health and fitness monitoring applications that can assist in diabetes management for type 1 and type 2 diabetes patients, have been developed. The widespread adoption of such health monitoring devices and the increased development and distribution of mobile health applications have improved health management, and more specifically, chronic disease management, in the healthcare field. In particular, the use of such health monitoring devices in conjunction with mobile health applications represents a more scalable and potentially more cost-effective alternative to traditional interventions, providing a means to improve health and chronic disease management by expanding the reach of healthcare services and improving user access to health-related information and interventions.

[0018] Mobile health applications allow users to have greater involvement in their medical care by granting them access to and control over their health information. In particular, mobile health applications allow users to access, monitor, record, and update their health information regardless of physical constraints such as time and location. In particular, various intervention applications have been developed to provide guidance that may assist patients, caregivers, healthcare providers, or other users in improving their lifestyle or clinical / patient outcomes by addressing various issues such as analyte management, exercise, and / or other health factors. For example, diabetes intervention applications may assist patients, caregivers, healthcare providers, or other users in overnight glucose management (e.g., reducing the incidence of hypoglycemic events or hyperglycemic deviations), mealtime and post-meal glucose management (e.g., using historical information and trends to improve glycemic control), hyperglycemic correction (e.g., increasing time in target zone while avoiding hypoglycemic events due to overcorrection), and / or hypoglycemic treatment (e.g., addressing hypoglycemia while avoiding "rebound" hyperglycemia), to name a few.

[0019] Many mobile health applications designed to support the management of chronic diseases or health conditions have suffered from low user participation and high user attrition rates. Reasons for low user participation and / or high user attrition rates may include the inability of the mobile health application to provide users with individualized or personalized user interaction policies (as further defined below). If the mobile health application is unable to provide individualized and / or personalized user interaction policies, users of the application may perceive their interactions as ineffective in implementing a holistic approach to managing their health (e.g., disease, condition, health condition, etc.). Furthermore, user interaction policies that are not tailored to an individual (i.e., not "individualized") may result in suboptimal health outcomes. In addition, user interaction policies that are not tailored to a cohort of which an individual is a member (i.e., not "personalized") may also result in suboptimal health outcomes. Furthermore, user interaction policies that are not individualized and / or personalized may result in user interaction policies that are not relevant to the user. Thus, user participation associated with such mobile health applications may decrease, and thus user attrition rates may increase.

[0020] Correspondingly, embodiments described herein are directed to a health monitoring platform that uses a contextual multi-armed bandit (MAB) model to provide users with individualized or personalized user interaction policies based on, for example, corresponding contextual data of the users. A user interaction policy refers to various ways or means by which a software application interacts with a user. For example, a user interaction policy may include recommendations (e.g., to take more steps, get more sleep, possible interest in intermittent fasting, etc.), insights, content, information, etc., provided to the user. A user interaction policy may also include delivery methods (e.g., video, audio, text, phone, email, push notification, in-app message, etc.) for delivering such recommendations, insights, content, information, etc., to the user. A user interaction policy may further include, for example, UI configurations associated with different layouts of the software application used by the user. The objective of the software application is to individualize or personalize the user interaction policy for each user by selecting and providing to each user a user interaction policy that maximizes the likelihood of user participation by the particular user and / or achievement of the particular user's goals.

[0021] To achieve this level of individualization and / or personalization of user interaction policies, the health monitoring platform may utilize a contextual MAB model that receives as input contextual data corresponding to users and outputs individualized and / or personalized user interaction policies for each user. The user's contextual data may include the user's demographic information (e.g., age, gender, ethnicity, etc.), physiological information (e.g., analyte information generated by one or more analyte sensors), non-analyte health information (e.g., heart rate, body temperature, or other data generated by non-analyte sensors), illness information, and any other health-related information, as well as any other relevant user-specific information.

[0022] The user's contextual data also includes psychographic data. As used herein, psychographic data may include various data related to goals, interests, values, attitudes, personality traits, behavior-related data, and the like. For example, user goals may include sleep goals, exercise goals, diet goals, glucose management goals, and the like. User interests may include interest in a type of exercise, a type of activity, and the like. Having access to the user's psychographic data plays an important role in providing the user with an individualized or personalized user interaction policy that maximizes user participation. If the user's goals are unknown, the user interaction policy (e.g., recommendations) provided to the user may not be beneficial in helping the user achieve their goal. For example, if a diabetic user's goal of getting continuous sleep is not known by the health monitoring platform, it may not be possible to provide recommendations that help the user achieve their goal. Similarly, if the user's interest in running (as opposed to other types of exercise) is not known to the health monitoring platform, recommending that the user take part in weightlifting to achieve a weight loss goal may not be helpful to the user. Therefore, it is advantageous to obtain psychographic data of a user and utilize that data as part of the contextual data used as input to a contextual MAB model for providing user interaction policies.

[0023] However, there are technical challenges associated with obtaining psychographic data, as it is not always fully available for each user. For example, unlike demographic and / or physiological information that may be available to a new user who is just beginning to use a health monitoring platform, psychographic data may not be available. Furthermore, even if a user is requested to provide psychographic data, the user may not necessarily respond to such requests. At the same time, overburdening a user to continually or frequently seek out such information and provide input may be negative experience and may even be undesirable, resulting in higher user attrition rates. Furthermore, continually requesting a user to provide input regarding psychographic data and updating such information for processing may be technically disadvantageous, as such a process would likely use limited processing, memory, battery, and network resources available to the computing device used by the user.

[0024] Accordingly, certain embodiments herein are directed to performing a psychographic data collection and development phase, as described further below, to collect psychographic data for multiple users and train imputation models to infer user psychographic data unavailable to other users. Obtaining psychographic data for users allows for the creation of a more complete training data set that can be used in training the contextual MAB model described above for determining user interaction policies during the later exploration-exploitation phase.

[0025] By utilizing imputation models, the psychographic data collection and development phase can generate psychographic data that is not available and provide more useful and complete training data for use by the contextual MAB model. In particular, imputation models can be used to infer psychographic data for new users and / or users for whom no or limited information is available. For example, for new users, demographic information may be available, but information about the user's goals and / or interests may not be available. For each of the multiple psychographic features (e.g., goals, behaviors (e.g., diabetes distress, fear of hyperglycemia, psychological well-being), interests, concerns, psychological well-being, etc.) that may be used by exploration-exploitation to provide user interaction policies, there may be a different imputation model for inferring a corresponding value for the psychographic feature, as described further below.

[0026] The systems, devices, and methods of the embodiments described herein can be used in conjunction with any type of analyte sensor for any measurable analyte. Additionally, the systems, devices, and methods of the embodiments described herein may be used in conjunction with any health-related application that is provided to a user to improve the user's health. For example, the health-related application may assist in treating a user's particular disease or assist in improving the health of a user who has not necessarily been diagnosed with a disease.

[0027] Exemplary Environment Having a Data Analytics Platform for Determining User Interaction Policies Based on User Psychographics 1 illustrates an environment 100 in an exemplary implementation operable to employ the techniques described herein. The illustrated environment 100 includes a human 102, who is shown wearing an analyte monitoring system 104, a drug delivery system 106, and a computing device 108. The illustrated environment 100 also includes other users of an analyte monitoring device user population 110, a health monitoring platform 112, and an Internet of Things 114 (IoT 114). The analyte monitoring system 104, the drug delivery system 106, the computing device 108, the user population 110, the health monitoring platform 112, and the IoT 114 are communicatively coupled via a network 116.

[0028] Alternatively or additionally, one or more of the analyte monitoring system 104, the drug delivery system 106, and the computing device 108 are communicatively coupled in other manners, such as using one or more short-range communication protocols and / or techniques. For example, the analyte monitoring device 104, the drug delivery system 106, and the computing device 108 may communicate with each other using one or more of Bluetooth, Near-Field Communication (NFC), 5G, etc. The analyte monitoring system 104, the drug delivery system 106, and the computing device 108 may utilize such types of communication to form a closed-loop system between each other. In this manner, the drug delivery system 106 may deliver a drug (e.g., insulin) based on a prediction of an analyte level (e.g., glucose level) calculated in real time (e.g., by the computing device 108) as analyte measurements are obtained by the analyte monitoring device 104.

[0029] In one or more implementations, the analyte monitoring device 104 is a continuous glucose monitoring ("CGM") device. As used herein, the term "continuous" when used in connection with analyte monitoring may refer to the ability of a device to generate measurements substantially continuously, such that the device may be configured to generate analyte measurements at regular or irregular time intervals (e.g., about every hour, about every 30 minutes, about every 5 minutes, etc.), such as upon establishing a communication coupling with a different device (e.g., when the computing device 108 establishes a wireless connection with the analyte monitoring system 104 to retrieve one or more of the measurements). This functionality, along with further aspects of the configuration of the analyte monitoring device 104, are described in more detail in connection with FIG. 2.

[0030] In one or more embodiments, the analyte monitoring device 104 transmits the analyte measurements 118 to one or more of the computing device 108, the drug delivery system 106, or some other device, e.g., via Bluetooth. The analyte monitoring device 104 may communicate these measurements in real time, e.g., as these measurements are generated using an analyte sensor. Alternatively or additionally, the analyte monitoring device 104 may communicate the analyte measurements 118 at set time intervals, e.g., every 30 seconds, every minute, every 5 minutes, every hour, every 6 hours, daily, etc. Furthermore, the analyte monitoring device 104 can communicate these measurements in response to a request for measurements, which is communicated to the analyte monitoring device 104, e.g., when the computing device 108 causes a user interface with information regarding the glucose level of the human 102 to be displayed, updates such display, predicts the next glucose level of the human 102 for purposes of delivering insulin, etc. Thus, the computing device 108 may at least temporarily maintain the analyte measurements 118 of the human 102 , for example, within a computer-readable storage medium of the computing device 108 .

[0031] In the particular example, it is assumed that the analyte monitoring system 104 is a glucose monitoring system, although it is noted that the analyte monitoring system 104 may be operative to monitor one or more additional or alternative analytes. As used herein, the term "analyte" is a broad term and is to be given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to a substance or chemical constituent in the body or a biological sample (e.g., bodily fluids including blood, serum, plasma, interstitial fluid, cerebrospinal fluid, lymphatic fluid, ocular fluid, saliva, oral fluid, urine, excreta, or exudates). Analytes may include naturally occurring substances, man-made substances, metabolites, and / or reaction products. In some embodiments, the analytes for measurement by the sensing regions, devices, and methods are albumin, alkaline phosphatase, alanine transaminase, aspartate aminotransferase, bilirubin, blood urea nitrogen, calcium, CO2, chloride, creatinine, glucose, gamma-glutamyl transpeptidase, hematocrit, lactate, lactate dehydrogenase, magnesium, oxygen, pH, phosphorus, potassium, sodium, total protein, uric acid, metabolic markers, and drugs.

[0032] Other analytes are contemplated as well, including acetaminophen, dopamine, ephedrine, terbutaline, ascorbate, uric acid, oxygen, d-amino acid oxidase, plasma amine oxidase, xanthine oxidase, NADPH oxidase, alcohol oxidase, alcohol dehydrogenase, pyruvate dehydrogenase, diols, Ros, NO, bilirubin, cholesterol, triglycerides, gentisic acid, ibuprofen, L-dopa, methyldopa, salicylate, tetracycline, tolazamide, tolbutamide, acarboxyprothrombin, riboflavin ... ; acylcarnitines; adenine phosphoribosyltransferase; adenosine deaminase; albumin; α-fetoprotein; amino acid profile (arginine (Krebs cycle), histidine / urocanic acid, homocysteine, phenylalanine / tyrosine, tryptophan); andrenostenedione; antipyrine; arabinitol enantiomers; arginase; benzoylecgonine (cocaine); biotinidase; biopterin; c-reactive protein; carnitine; carnosinase; CD4; ceruloplasmin; chenodeoxycholine; Oxycholic acid;Chloroquine;Cholesterol;Cholinesterase;Conjugated 1-β-hydroxycholic acid;Cortisol;Creatine kinase;Creatine kinase MM isoenzyme;Cyclosporin A;d-Penicillamine;Desethylchloroquine;Dehydroepiandrosterone sulfate;DNA (acetylation polymorphism, alcohol dehydrogenase, α1-antitrypsin, cystic fibrosis, Duchenne / Becker muscular dystrophy, glucose-6-phosphate dehydrogenase, hemoglobin A, hemoglobin S, hemoglobin C, hemoglobin globin D, hemoglobin E, hemoglobin F, D-Punjab, beta-thalassemia, hepatitis B virus, HCMV, HIV-1, HTLV-1, Leber's hereditary optic neuropathy, MCAD, RNA, PKU, Plasmodium vivax, sex differentiation, 21-deoxycortisol);desbutylhalofantrine;dihydropteridine reductase;diphtheria / tetanus antitoxin;erythrocyte arginase;erythrocyte protoporphyrin;esterase D;fatty acids / acylglycines;free beta-human chorionic gonadotropin;free erythrocyte porphyrin;free thyroxine (FT4);Free triiodothyronine tri-iodothyronine, FT3); fumarylacetoacetase; galactose / gal-1-phosphate; galactose-1-phosphate uridyltransferase; gentamicin; glucose-6-phosphate dehydrogenase; glutathione; glutathione peroxidase; glycocholate; glycosylated hemoglobin; halofantrine; hemoglobin variants; hexosaminidase A; human erythrocyte carbonic anhydrase I; 17-α-hydroxyprogesterone; hypoxanthine phosphoribosyltransferase; immunoreactive trypsin; lactate; lead; lipoproteins ((a), B / A-1, β); lysozyme; mefloquine; netilmicin; phenobarbitone; phenytoin; phytanic acid / pristanic acid; progesterone; prolactin; prolidase; purine nucleoside phosphorylase; quinine; reverse triiodothyronine (reverse tri-iodothyronine, rT3); selenium; serum pancreatic lipase; sisomicin; somatomedin C; specific antibodies (adenovirus, antinuclear antibody, anti-zeta antibody, arbovirus, pseudorabies virus, dengue virus, guinea worm, tapeworm, ameba histolytica, enterovirus, giardiasis, Helicobacter pylori, hepatitis B virus, herpes virus, HIV-1, IgE (atopic disease), influenza virus, Leishmania donovani, Leptospira, measles / mumps / rubella, Mycobacterium leprae, Mycoplasma pneumoniae, myoglobin, Onchocerca volvulus, parainfluenza Viruses, malaria parasites, poliovirus, Pseudomonas aeruginosa, respiratory syncytial virus, Rickettsia (tsutsugamushi disease), Schistosoma mansoni, Toxoplasma gondii, Treponema pallidum, Trypanosoma cruzi / langeri, Vesicular stomatitis virus, Wuchereria bancrofti, Yellow fever virus);Specific antigens (Hepatitis B virus, HIV-1);Succinylacetone;Sulfadoxine;Theophylline;Thyrotropin (TSH);Thyroxine (T4);Thyroxine-binding globulin;Trace elements;Transferrin;UDP-galactose-4-epimerase;Urea;Uroporphyrinogen I synthase;Vitamin A;Leukocytes;and zinc protoporphyrin. Salts, sugars, proteins, fats, vitamins, and hormones naturally present in blood or interstitial fluid can also constitute analytes in certain embodiments.

[0033] The analyte may be naturally present in a biological fluid, e.g., a metabolite, hormone, antigen, antibody, etc. Alternatively, the analyte may be introduced into the body, e.g., a contrast agent for imaging, a radioisotope, a chemical agent, a fluorocarbon-based synthetic blood, or a drug or pharmaceutical composition, including, but not limited to, insulin; ethanol; cannabis (marijuana, tetrahydrocannabinol, hashish); inhalants (nitrous oxide, amyl nitrite, butyl nitrite, chlorohydrocarbons, hydrocarbons); cocaine (crack cocaine); stimulants (amphetamine, methamphetamine, ritalin, silurt, preludine, didrex, prestate, borane, sandrex, pregin); antidepressants (barbiturates, methaqualone, valium salts, etc.); , Librium, Miltown, Serax, Equuanil, Tranxine, and other tranquilizers; hallucinogens (phencyclidine, lysergic acid, mescaline, peyote, psilocybin); narcotics (heroin, codeine, morphine, opium, meperidine, Percocet, Percodan, Tasionex, fentanyl, Darvon, Talwin, Lomotil); synthetic narcotics (fentanyl, meperidine, amphetamine, methamphetamine, and analogs of phencyclidine, e.g., ecstasy); anabolic steroids; and nicotine. Metabolites of drugs and pharmaceutical compositions are also contemplated analytes. Analytes such as neurochemicals and other chemicals produced in the body, such as, for example, ascorbic acid, uric acid, dopamine, noradrenaline, 3-methoxytyramine (3MT), 3,4-dihydroxyphenylacetic Acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT), histamine, Advanced Glycation End Products (AGEs), and 5-hydroxyindoleacetic acid (FHIAA), can also be analyzed.

[0034] Although the computing device 108 is illustrated as a wearable device (e.g., a smart watch), it may be configured in a variety of ways without departing from the spirit or scope of the described techniques. By way of example and not limitation, the computing device 108 may be configured as a different type of mobile device (e.g., a mobile phone, a tablet device, a smart ring, or other wearable device), or the computing device 108 may be configured as a dedicated device associated with the health monitoring platform 112, having the functionality to, for example, obtain analyte measurements 118 from the analyte monitoring device 104, perform various calculations on the analyte measurements 118, display information related to the analyte measurements 118 and the health monitoring platform 112, communicate the analyte measurements 118 to the health monitoring platform 112, etc. In contrast to an embodiment in which the computing device 108 is configured as a mobile phone, when the computing device 108 is configured as a dedicated analyte monitoring device, it may not have the functionality available in a mobile phone or wearable configuration, such as the functionality to make phone calls, camera functionality, or utilize social networking applications.

[0035] Additionally, computing device 108 may represent more than one device in accordance with the described techniques. In one or more scenarios, for example, computing device 108 may correspond to both a wearable device (e.g., a smart watch) and a mobile phone. In such a scenario, both of these devices may perform at least some of the same operations, such as receiving glucose measurements 118 from analyte monitoring device 104, communicating those measurements over network 116 to health monitoring platform 112, and displaying information related to analyte measurements 118.

[0036] Alternatively or additionally, different devices may have different capabilities that the other device does not have or that are limited to a designated device through computational instructions. In a scenario where the computing device 108 corresponds to a separate smartwatch and cell phone, for example, the smartwatch may be configured with various sensors and capabilities to measure various physiological markers (e.g., heart rate, respiration, blood flow rate, etc.) and activity (e.g., walking or other movement) of the human 102. In this scenario, the cell phone may not be configured with these sensors and capabilities or may include a limited amount of that functionality - although in other scenarios, the cell phone may be capable of providing the same functionality. Continuing with this particular scenario, the cell phone may have capabilities that the smartwatch does not have, such as an amount of computing resources (e.g., battery and processing speed) that allows the cell phone to more efficiently perform calculations related to the analyte measurements 118. Even in scenarios where the smartwatch is capable of performing such calculations, the computing instructions may limit the execution of those calculations to the cell phone to not burden both devices and efficiently utilize available resources. To this extent, computing device 108 may be configured in different ways and represent a different number of devices than described herein without departing from the spirit and scope of the described technology.

[0037] As described above, the computing device 108 communicates the analyte measurements 118 to the health monitoring platform 112. In the illustrated environment 100, the analyte measurements 118 are shown stored in a storage device 122 of the health monitoring platform 112 as part of the analyte data 120. The storage device 122 may represent one or more databases and other types of storage capable of storing the analyte data 120. The analyte data 120 also includes a user profile 124. In accordance with the described techniques, the human 102 corresponds to at least a user of the health monitoring platform 112 and may also be a user of one or more other third party service providers. To this end, the human 102 may be associated with a username and may be required at some point to provide authentication information (e.g., password, biometric data, etc.) to access the health monitoring platform 112 using the username. This information may be captured in the user profile 124. The user profile 124 may also include various other information about the user, such as demographic information describing the human 102, information about health care providers, payment information, prescription information, determined health indicators, user preferences, account information for other service provider systems (e.g., service providers associated with wearables, social networking systems, etc.) The user profile 124 may include different information about the user within the spirit and scope of the described techniques.

[0038] Moreover, the analyte data 120 not only represents data for the user corresponding to the human 102, but also represents data for other users within the user population 110. With this in mind, the analyte measurements 118 in the storage device 122 include analyte measurements from analyte sensors of the analyte monitoring devices 104 worn by the human 102, as well as analyte measurements from analyte sensors of analyte systems worn by humans corresponding to other users within the user population 110. The analyte measurements 118 of these other users are also communicated by their respective devices over the network 116 to the health monitoring platform 112, and these other users have respective user profiles 124 with the health monitoring platform 112.

[0039] The data analytics platform 126 represents functionality to process the analyte data 120 to generate various predictions, such as by using various machine learning models. Based on these predictions, the health monitoring platform 112 may provide recommended actions. For example, the health monitoring platform 112 may provide decision support output directly to the user or to a medical professional associated with the user. Although depicted separately from the computing device 108, portions or the entirety of the data analytics platform 126 may alternatively or additionally be configured to be implemented on the computing device 108. The data analytics platform 126 is also configured to use data in addition to the analyte measurements 118, such as additional data obtained via the IoT 114, to generate these predictions.

[0040] It should be understood that the IoT 114 represents various sources that can provide data describing the activities and real-world activities of the human 102 and human 102' as users of one or more service providers. By way of example, the IoT 114 may include the user's various devices, e.g., cameras, mobile phones, laptops, etc. Thus, the IoT 114 may provide information regarding the user's interactions with the various devices, e.g., interactions with web-based applications, photos taken, communications with other users, etc. The IoT 114 may also include various real-world items (e.g., shoes, clothing, sporting goods, appliances, automobiles, etc.) that are comprised of sensors that provide information describing behavior, e.g., steps taken, foot force on ground, stride length, the user's body temperature (and other physiological measurements), the user's environmental temperature, types of food stored in the refrigerator, types of food removed from the refrigerator, driving habits, etc.

[0041] The IoT 114 may include third parties to the health monitoring platform 112, such as healthcare providers (e.g., healthcare providers of the human 102) and manufacturers (e.g., manufacturers of the analyte monitoring devices 104, drug delivery systems 106, or computing devices 108) that can provide medical and manufacturing data, respectively, that can be leveraged by the data analytics platform 126. Certainly, the IoT 114 may include devices and sensors capable of providing rich data relevant to recommendations based on analyte monitoring (e.g., continuous glucose monitoring) without departing from the spirit or scope of the described techniques. Consider the following description of FIG. 2 in the context of measuring analytes, for example, continuously, and obtaining data describing such measurements.

[0042] 2 illustrates in greater detail an example implementation 200 of the analyte monitoring device 104 of FIG. 1 in accordance with certain embodiments of the present disclosure. In particular, the illustrated example 200 includes a top view and a corresponding side view of the analyte monitoring device 104. It should be understood from the following description that the analyte monitoring device 104 may be modified in various ways in implementation without departing from the spirit or scope of the described techniques.

[0043] In this example 200, the analyte monitoring device 104 is illustrated to include an analyte sensor 202 (e.g., a glucose sensor) and a sensor module 204. Here, the analyte sensor 202 is shown in a side view inserted, for example, under the skin 206 of the human 102. The sensor module 204 is approximated as a dashed rectangle in a top view. The analyte monitoring device 104 also includes a transmitter 208 in the illustrated example 200. The use of a dashed rectangle for the sensor module 204 indicates that the sensor module may be contained or otherwise implemented within a housing of the transmitter 208. An antenna and / or other hardware used to enable generation of a signal for the transmitter 208 to communicate data, for example, via a wireless connection to the computing device 108, may also be contained or otherwise implemented within a housing of the transmitter 208. In this example 200, the analyte monitoring device 104 further includes an adhesive pad 210.

[0044] In operation, the analyte sensor 202 and adhesive pad 210 may be assembled to form an applicator assembly, which is configured to be applied to the skin 206 such that the analyte sensor 202 is inserted subcutaneously as shown. In such a scenario, the transmitter 208 may be attached to the assembly after affixing to the skin 206 via an attachment mechanism (not shown). Alternatively, the transmitter 208 may be incorporated as part of the applicator assembly, such that the analyte sensor 202, adhesive pad 210, and transmitter 208 (with sensor module 204) can be affixed to the skin 206 substantially all at once. In one or more implementations, this applicator assembly is affixed to the skin 206 using a separate sensor applicator (not shown). Unlike the finger prick required by conventional blood glucose meters, user-initiated application of the analyte monitoring device 104 with the sensor applicator is substantially painless and does not require the drawing of blood. Additionally, an automated sensor applicator generally allows the human 102 to implant the analyte sensor 202 beneath the skin 206 without the assistance of a clinician or health care provider.

[0045] The analyte monitoring device 104 can also be removed by peeling the adhesive pad 210 from the skin 206. It should be understood that the analyte monitoring device 104 and its various components as shown are simply one exemplary form factor, and that the analyte monitoring device 104 and its components can have different form factors without departing from the spirit or scope of the described techniques.

[0046] During operation, the analyte sensor 202 is communicatively coupled to the sensor module 204 via at least one communication channel, which may be a wireless connection or a wired connection. Communications from the analyte sensor 202 to the sensor module 204 or from the sensor module 204 to the analyte sensor 202 may be performed actively or passively, and these communications may be continuous (e.g., analog) or discrete (e.g., digital).

[0047] The analyte sensor 202 may be a device, molecule, and / or chemical that changes or causes a change in response to an event that is at least partially independent of the analyte sensor 202. The sensor module 204 is implemented to receive an indication of a change to or caused by the analyte sensor 202. For example, the analyte sensor 202 may include glucose oxidase that reacts with glucose and oxygen to form hydrogen peroxide that is electrochemically detectable by the sensor module 204, which may include electrodes. In this example, the analyte sensor 202 may be configured as or include a glucose sensor configured to detect an analyte in blood or interstitial fluid that is indicative of glucose levels using one or more measurement techniques. In one or more embodiments, the analyte sensor 202 may also be configured to detect analytes in blood or interstitial fluid that are indicative of other markers, such as lactate levels, ketones, or ionic potassium, which may improve accuracy in identifying or predicting glucose-based events. Additionally or alternatively, the analyte monitoring device 104 may include additional sensors and / or architectures to the analyte sensor 202 for detecting those analytes indicative of other markers.

[0048] In another example, the analyte sensor 202 (or additional, not shown, sensors of the analyte monitoring device 104) can include first and second conductors, and the sensor module 204 can electrically detect a change in electrical potential across the first and second conductors of the analyte sensor 202. In this example, the sensor module 204 and the analyte sensor 202 are configured as a thermocouple, such that the change in electrical potential corresponds to a change in temperature. In some examples, the sensor module 204 and the analyte sensor 202 are configured to detect a single analyte, e.g., glucose. In other examples, the sensor module 204 and the analyte sensor 202 are configured to use multiple sensing modes to detect multiple analytes, e.g., ionic sodium, ionic potassium, carbon dioxide, and glucose. Alternatively or additionally, the analyte monitoring device 104 includes multiple sensors to detect not only one or more analytes (e.g., ionic sodium, ionic potassium, carbon dioxide, glucose, and insulin), but also one or more environmental conditions (e.g., temperature). Thus, the sensor module 204 and analyte sensor 202 (and any additional sensors) may detect the presence of one or more analytes, the absence of one or more analytes, and / or changes in one or more environmental conditions. As described above, the analyte monitoring device 104 may be configured to generate data describing a single analyte (e.g., glucose) or multiple analytes. Furthermore, the combination of analytes for which the analyte monitoring device is configured may vary across different lots of monitoring devices manufactured (e.g., by the health monitoring platform 112), such that wearable analyte monitoring devices having different architectures may be configured for use by different patient populations and / or for different health conditions.

[0049] In one or more embodiments, the sensor module 204 may include a processor and memory (not shown). The sensor module 204 may utilize the processor to generate an analyte measurement 118 based on communication with the analyte sensor 202 indicating the change described above. Based on the communication from the analyte sensor 202, the sensor module 204 is further configured to generate a communicable data package including the at least one analyte measurement 118. In this example 200, the analyte data 116 represents these data packages. Additionally or alternatively, the sensor module 204 may configure the analyte data 116 to include additional data including, by way of example, supplemental sensor information 212. The supplemental sensor information 212 may include a sensor identifier, a sensor status, a temperature corresponding to the analyte measurement 118, measurements of other analytes corresponding to the analyte measurement 118, and the like. It should be appreciated that the supplemental sensor information 212 may include a variety of data supplemental to the at least one analyte measurement 118 without departing from the spirit or scope of the described techniques.

[0050] In embodiments in which the analyte monitoring device 104 is configured for wireless transmission, the transmitter 208 may transmit the analyte data 116 as a data stream to a computing device. Alternatively or additionally, the sensor module 204 may buffer the analyte measurements 118 and / or the supplemental sensor information 212 (e.g., in a memory of the sensor module 204 and / or in another physical computer-readable storage medium of the wearable analyte monitoring device 104) and later have the transmitter 208 transmit the buffered analyte data 120 upon request from another device at various regular or irregular intervals, such as time intervals (about every second, about every 30 seconds, about every minute, about every 5 minutes, about every hour, etc.), storage intervals (when the buffered analyte measurements 118 and / or supplemental sensor information 212 reach a threshold amount of data or number of measurements), etc.

[0051] Having discussed an exemplary environment and an exemplary analyte monitoring system, a description of some exemplary details of the present technology according to one or more embodiments will now be considered.

[0052] Exemplary Process for Determining User Interaction Policies Based on Psychographics 3 is a flow diagram illustrating a process for determining a user interaction policy based on contextual data using a contextual MAB model, according to an embodiment of the present disclosure. As described above, to obtain a more complete training data set including psychographic data to train the contextual MAB model, the process 300 first performs a psychographic data collection and development phase 304, and then uses the more complete training data set including the collected and developed psychographic data to train the contextual MAB model used in the exploration-exploitation phase 306.

[0053] At block 302, in some embodiments, the process 300 includes performing a psychographic data collection stage. The data collection stage includes the health monitoring platform (e.g., health monitoring platform 112) collecting psychographic data from the user, such as by providing the user with a digital survey or quiz that asks the user to identify their goals, interests, and other types of psychographic data. In such an example, the user may be provided with various user interfaces ("UI") on a mobile health application that asks for the psychographic data. For example, a UI may be presented to the user asking what the user's goals are and listing various goals (e.g., lower A1C, spend more time in range, feel better, lower stress, better sleep, more energy) from which the user may select. Other types of data about the user (e.g., non-psychographic data) may be collected as part of block 302 or separately. For example, the health monitoring platform may separately collect various other information about the user, such as, but not limited to, device information, account information, demographic information, food consumption information, activity information (e.g., sleep and exercise), patient status information, medication (e.g., insulin) information, information from connected sensors such as blood glucose, etc.

[0054] In some embodiments, the data collection phase (e.g., including psychographic data collection or collection of other types of data) may be continuous and ongoing, or may be based on a trigger event. For example, the psychographic data collection phase may be a gradual rollout where the user's mobile health application requests information from the user as the user uses the application. Other types of data, such as sensor data, may be collected on an ongoing basis. In another example, the data collection phase may be initiated when the user's mobile health application determines that the confidence score for available or inferred data, such as psychographic data, is low. In such an example, the mobile health application may explicitly request user input.

[0055] As mentioned above, not all users will respond to a request for psychographic data. For example, a survey, quiz, or other type of request for psychographic data may be sent to 100,000 users. However, only 20,000 users may respond to all psychographic-related questions. Thus, a user may have an incomplete psychographic profile that is only partially complete or not complete at all. Therefore, multiple imputation models may be trained to infer or predict the psychographic data of the remaining 80,000 users, each imputation model being configured to infer or predict a different psychographic feature, such as goals, interests, actions, concerns, etc. It should be noted that a psychographic profile is a type of contextual profile (e.g., a user profile with contextual information about a user, which may also include psychographic data).

[0056] At block 302, the process 300 includes training an imputation model using a training data set that includes the psychographic data and / or other types of contextual data collected during block 301. For example, the training data set may be prepared using data collected from 20,000 users, where the training data set includes all kinds of demographic, physiological, and other types of relevant data, as well as psychographic data. Using the training data set, multiple imputation models may be trained, where each imputation model is configured to infer or predict values ​​of different psychographic features, such as goals, interests, actions, concerns, etc. Based on what the imputation model is configured to predict, the training data set may be labeled differently. For example, if the imputation model is trained to predict diabetes distress scores, the training data set is labeled with diabetes distress scores. The imputation model may be trained using one of a variety of artificial intelligence / machine learning algorithms, such as supervised learning algorithms. In an embodiment, the imputation model may be trained as a classification, regression, or other type of model.

[0057] At block 303, process 300 includes using the trained imputation model to develop psychographic data and / or other types of contextual data for one or more users. Continuing with the above example, the trained imputation model may be used to develop psychographic profiles for the 80,000 remaining users for whom the psychographic profiles were incomplete. For example, an imputation model trained to predict a user's diabetes distress score is used to output a predicted diabetes distress score for the remaining 80,000 users. In another example, an imputation model trained to predict a user's hyperglycemia fear score is used to output a predicted hyperglycemia fear score for the remaining 80,000 users. In yet another example, an imputation model trained to predict a type of exercise that a user is interested in is used to output a type of exercise that each of the remaining 80,000 users is interested in. Similarly, an imputation model trained to predict a type of glucose control goal that a user is interested in is used to output a predicted glucose control goal for each of the remaining 80,000 users.

[0058] The imputed psychographic data for the remaining 80,000 users is then used to construct a training dataset that can be used to train a contextual MAB model that is utilized in block 306 to determine user interaction policies.

[0059] At block 306, the process 300 may include performing an exploration-exploitation phase to determine an optimal user interaction policy. The exploration-exploitation phase may be implemented using a contextual MAB model and may include successive explorations on a portion of the user population to refine the contextual MAB model. The exploration-exploitation phase includes an individualized or personalized context model to determine the user interaction policy. In an embodiment, the exploration-exploitation phase also includes utilizing the observed results to train and / or update the contextual MAB model and / or the imputation model discussed above. For example, a health monitoring platform may use feedback telemetry associated with a user interaction policy provided to a user to maximize user participation and / or optimize health outcomes resulting from the user interaction policy.

[0060] As described below in connection with FIG. 4, in the illustrated example exploration-exploitation phase 306, the health monitoring platform divides the set of users into an exploitation subset and an exploration subset. For users in the exploitation subset, the health monitoring platform uses a contextual MAB model (e.g., a predictive model using contextual data) to determine a user interaction policy based on the users' contextual data, including psychographic data. For users in the exploration subset, the computing device randomly assigns user interaction policies and measures the results. In an embodiment, the measured results from the exploration subset of users are more favorable and can be used to explore potential user interaction policies that can be assigned to the exploitation subset of users later. Thus, the exploration-exploitation phase and the contextual MAB model can be utilized to determine an optimal user interaction policy.

[0061] At block 308, the process 300 may also include the health monitoring platform retraining the contextual model and / or the imputation model. In some embodiments, the health monitoring platform may decide to train the model continuously. In some embodiments, the health monitoring platform may decide to retrain the model in response to a trigger. For example, the health monitoring platform may decide to retrain periodically (e.g., monthly regardless of performance). In some embodiments, the process 300 may decide to retrain the model whenever a sample size of a search set of a user's user interaction policies and observed outcomes reaches a threshold number (e.g., 10,000). In certain embodiments, the health monitoring platform retrains the imputation and / or contextual MAB models using user feedback telemetry and / or measured outcomes / rewards associated with the user interaction policies.

[0062] Below, blocks 304 and 306 are described in more detail with reference to Figure 4 which follows. In particular, block 304 of Figure 3 is described in more detail with reference to blocks 406-412 of Figure 4. Additionally, block 306 of Figure 3 is described in more detail with reference to blocks 422-434 of Figure 4.

[0063] 1. Block 304: Psychographics Data Collection and Development Stage As described above with reference to FIG. 3, the process 300 may include performing a psychographic data collection and development stage 304. FIG. 4 is an example operational process for performing a psychographic data collection and development stage in combination with the exploration-exploitation stage according to some embodiments of the present disclosure. The psychographic data collection and development stage may be performed to initially train an imputation model or to retrain an already trained imputation model based on other user data (e.g., device, account, etc.) and / or user feedback telemetry. As explained above, an imputation model is a machine learning model configured to infer psychographic data (e.g., user interests, goals, etc.) for each user when psychographic data is not available (or not sufficiently available). The inferred psychographic data may be used to generate a psychographic profile of the user, which may be used as part of a training dataset for training a contextual MAB model used in the exploration-exploitation stage.

[0064] At block 406, psychographic data is collected for one or more users. For example, psychographic data may be collected using a quiz that is provided to each user, and questions may be provided to the user via various UIs, as described further below. In some embodiments, each user's participation with the UI may also provide psychographic data. For example, if a user engages with content (e.g., video, audio, text) that includes a particular subject matter, that subject matter may be used to collect psychographic data for the user that may indicate the user's interest in such content. In some embodiments, any metadata (e.g., a user's search terms, time spent on a particular UI element, etc.) may be used to collect psychographic data for the user. As described above, psychographic data may be collected for a subset of one or more users, while the remaining users may not be responsive or provide incomplete information.

[0065] Once the psychographic data has been collected for a subset of users, in block 408, the psychographic data may be provided to construct a psychographic profile (i.e., psychographic segmentation). For example, if the user provided an answer to a quiz, the answer may be used to generate a psychographic profile that includes splitting the user's psychographic data into one or more data entries that influence the user's behavior. For example, a survey (e.g., a quiz) may include a series of questions that the user can answer to determine the user's diabetes distress score. In other examples, a survey may be provided to the user about fear of hyperglycemia or psychological well-being. If data is available, by issuing one or more surveys, the answers to the user may be used to generate a training dataset, which becomes an input to a subsequent contextual MAB model, as described further below.

[0066] For the remaining users for whom psychographic data is unavailable or incomplete, the psychographic data collected for the subset of responding users may be used to build imputation models in the training data set. Thus, for users without psychographic data, the health monitoring platform may predict their psychographic data. For example, in block 410, the health monitoring platform may run one or more imputation models to infer the remaining users' psychographic data. There may be an imputation model corresponding to each particular feature (e.g., diabetes distress, hyperglycemia fear, psychological well-being, etc.) in the user's psychographic profile. In other words, for each of these features (i.e., diabetes distress, hyperglycemia fear, psychological well-being), there may be an imputation model to predict the corresponding feature value. Each imputation model may be trained to take as input the user's demographics, physiological information, and other types of relevant information to predict the feature value of a certain psychographic feature in the user's psychographic profile.

[0067] In some embodiments, the imputation model may infer psychographic data (and / or other types of contextual data) based on the best available data. If the inferred psychographic data is unreliable or there is adequate time, the health monitoring platform may periodically request the information from the user to validate or retrain the imputation model. For example, for a user who has demonstrated type 2 diabetes, the imputation model may infer that their goal is better sleep. However, the confidence level for the predicted goal of better sleep for this user may be low. Thus, the health monitoring platform may ask the user to confirm whether better sleep is a goal for the user. In some embodiments, even if the confidence level for the predicted goal of better sleep is high, the health monitoring platform may confirm it with the user.

[0068] The imputation model may also be trained and retrained using user feedback telemetry from the exploration-exploitation phase 306, as shown in block 412. For example, after providing a user interaction policy (e.g., insights, recommendations, etc.) to the user, the user may be provided with a feedback element (e.g., thumbs up, thumbs down, 1 to 5 stars, etc.), and the user's feedback telemetry may be used to refine the imputation model. As an example, the imputation model may predict a user's interests, such as running, and an exercise recommendation to the user may always indicate that the user should engage in running. However, the user may instead be interested in weightlifting. Thus, the user feedback telemetry may indicate that the user is not interested in running, and thus may be used to retrain the imputation model.

[0069] 2. Block 306: Perform exploration-exploitation phase As described above with reference to FIG. 3, the process 300 may include performing an exploration-exploitation phase using a contextual MAB model. FIG. 4 is an exemplary operational process for performing the exploration-exploitation phase in combination with a psychographic data collection and development phase, according to some embodiments of the present disclosure. The exploration-exploitation phase 306 may utilize the contextual MAB model in block 430 to optimize and assign user interaction policies as efficiently as possible. For example, the exploration-exploitation phase avoids the need to "test" many different user interaction policies to see what works best for an individual user. Instead, the exploration-exploitation phase involves training an algorithm to learn what the relationship is between user characteristics (e.g., psychographic data) and the optimal user interaction policy (e.g., recommended action) that results in the best outcome for the user. In an embodiment, the exploration-exploitation phase determines the user interaction policy in a manner that targets both: The user interaction policies are optimized to (i) maximize the likelihood that the user interaction policies will lead to optimal outcomes (e.g., increased user participation, better health outcomes, etc.) and (ii) explore new user interaction policies while leveraging already detected relationships between psychographic data (and / or other types of contextual data) and user interaction policies.

[0070] At block 422, the exploration-exploitation phase 306 may include dividing the set of users into an exploration subset and an exploitation subset. As shown at block 424, the set of users is divided into an exploration subset and an exploitation subset using a sampling criterion. For example, the set of users may be divided, where a portion of ε of the users is randomly selected and used to identify the exploration subset, and the remaining (1-ε) portion of the users is used to develop the exploitation subset. In some embodiments, the magnitude of the exploration ratio ε is decreased after each iteration of the exploration-exploitation phase. In some embodiments, the magnitude of the exploration ratio ε is set to 0 after a predetermined number of iterations of the exploration-exploitation phase. In some embodiments, the magnitude of the decrease in the exploration ratio ε may be a function of the user interaction policy accuracy (which may improve over time). In addition, the exploration ratio ε may also have a lower bound >0 to ensure that some degree of exploration is always performed, since what works best for a user may change over time.

[0071] For the search subset of users, the health monitoring platform may randomly assign a user interaction policy (e.g., a "random" action) in block 426. In block 428, the health monitoring platform may measure results that may be used to train / retrain the contextual MAB model in block 430. In some embodiments, user feedback is also received in block 434 and may be used to train / retrain the contextual MAB and / or imputation models in block 410.

[0072] For the utilization subset of users, the exploration-utilization phase may include running a context model using the context data (e.g., including psychographic data) for each user or user cohort to determine a user interaction policy, at block 430. For example, one of many context models may be a decision support model, which is a user-aware algorithm that determines a user interaction policy, such as decision support recommendations for a user, based on the user's context data.

[0073] At block 432, the user interaction policy may be provided to each user of the utilization subset of users. At block 434, the health monitoring platform may receive user feedback telemetry indicating the effectiveness of the corresponding user interaction policy, e.g., on user participation, health, etc. For example, when the user interaction policy recommends a video, the user's participation in the video may indicate positive user feedback telemetry. However, if the user does not engage with the video, the user feedback telemetry may be determined to be negative. In another example, the recommended video may also include a rating element (e.g., a thumbs-up or thumbs-down element) that allows the user to provide a direct indication of positive or negative user feedback telemetry. In yet another example, the user interaction policy may recommend to the user to perform some action (e.g., walk 5,000 steps today). The user may perform the action, thereby indicating positive user feedback telemetry, or the user may decline to perform the action, thereby indicating negative user feedback telemetry.

[0074] In some embodiments, after implementing a user interaction policy (e.g., watching a recommended video or performing a recommended action), results (e.g., user metrics, such as physiological metrics (e.g., glucose trends)) may be observed in block 428 to measure whether the user's metrics have improved or deteriorated. For example, if a user is provided with recommendations but the user's metrics have deteriorated, the measured results may indicate a low reward. In another example, if a user is provided with recommendations but the user's metrics have not improved or have not improved sufficiently, the measured results may indicate a low reward. As discussed above, user feedback telemetry

[0075] Implementations and examples of the CMAB model are further defined in U.S. Patent Application Publication No. 17 / 931,531, which is incorporated herein by reference. Additionally, implementations and examples of using the recommender model to provide decision support recommendations based on user goals and interests are further defined in U.S. Patent Application Publication No. 17 / 241,919, which is incorporated herein by reference.

[0076] Exemplary User Interface for Providing User Interaction Policies and Collecting Data 5-38 illustrate various UIs provided by a mobile health application executing on a display device (e.g., display device 108). Many of these UIs are used to provide user interaction policies to the user. For example, some of the UIs are used to provide recommendations to the user, while others are used to provide insights or other information. In addition, user feedback telemetry may be received using one or more of the UIs, as described further below.

[0077] Some other UIs may be used to collect data. As further described above, the present embodiment may include a data collection stage (and other data collection steps) for collecting psychographic data and / or other types of data from the user. Such data may be used (1) to train imputation models to infer psychographic data (or other types of data) for users for whom psychographic data is not available (e.g., new users or users who did not respond to a data collection effort) and (2) as context data to provide optimal user interaction policies to the user. Thus, for example, a user may directly provide psychographic data via one or more UIs, as further described below.

[0078] Exemplary UIs shown in FIGS. 5-38 may include an Achievement UI, a Family Collaboration UI, a Travel UI, a Glucose Inspiration UI, a Follower Highlights UI, an Expert Portal UI, a Personalization UI, a Booking Preparation UI, a Video Story UI, and a Health Hub UI. Each of these UIs may be used to collect psychographic data or other types of data, provide user interaction policies, and receive user feedback telemetry. For example, the Personalization UI may present a quiz element that allows users to provide psychographic data. Thus, some users may tell you what their goals are and what they would like to optimize. This psychographic data may be used as training data to build imputation models and generate psychographic segmentations to train the contextual models of the exploration-exploitation phase, as further described above. In this example, the collection of psychographic data may be part of the incremental rollout. In some embodiments, it may be optional. In some embodiments, a user may not provide an answer to a quiz element, and thus the embodiment may use a trained imputation model to infer psychographic data, and may again attempt to collect or verify such data at a later stage in their progressive release. Thus, the embodiment may enable the health monitoring platform to impute psychographic data (e.g., psychographic attributes such as goals and interests) before they are available.

[0079] Achievement UI In an embodiment such as that shown in FIGS. 5-8, users earn badges for achieving milestones or completing challenges related to their diabetes management. In an embodiment, the challenges may be selected based on the user's psychographic profile generated during the psychographic data collection and development phase. Milestones and challenges vary in difficulty, with some being relatively easier to accomplish than others. A mobile health application (e.g., an analyte monitoring application) prompts the user with challenges related to exercise, healthy eating, stress management, glucose levels, etc. Each time a user completes a challenge or achieves a milestone based on their analyte data, the user may earn a unique badge that the application tracks on the mobile health application's achievements page. Users may also choose to join a team with other users of the application and work together to complete team challenges in addition to their own individual challenges.

[0080] 5 illustrates an example of an achievement UI that may be displayed in one or more implementations according to an embodiment of the present disclosure. In example 500, a user interface 502 includes a prompt element 504 with a challenge for the user. For example, the prompt element 504 may indicate, "This week's challenge is to track your sleep time each night and see if you notice any relevant patterns in your glucose. Would you like to participate?" The user may then accept the challenge using a button element (i.e., an "Accept Challenge" button) 506 indicating that the user accepts the presented challenge.

[0081] FIG. 6 illustrates another example of a displayed achievement UI, according to an embodiment of the present disclosure. In example 600, achievement UI 602 includes a text element celebrating the user's achievement ("You did it! You earned a sleep detection badge!") and includes a unique badge 606 earned by the user by completing a challenge. In addition, achievement UI 610 includes badges / rewards 612 earned based on the user's analyte data. In example 600, the badges displayed in interface 610 correspond to the user staying within glucose range for a percentage of the day, e.g., staying between low and high glucose thresholds for 90% of the day. Notably, badges and awards may be based on metrics other than just analyte data. For example, badges / rewards 606 displayed in achievement UI 602 are based on the user's sleep data.

[0082] 7 illustrates another example of a displayed achievement UI, according to an embodiment of the present disclosure. In example 700, the achievement UI 702 includes text elements, e.g., an "Achievements" page, and displays badges earned by the user. In example 700, the badges correspond to different milestones associated with different categories, such as glucose milestone 706 and badge 708, emotional milestone 710 and badge 712, athletic milestone 714 and badge 716, and nutrition milestone 718 and badge 720. Notably, a user may earn multiple awards for each category by achieving various different milestones or completing challenges. A first glucose milestone 706 may correspond, for example, to staying within glucose range for 50% of the day, and a second glucose milestone may correspond to staying within glucose range for 75% of the day.

[0083] 8 illustrates example badges displayed via a user's achievement UI and via an achievement UI for a team of users competing together, according to an embodiment of the present disclosure. In example 800, badges are shown for user 802, e.g., based on individual challenges in which the users participate. In example 800, badges are also shown for user 804's team, e.g., based on team challenges in which the users participate. The badges from example 800 are displayed via the respective achievement UIs.

[0084] Family joint UI In embodiments such as those shown in FIGS. 9-12, the mobile health application allows groups of users associated with a particular user (e.g., family, groups of friends, sports teams, etc.) to interact in various ways to support the particular user. For example, the user's entire family may use the application together to better support the user. In one or more embodiments, when first logging into the application, the user may take a short quiz that collects the user's psychographic data, including the user's needs and preferences regarding the assistance they would like in managing their health (e.g., diabetes). Each person in the user's group (e.g., the user's family) takes a similar quiz about their own preferences with respect to supporting the user, providing psychographic data. The application then generates recommendations and educational suggestions on topics to encourage conversations about adverse conditions related to analyte monitoring, such as type 2 diabetes. From there, the application allows users to send messages to each other in appreciation of positive behaviors, such as buying diabetes-friendly groceries or going for walks together.

[0085] 9 illustrates an example of a displayed family collaboration UI, according to an embodiment of the present disclosure. In the example 900, the family collaboration UI 902 illustrates at least a portion of the quizzes taken by a supporter (e.g., a family member) of a particular user. For example, the family collaboration UI 902 includes a text element ("Supporter Needs Assessment"), a first question element 906 (i.e., "Please Help"), a first answer element 908 that allows the supporter to answer the first question element 906, a second question element 910 ("I'll help you"), a second answer element 912 that allows the supporter to answer the second question element 910, a third question element 914 ("How do you feel:"), and a third answer element 916 that allows the supporter to answer the third question element 914. Additionally, the family collaboration UI 920 illustrates at least a portion of the quizzes taken by a particular user. For example, family collaboration UI 920 includes a text element (i.e., "Type 2 Needs Assessment"), a first question element 924 ("Diabetes management, support needed:"), a first answer element 926 that allows the user to respond to the first question element 924, a second question element 928 ("What time is convenient for you:"), a second answer element 930 that allows the user to respond to the second question element 928, and a third question element 932 ("How do you feel:"), and a third answer element 934 that allows the user to respond to the third question element 932.

[0086] 10 illustrates another example of a displayed family collaboration UI, according to an embodiment of the present disclosure. In example 1000, family collaboration UIs 1002, 1010, 1020 illustrate personalized recommendations displayed and / or generated by a mobile health application to prompt a conversation regarding adverse conditions or ways to provide support related to analyte monitoring, such as type 2 diabetes. For example, family collaboration UI 1002 includes personalized recommendation 1004 for Casey, including issue 1006 and information 1008 about type 2 diabetes. Further, family collaboration UI 1010 includes personalized recommendation 1012 for Alex, including issue 1014 and information about type 2 diabetes 1016. Additionally, family collaboration UI 1020 includes personalized recommendation 1022 for Morgan, including issue 1024 and information about type 2 diabetes 1026.

[0087] 11 illustrates another example of a displayed family collaboration UI, according to an embodiment of the present disclosure. In example 1100, the family collaboration UI 1102 displays interface elements that allow users to send and receive messages in recognition of positive or helpful behaviors. The family collaboration UI 1102 includes a message box element 1104 that allows group participants to use a button element 1106 (i.e., a "send" button) to send a message to the user thanking the user for taking action to improve diabetes management (e.g., by thanking the user for bringing home healthy groceries).

[0088] 12 illustrates another example of a family collaboration UI that may be displayed in one or more implementations according to an embodiment of the present disclosure. In example 1200, the family collaboration UI 1202 displays interface elements that enable feedback indicating how a user feels about a received message. For example, the family collaboration UI 1202 may display a received message 1206 and include a "like" button (i.e., a thumbs up button) that enables a recipient to provide feedback regarding a received message, such as one sent from another user via the family collaboration UI 1102 shown in example 1100.

[0089] 9-12 may allow users who have established accounts in association with a particular user to inform the particular user that they are being supported. The mobile health application may be used by a family (or another group) together to better support a person with an adverse health condition associated with the analyte being monitored, e.g., type 2 diabetes.

[0090] Process sharing UI In an embodiment such as that shown in Figs. 13-15, a particular user may share / initiate the particular user's analyte monitoring process with other users. When the user purchases an analyte monitoring device, the user is presented (e.g., displayed) with an option to purchase a kit including the analyte monitoring device for themselves along with additional analyte monitoring device(s) for one or more of the user's supporters to try for a period of time, e.g., one month. Thus, the user and the selected supporter may learn together about the analyte monitoring device by each wearing an analyte monitoring device and sharing data via the mobile health application. Thus, the application may collect psychographic data from the user and the selected supporter and provide user interaction policies to optimize the user's health outcomes and the supporter's ability to provide support to the user. The application allows a group of associated users, whether they have diabetes or not, to share the experience of reviewing glucose data. The user and their selected group may even challenge each other as a team, for example, to achieve the user's health goals together through fun games. At the end of the month, the mobile health application sends a report to one or more of the users with everyone's analyte data and, optionally, one or more educational insights based on the analyte trends.

[0091] 13 illustrates an example of a displayed process sharing UI and process, according to an embodiment of the present disclosure. In example 1300, the process sharing UI 1302 (displayed via a display device) includes a first text element 1304 that displays information about the CGM system and an option for the user to initiate a "process sharing" program with one or more other users, for example, the user and the other users wear an analyte monitoring device for a period of time. For example, the process sharing UI 1302 includes a second text element 1306 (i.e., "Add a Process Sharing Kit?") and a button element 1308 (i.e., "Check Here" button) that enables the user to initiate the "Process Sharing" program.

[0092] 14 illustrates another example of a displayed process sharing UI, according to an embodiment of the present disclosure. In example 1400, the process sharing UI 1402 illustrates a game used as part of a "process sharing" program. The process sharing UI 1402 includes a first text element 1404 that identifies a group (i.e., "team name") and visual elements 1406 that indicate various users. Here, the game is displayed on a bingo board 1408 that includes analyte-related behavior boxes. Users can perform various analyte-related behaviors (e.g., attempting a new workout 1410, staying within glucose range for a certain percentage of the day 1412, switching to a low carbohydrate diet 1414, or managing stress levels 1416) to "fill in" the associated squares on the bingo board 1408.

[0093] 15 illustrates another example of a displayed process sharing UI, according to an embodiment of the present disclosure. In example 1500, the process sharing UI 1502 includes a first text element 1504 that displays a summary report of an associated group of users, e.g., the user's family, friends, or colleagues, selected via the process sharing UI 1302 of FIG. 13. The process sharing UI 1502 also includes a visual element 1506 that shows various user and analyte data (e.g., glucose traces) for all users in the group, and / or one or more educational insights for those users based on determined trends. In an embodiment, the visual element 1506 may highlight the glucose data for each user.

[0094] Glucose Inhalation UI In an embodiment such as that shown in FIGS. 16-18, the mobile health application allows a user to compare how the user's glucose is at one or more time points against other users on a similar journey. The user is presented with an option to join a group of users' choices, which may be selected by the computing device based on the user's psychographic profile. Additionally, the application compares metrics determined from the user's monitored analyte data (e.g., time in range) with those of other users, e.g., other users with type 2 diabetes. The application allows the user to send congratulations to others on the team, as an option. For example, if a user notices that another user is doing well, they can send a congratulatory message to encourage the other user to continue the good work.

[0095] FIG. 16 illustrates an example of a displayed glucose inhale UI, according to an embodiment of the present disclosure. In example 1600, the glucose inhale UI 1602 displays analyte-based metric data for users in a group that a particular user has selected to join and how the analyte-based metric data of the group members for a period of time (e.g., a day) compares. For example, the glucose inhale UI 1602 includes a first text element 1604 (i.e., “dog lovers”) that identifies the group and the number of members (shown as having 34 members). Additionally, the glucose inspiration UI 1602 also includes a ranking list element 1606 that ranks users based on analyte-based metric data (e.g., percentage of time in range). In the illustrated example, the ranking list element 1606 displays the user (i.e., “me”) and a ranking text element 1608 that indicates that the user has moved to second place. The glucose inhale UI 1602 can also include a button element 1610 that allows the user to chat with the group.

[0096] 17 shows another example of a displayed glucose intake UI, according to an embodiment of the present disclosure. In example 1700, glucose intake UI 1702 includes text element 1704 ("Join a Team") and button elements 1706-1714 that display groups that a user can select to join, for example, to compare the user's metrics to those of other users in the group. Groups may be interest-based, condition-based, time-based, age-based, and location-based, to name a few.

[0097] 18 illustrates another example of a displayed glucose inhale UI, according to an embodiment of the present disclosure. In example 1800, the glucose inhale UI 1802 includes a first text element 1804 (i.e., "Dog Lovers Chat") that identifies the group and the number of members (shown as 34 members). The glucose inhale UI 1802 also includes messaging elements from members in the selected group (e.g., Messages from Greg 1806 and Messages from Patricia 1808) that enable members of the group to send and receive messages to other members in the group.

[0098] Follower Highlight UI In an embodiment such as that shown in FIGS. 19-21, the mobile health application allows a user to receive recognition from family and friends based on the user's psychographic data, such as analyte data and the user's goals and interests. The user can choose to allow one or more supporter(s) to follow the user's analyte data, and the authorized users can send or view daily highlights along with descriptions of the user's achievements for the day. One or more supporting users can also receive real-time notifications on their respective mobile health applications when the user is experiencing good glucose times. The mobile health application allows supporting users to quickly send messages to the user, for example, a user sending a timely congratulations to show their support. In an embodiment, messages may be selected based on the user's psychographic profile, and a context model determines the optimal user interaction policy.

[0099] FIG. 19 illustrates an example of a displayed follower highlight UI, according to an embodiment of the present disclosure. In example 1900, follower highlight UI 1902 displays the user's analyte data (e.g., glucose data) and / or analyte highlights via the support user's display device, thereby allowing the support user to view the user's accomplishments. For example, follower highlight UI 1902 includes a first highlight element 1906 that provides highlights of the user's analyte data (i.e., "Taylor achieved 94% in one day. Amazing!"). Follower highlight UI 1902 also includes a button element 1908 that allows the support user to send a message to the user. Additionally, follower highlight UI 1902 also includes a second highlight element 1910 that provides details of the highlights of the user's analyte data, such as, but not limited to, a glucose trend graph.

[0100] 20 illustrates another example of a displayed follower highlight UI, according to an embodiment of the present disclosure. In example 2000, follower highlight UI 2002 displays elements that allow a support user to interact to send a quick message to the user to show support. For example, follower highlight UI 2002 includes a prompt element 2004 (i.e., "Select a message to send to Taylor") and a selection of a graphic element 2008 that includes a selection of a first message 2006 (i.e., "Nice job, Taylor, you made 94% of your range for the day!"), a second message 2008 ("Congratulations, Taylor, you were in range today, awesome!"), and a graphic (e.g., emoji).

[0101] 21 illustrates another example of a displayed follower highlight UI, according to an embodiment of the present disclosure. In example 2100, follower highlight UI 2102 displays a message that the user received from a support user. For example, follower highlight UI 2102 includes a message 2106 sent by the support user ("Congratulations, Taylor, you were in range today, awesome!").

[0102] In one or more embodiments, the mobile health application allows users to send congratulations to other users, e.g., family and friends, based on their monitored analyte data. The mobile health application allows support users to track a particular user's analyte data and see daily highlights along with an explanation of what went well that day for that particular user. The support user can also receive real-time notifications that users they follow are experiencing good glucose moments. The mobile health application allows support users to quickly send congratulatory messages to followed users to show support to the followed users at the right moment.

[0103] Expert Portal UI In an implementation such as that shown in Figures 22-24, the mobile health application provides an expert portal. In some embodiments, the expert portal may be curated based on the user's interaction policies, as described above. Through the expert portal, the user is provided with access to content from a pool of diabetes experts who support the user. In some embodiments, the content may be personalized to the user's interests. The user is provided with access to content on diabetes-related topics from medical experts with different areas of expertise. The experts available through the portal may include, by way of example and not limitation, physicians, diabetes educators, nutritionists, life coaches, and psychologists, to name a few. In some embodiments, if a user has a question that is not yet addressed by the content in the portal's library, the user may submit a question and receive a response from one of the experts in the portal. In some embodiments, the user may submit a question and receive a content match that may answer their question.

[0104] FIG. 22 illustrates an example of a displayed expert portal UI, according to an embodiment of the present disclosure. In example 2200, the expert portal UI 2202 illustrates elements that allow a user to search a library of content (e.g., videos, podcasts, and articles) from diabetes experts. The expert portal UI allows a user to view, read, and / or listen to content on topics that may be related to the user's search. For example, the expert portal UI 2202 includes a search element 2204 that allows a user to search the library of content. In such an example, the expert portal UI 2002 also provides results when a search term is provided by a user to the search element 2204. In other examples, the expert portal UI 2002 can preload content from the library of content before the user launches the search element 2204. In both examples, the expert portal UI 2202 can provide link elements (e.g., video links 2208, audio links 2210, or text links 2212) to expert content.

[0105] 23 illustrates another example of a displayed expert portal UI, according to an embodiment of the present disclosure. In example 2300, the expert portal UI 2302 includes a search element 2204 that allows a user to search a library of content. If the library of content does not contain the user's search, the expert portal UI 2302 includes a text element 2306 that indicates that no match was found and asks the user if they would like to be notified when an answer becomes available. In such an example, the expert portal UI 2302 also includes a Yes button 2308 and a No button 2310. When the Yes button 2308 is indicated by the user, the user can be notified later when a match is found.

[0106] FIG. 24 illustrates another example of a displayed expert portal UI, according to an embodiment of the present disclosure. In example 2400, the expert portal UI 2402 includes a notification element 2404 that notifies the user when new content that may be relevant to the user is discovered. For example, the user receives a response or is otherwise notified when a resource related to a submitted search or question is added to the portal. In the illustrated example, the expert portal UI 2402 includes a video link 2406 to a video prepared by an expert. The video may be related to the submitted search or may answer a question submitted by the user. In one or more implementations, popular topics for a period of time are presented via the expert portal UI 2402. For example, topics that have been asked most frequently by other users during that period of time (e.g., a month) are presented via the expert portal UI 2402. In one or more implementations, the expert portal UI 2402 includes an element selectable by the user to vote for one or more of the popular topics, which further increases the popularity of the hot topic and the relative importance of the responding expert.

[0107] Personalized UI In an embodiment such as that shown in FIG. 25, the mobile health application allows the user to set and track personalized analyte-based goals. In this manner, the computing device may receive user psychographic data directly from the user. For example, the mobile health application may allow the user to personalize their analyte monitoring experience based on the user's own individual goals. When the user first sets up the mobile health application, for example, they may be prompted to take a quick quiz to set goals related to a health condition (e.g., diabetes-related goals). The tips and resources displayed to the user in the mobile health application are tailored to the user's needs and are user-specific. Because the user's priorities may change over time, the application allows the user to adjust the application's settings at any point in time.

[0108] 25 illustrates an example of a displayed personalization UI, according to an embodiment of the present disclosure. In example 2500, personalization UI 2502 includes a quiz element 2504 that allows a user to take a quiz to set one or more health condition (e.g., diabetes) related goals. In an embodiment, the quiz may be presented at various times, such as upon initialization. Alternatively or additionally, the user may be directed to, for example, the application's settings or some other interface to initiate the quiz. Personalization UI 2502 lists a variety of different goals, such as spending more time in range, feeling better, lower stress, better sleep, more energy, etc., that may be selected by the user via one or more selection elements (e.g., selection element 2506 for "low A1C").

[0109] After selection, the user may be provided with a personalized UI interface 2510 that displays decision support outputs (e.g., tips and resources) curated for the user based on the user's selected goal and other data associated with the user's goal. For example, the personalized UI 2510 includes a first decision support output 25122 (i.e., "low carb recipes") and a second decision support output 2514 (i.e., "hear from others who share the same goal"). In an embodiment, the tips and resources may be curated based on other user data. In one or more implementations, the mobile health application informs the user of goal achievement, for example, by presenting a congratulatory message via the user interface.

[0110] The personalization UI may allow the user to update and / or provide additional data related to their goals. For example, the personalization UI 2520 displays an element that allows the user to take the next quiz at the next time, where the next quiz allows the user to select one or more health condition-related goals. In one or more implementations, the application prompts the user to take the subsequent quiz. The personalization UI 2520 includes a quiz element 2522 that allows the user to take a quiz to update one or more health condition (e.g., diabetes)-related goals. In the illustrated example, the user selects "Sleep Better" as the updated goal using the selection element 2524.

[0111] After selection, the user may be provided with a personalization UI interface 2530 that displays decision support outputs (e.g., tips and resources) curated for the user based on the user's updated goals, data associated with the user's goals, or other user data. For example, the personalization UI 2530 includes a first decision support output 2534 (i.e., "Get ready for sleep") and a second decision support output 2536 (i.e., "Hear from people who share the same goals").

[0112] Reservation preparation UI In an embodiment such as that shown in FIGS. 26-31, the mobile health application allows the user to prepare for appointments with a healthcare provider. The benefit of this application is that the user can be better prepared for appointments with a healthcare provider. In the application, the user interface displays the user's upcoming appointments. Before each appointment, the mobile health application generates and provides a list of questions to ask the healthcare provider that are editable by the user to fit the user's needs. In an embodiment, the list of questions can be based on user interaction policies. Furthermore, based on the user's analyte data and trends, the mobile health application may suggest questions to add to the action list. Additionally or alternatively, the user can flag days or moments of interest in the analyte data (e.g., a particular glucose spike or pattern (e.g., hypoglycemia at a certain time)) and add a note to talk to the healthcare provider about the analyte data for that day. Before the user proceeds to each appointment, the mobile health application provides a checklist with reminders of what to do ahead of time to make the appointment as efficient and valuable as possible. The user can store important takeaways / answers to questions after the appointment so the user can refer to them at any time.

[0113] 26 illustrates an example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. In example 2600, appointment preparation UI 2602 includes a calendar element 2604 and a text element 2606 that displays upcoming appointments and reminder notifications, for example, with a healthcare provider.

[0114] 27 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. In example 2700, appointment preparation UI 2702 includes a list element 2704 that includes appointment preparation questions for the user to ask at the time of the appointment. Appointment preparation UI 2702 also includes an add button element (i.e., "add question") that allows the user to add a question to a list of questions associated with the upcoming appointment. Appointment preparation UI 2702 also includes a send button element 2708 that is selectable by the user to send the list of questions and the user's analyte data to a healthcare provider associated with the user. The application builds a list of questions for the user during the appointment for the user to ask their healthcare professional. The list of questions can include questions determined by the application and / or questions selected by the user.

[0115] FIG. 28 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. In example 2800, appointment preparation UI 2802 includes an element for adding a question to the user's list, where the user is prompted to add a question from an application based on an analysis of the user's analyte data over a period of time. For example, based on the user's glucose data, the system may determine that the user has been experiencing hypoglycemia every night during the past week. Based on this determination, the system asks the user if the user would like to question their physician about the pattern of nighttime hypoglycemia. In the illustrated example, appointment preparation UI 2802 includes a prompt element 2804 that displays the user's analyte data (i.e., "I have been experiencing hypoglycemia every night this week") and asks the user additionally ("Would you like to ask your physician about this?"). Appointment preparation UI 2802 also includes an add button element 2806 (a button element indicating "Yes, add to my preparation list") and a reject button element 2808 (a button element indicating "No, thank you").

[0116] 29 shows another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. In example 2900, appointment preparation UI 2902 includes a trend graph element 2904 (e.g., showing the user's analyte concentration over a period of time) and a flagging element 2906 that allows the user to flag a period of time (e.g., days) and may cause the application to remind the user to discuss the analyte data for that period with the user's medical professional. Appointment preparation UI 2902 also includes an add button element 2908 (a button element that indicates "Yes, add to my appointment list").

[0117] 30 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. In example 3000, the appointment preparation UI 3002 displays a checklist element 3004 to provide the user with a checklist (e.g., "Review glucose data," "Download report," etc.) that can be reviewed by the user prior to an appointment with a healthcare provider to, for example, help the user and healthcare provider get the most out of the appointment.

[0118] 31 illustrates another example of a displayed appointment preparation UI, according to an embodiment of the present disclosure. In example 3100, appointment preparation UI 3102 includes a notes list element 3104 that allows a user to document important takeaways / answers to post-appointment questions.

[0119] Video Story UI In an embodiment such as that shown in FIGS. 32-35, the mobile health application recommends digital content, e.g., videos, based on the user's psychographic profile. The mobile health application allows the user to hear and / or watch stories about other people with health conditions, e.g., type 2 diabetes. Once the user signs into the mobile health application and / or corresponding website, the user interface presents high-quality videos (or other digital content) curated to the user based on a variety of factors, including analyte data over a period of time (e.g., what the user's glucose looks like during the week) and other user data (e.g., demographics, digital content selection history, similarity to other users, activity data, dietary data, medication data, etc.). The application also allows the user to interact with elements to filter videos based on categories of interest to the user. Additionally, the application allows the user to provide feedback on videos viewed, and the next videos recommended by the application are adjusted based on the feedback.

[0120] 32 illustrates an example of a displayed video story UI, according to an embodiment of the present disclosure. In example 3200, video story UI 3202 displays elements that allow a user to browse a collection of digital content. For example, video story UI 3202 includes video links organized based on different categories of videos ("Top Viewed" 3206 and "Newly Updated" 3208).

[0121] 33 illustrates another example of a displayed video story UI, according to an embodiment of the present disclosure. In example 3300, video story UI 3302 displays selectable elements to filter digital content based on the type of content the user is interested in consuming. For example, video story UI 3302 has filter buttons 3304 that allow the user to filter digital content based on "location," "interests," "time since diagnosis," etc.

[0122] 34 illustrates another example of a displayed video story UI, according to an embodiment of the present disclosure. In example 3400, video story UI 3402 includes feedback elements that enable a user to provide feedback regarding digital content output via a computing device, such as displayed via user interface 3402. For example, video story UI 3402 may include a prompt element 3404 ("Is this video relevant?"), a thumbs up element 3406, and a thumbs down element 3408.

[0123] 35 illustrates another example of a displayed video story UI, according to an embodiment of the present disclosure. In example 3500, video story UI 3502 includes a suggestions element 3504 that displays recommended digital content tailored to the user's preferences, for example, based on the user's feedback or based on the user's other viewed videos.

[0124] In one or more implementations, the mobile health application outputs a notification when it determines that at least one item of digital content may provide useful insight to the user, e.g., based on the user's analyte data over a period of time and / or based on metrics (e.g., trends) derived from the data over time.

[0125] Health Hub UI In an embodiment such as that shown in Figures 36-38, the mobile health application interfaces with a marketplace that enables interaction with additional third party applications that are related to the mobile health application and may be based on the user's psychographic profile. The mobile health application connects to other applications, for example, health related applications for a holistic experience. In the mobile health application, the user can select other applications that will connect to the mobile health application to view data. Based on the user selection, the mobile health application has access to high level metrics from each selected application, all in one central location.

[0126] 36 illustrates an example of a displayed health hub UI, according to an embodiment of the present disclosure. In example 3600, the health hub UI 3602 includes a trend graph element 3604 (e.g., a graph plotting trends of analytes and / or analyte-based metrics over time) and a selection button element 3606 that the user can select to interact with one or more other applications. In other words, the health hub UI 3602 allows the user access to a marketplace of other connected applications from the mobile health application's homepage (or other pages).

[0127] 37 illustrates another example of a displayed health hub UI, according to an embodiment of the present disclosure. In example 3700, the health hub UI 3702 displays a number of other applications selectable for connecting with the mobile health application. For example, the health hub UI 3702 includes a first button element ("headspace"), a second button element ("0"), a third button element ("fitbit"), a fourth button element ("Loose it!"), and a fifth button element ("weight watchers").

[0128] 38 illustrates another example of a displayed health hub UI, according to an embodiment of the disclosure. In example 3800, the health hub UI 3802 includes a presentation element 3804 that shows information obtained from a selected other application ("fitbit"), the health hub UI 3802 being displayed by the mobile health application. In this manner, the mobile health application presents information from the other application without navigating away from the mobile health application. For example, the mobile health application presents metrics (e.g., "high level" or "summary" metrics) from a selected application without leaving the mobile health application.

[0129] Having described exemplary procedures according to one or more implementations, exemplary systems and devices that can be utilized to implement the various techniques described herein are now discussed.

[0130] Exemplary Systems and Devices for Providing User Interaction Policies 39 illustrates an example system generally at 3900, including an example computing device 3902, which represents one or more computing systems and / or devices that may implement various techniques described herein. This is illustrated through the inclusion of the health monitoring platform 112. The computing device 3902 may be, for example, a server of a service provider, a device associated with a client (e.g., a client device), an on-chip system, and / or any other suitable computing device or computing system.

[0131] Computing device 3902 as shown includes a processing system 3904, one or more computer readable media 3906, and one or more I / O interfaces 3908 communicatively coupled to each other. Although not shown, computing device 3902 may further include a system bus or other data and command transfer system coupling various components together. The system bus may include any one or combination of different bus structures, such as a memory bus or memory controller, a peripheral bus, a universal serial bus, and / or a processor or local bus utilizing any of a variety of bus architectures. Various other examples, such as control and data lines, are also contemplated.

[0132] The processing system 3904 represents functionality for performing one or more operations using hardware. Thus, the processing system 3904 is illustrated as including hardware elements 3910, which may be configured as a processor, functional blocks, etc. This may include implementations in hardware as application specific integrated circuits or other logic devices formed using one or more semiconductors. The hardware elements 3910 are not limited by the materials from which they are formed, or the processing mechanisms employed therein. For example, a processor may be comprised of semiconductors and / or transistors (e.g., electronic integrated circuits (ICs)). In this context, processor-executable instructions may be electronically executable instructions.

[0133] The computer readable medium 3906 is illustrated as including memory / storage 3912. The memory / storage 3912 represents memory / storage capacity associated with one or more computer readable media. The memory / storage component 3912 may include volatile media (such as random access memory (RAM)) and / or non-volatile media (such as read only memory (ROM), flash memory, optical disks, magnetic disks, etc.). The memory / storage component 3912 may include fixed media (e.g., RAM, ROM, fixed hard drives, etc.) as well as removable media (e.g., flash memory, removable hard drives, optical disks, etc.). The computer readable medium 3906 may be configured in a variety of other ways, as further described below.

[0134] Input / output interface 3908 represents functionality that allows a user to input commands and information into computing device 3902 and that allows information to be presented to a user and / or other components or devices using various input / output devices. Examples of input devices include keyboards, cursor control devices (e.g., a mouse), microphones, scanners, touch capabilities (e.g., capacitive or other sensors configured to detect physical contact), cameras (e.g., employing visible wavelengths or non-visible wavelengths such as infrared frequencies to recognize movements as non-touch gestures), and the like. Examples of output devices include display devices (e.g., a monitor or projector), speakers, printers, network cards, haptic response devices, and the like. Thus, computing device 3902 can be configured in a variety of ways to support user interaction, as described further below.

[0135] Various techniques may be described herein in the general context of software, hardware elements, or program modules. Generally, such modules include routines, programs, objects, elements, components, data structures, etc. that perform particular tasks or implement particular abstract data types. As used herein, the terms "module," "function," and "component" generally refer to software, firmware, hardware, or combinations thereof. Aspects of the techniques described herein are platform-independent, meaning that the techniques may be implemented on a variety of commercial computing platforms having a variety of processors.

[0136] An implementation of the described modules and techniques may be stored on or transmitted across some form of computer-readable media. Computer-readable media may include a variety of media that may be accessed by computing device 3902. By way of example, and not limitation, computer-readable media may include "computer-readable storage media" and "computer-readable signal media."

[0137] A "computer-readable storage medium" may refer to a medium and / or device that allows for persistent and / or non-transient storage of information, as opposed to merely a signal transmission, carrier wave, or signal itself. Thus, a computer-readable storage medium refers to a non-signal-bearing medium. A computer-readable storage medium includes hardware, such as volatile and non-volatile, removable and non-removable media, and / or storage devices implemented in a manner or technology suitable for storage of information, such as computer-readable instructions, data structures, program modules, logic elements / circuits, or other data. Examples of computer-readable storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage, hard disk, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device, or other storage device, tangible media, or any article suitable for storing the desired information and that can be accessed by a computer.

[0138] A "computer-readable signal medium" may refer to a signal-bearing medium configured to transmit instructions to the hardware of the computing device 3902, such as via a network. Signal media may typically embody computer-readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave, data signal, or other transport mechanism. Signal media also includes any information delivery media. The term "modulated data signal" means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media include wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, RF, infrared, and other wireless media.

[0139] As previously described, the hardware elements 3910 and computer-readable media 3906 represent modules, programmable device logic, and / or fixed device logic implemented in hardware form that may be employed in some embodiments to implement at least some aspects of the techniques described herein, such as to execute one or more instructions. The hardware may include integrated circuits or components of on-chip systems, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), complex programmable logic devices (CPLDs), and other implementations in silicon or other hardware. In this context, the hardware may operate as a processing device that executes program tasks defined by instructions and / or logic embodied by the hardware, as well as hardware utilized to store instructions for execution, such as computer-readable storage media previously described.

[0140] Combinations of the foregoing may be used to implement various techniques described herein. Thus, software, hardware, or executable modules may be implemented as one or more instructions and / or logic embodied on some form of computer-readable storage medium and / or by one or more hardware elements 3910. Computing device 3902 may be configured to implement specific instructions and / or functions corresponding to software and / or hardware modules. Thus, implementation aspects of modules executable as software by computing device 3902 may be achieved at least partially in hardware, for example, through the use of computer-readable storage medium and / or hardware elements 3910 of processing system 3904. The instructions and / or functions may be executable / operable by one or more articles of manufacture (e.g., one or more computing devices 3902 and / or processing system 3904) to implement the techniques, modules, and examples described herein.

[0141] The techniques described herein may be supported by various configurations of computing device 3902 and are not limited to the specific examples of the techniques described herein. This functionality may also be implemented in whole or in part through the use of a distributed system, such as on a "cloud" 3914 via platform 3916, as described below.

[0142] Cloud 3914 includes and / or represents a platform 3916 for resources 3918. Platform 3916 abstracts the underlying functionality of the hardware (e.g., servers) and software resources of cloud 3914. Resources 3918 may include applications and / or data available while computer processing is running on a server remote from computing device 3902. Resources 3918 may also include services provided over the Internet and / or over a subscriber network, such as a cellular network or a Wi-Fi network.

[0143] The platform 3916 can abstract resources and functionality for connecting the computing device 3902 with other computing devices. The platform 3916 can also serve to abstract the scaling of resources to provide a level of scale corresponding to the demand faced by the resources 3918 implemented via the platform 3916. Thus, in an embodiment of interconnected devices, implementations of the functionality described herein can be distributed throughout the system 3900. For example, the functionality can be performed in part on the computing device 3902 as well as via the platform 3916 that abstracts the functionality of the cloud 3914.

[0144] Exemplary embodiments Embodiment 1: A non-transitory computer-readable medium including instructions that, when executed by at least one processor, cause the at least one processor to perform a method including: dividing a plurality of users into an exploration subset of users and an utilization subset of users; randomly assigning at least one user interaction policy for each exploration subset of users; and determining at least one user interaction policy for each utilization subset of users utilizing one or more context models trained using context data corresponding to the utilization subset of users, wherein the context data corresponding to the utilization subset of users includes at least some of a first set of context profiles and a second set of context profiles.

[0145] Embodiment 2: A non-transitory computer-readable medium as described in embodiment 1, wherein the exploration-exploitation phase is further performed by receiving user feedback telemetry from an exploitation subset of users, the feedback telemetry providing information regarding the effectiveness of at least one user interaction policy assigned to each user of the exploitation subset of users.

[0146] Embodiment 3: A non-transitory computer-readable medium as described in embodiment 2, wherein at least one of the one or more imputation models or at least one of the context models is retrained using user feedback telemetry.

[0147] Embodiment 4: The non-transitory computer-readable medium of embodiment 1, wherein the exploration-exploitation phase is further performed by measuring results associated with an exploitation subset of users, the measured results providing information regarding the effectiveness of at least one user interaction policy assigned to each user of the exploitation subset of users.

[0148] Embodiment 5: The non-transitory computer-readable medium of embodiment 4, wherein at least one of the context models is retrained using the measured results.

[0149] Embodiment 6: The non-transitory computer-readable medium of embodiment 1, wherein the method further includes: collecting context data for a first subset of the plurality of users; generating a first set of context profiles for the first subset of the plurality of users based on the collected context data; determining that the context data for a second subset of the plurality of users is incomplete or unavailable; training one or more imputation models based on the context data of the first subset of the plurality of users to create the context data for the second subset of the plurality of users; generating the context data for the second subset of the plurality of users using the one or more imputation models; and generating a second set of context profiles for the second subset of the plurality of users based on the generated context data for the second subset of the plurality of users.

[0150] Embodiment 7: A non-transitory computer-readable medium as described in embodiment 6, wherein the contextual data of a first subset of the plurality of users corresponds to psychographic data of the first subset of the plurality of users, the first set of contextual profiles of the first subset of the plurality of users corresponds to a first set of psychographic profiles of the first subset of the plurality of users, the contextual data of a second subset of the plurality of users corresponds to psychographic data of the second subset of the plurality of users, and the second set of contextual profiles of the second subset of the plurality of users corresponds to a second set of psychographic profiles for the second subset of the plurality of users.

[0151] Embodiment 8: A non-transitory computer-readable medium as described in embodiment 6, wherein determining that the contextual data of the second subset of the plurality of users is incomplete or unavailable includes requesting contextual data from the second subset of the plurality of users and determining at least one of: (1) some of the second subset of the plurality of users provided incomplete data in response to the request, or (2) some of the second subset of the plurality of users did not provide data in response to the request.

[0152] Embodiment 9: A non-transitory computer-readable medium including instructions that, when executed by at least one processor, cause the at least one processor to perform a method including: collecting context data for a first subset of a plurality of users; generating a first set of context profiles for the first subset of the plurality of users based on the collected context data; determining that context data for a second subset of the plurality of users is incomplete or unavailable; training one or more imputation models based on the context data of the first subset of the plurality of users to create context data for the second subset of the plurality of users; generating the context data for the second subset of the plurality of users using the one or more imputation models; and generating a second set of context profiles for the second subset of the plurality of users based on the generated context data for the second subset of the plurality of users.

[0153] Embodiment 10: A non-transitory computer-readable medium as described in embodiment 9, wherein the contextual data of a first subset of the plurality of users corresponds to psychographic data of the first subset of the plurality of users, the first set of contextual profiles of the first subset of the plurality of users corresponds to a first set of psychographic profiles of the first subset of the plurality of users, the contextual data of a second subset of the plurality of users corresponds to psychographic data of the second subset of the plurality of users, and the second set of contextual profiles of the second subset of the plurality of users corresponds to a second set of psychographic profiles for the second subset of the plurality of users.

[0154] Embodiment 11: A non-transitory computer-readable medium as described in embodiment 9, wherein determining that the contextual data of the second subset of the plurality of users is incomplete or unavailable includes requesting contextual data from the second subset of the plurality of users and determining at least one of: (1) some of the second subset of the plurality of users provided incomplete data in response to the request, or (2) some of the second subset of the plurality of users did not provide data in response to the request.

[0155] Embodiment 12: The method further includes dividing a plurality of users into an exploration subset of users and an utilization subset of users, randomly assigning at least one user interaction policy for each exploration subset of users, and determining at least one user interaction policy for each utilization subset of users utilizing one or more context models trained using context data corresponding to the utilization subset of users, wherein the context data corresponding to the utilization subset of users includes at least some of a first set of context profiles and a second set of context profiles. A non-transitory computer-readable medium as described in embodiment 9.

[0156] Embodiment 13: A non-transitory computer-readable medium as described in embodiment 12, wherein the exploration-exploitation phase is further performed by receiving user feedback telemetry from an exploitation subset of users, the feedback telemetry providing information regarding the effectiveness of at least one user interaction policy assigned to each user of the exploitation subset of users.

[0157] Embodiment 14: A non-transitory computer-readable medium as described in embodiment 13, wherein at least one of the one or more imputation models or at least one of the context models is retrained using user feedback telemetry.

[0158] Embodiment 15: A non-transitory computer-readable medium as described in embodiment 12, wherein the exploration-exploitation phase is further performed by measuring results associated with an exploitation subset of users, the measured results providing information regarding the effectiveness of at least one user interaction policy assigned to each user of the exploitation subset of users.

[0159] Embodiment 16: A non-transitory computer-readable medium as described in embodiment 15, wherein at least one of the context models is retrained using the measured results. [Explanation of symbols]

[0160] 100 Environment 102 Humans 104 Analyte Monitoring Systems 106 Drug Delivery Systems 108 Computing Devices 110 User Group 112 Health Monitoring Platform 114 Internet of Things 116 Network 118 Analyte Measurements 120 Analyte Data 122 Storage Devices 124 User Profile 126 Data Analysis Platform 200 Examples 202 Analyte Sensor 204 Sensor Module 206 Skin 208 Transmitter 210 Adhesive Pad 212 Supplemental Sensor Information 3900 System 3902 Computing Devices 3904 Processing System 3906 Computer-readable medium 3908 I / O Interface 3910 Hardware Elements 3912 Memory / Storage 3914 Cloud 3916 Platform 3918 resources

Claims

1. A non-transitory computer-readable medium containing instructions that, when executed by at least one processor, cause the at least one processor to: collecting context data for a first subset of a plurality of users; generating a first set of context profiles for the first subset of the plurality of users based on the collected context data; determining that context data for a second subset of the plurality of users is incomplete or unavailable; training one or more imputation models based on the context data of the first subset of the plurality of users to generate the context data of the second subset of the plurality of users; generating the context data for the second subset of the plurality of users using the one or more imputation models; and generating a second set of context profiles for the second subset of the plurality of users based on the generated context data for the second subset of the plurality of users.

2. the contextual data of the first subset of the plurality of users corresponds to psychographic data of the first subset of the plurality of users; the first set of contextual profiles of the first subset of the plurality of users corresponds to a first set of psychographic profiles of the first subset of the plurality of users; the contextual data of the second subset of the plurality of users corresponds to psychographic data of the second subset of the plurality of users; 2. The non-transitory computer-readable medium of claim 1 , wherein the second set of contextual profiles for the second subset of the plurality of users corresponds to a second set of psychographic profiles for the second subset of the plurality of users.

3. The method further includes performing an exploration-exploitation step, the exploration-exploitation step comprising: Partitioning the plurality of users into an exploration subset of users and an utilization subset of users; randomly assigning at least one user interaction policy to each of said search subsets of users; and determining at least one user interaction policy for each of the utilization subsets of users utilizing one or more context models trained using context data corresponding to the utilization subsets of users, the context data including at least some of the first set of context profiles and the second set of context profiles.

4. The exploration-exploitation phase includes:

4. The non-transitory computer-readable medium of claim 3, further performed by receiving user feedback telemetry from the utilizing subset of users, the feedback telemetry providing information regarding effectiveness of the at least one user interaction policy assigned to each user of the utilizing subset of users.

5. 5. The non-transitory computer-readable medium of claim 4, wherein at least one of the one or more imputation models or at least one of the contextual models is retrained using the user feedback telemetry.

6. The exploration-exploitation phase includes:

4. The non-transitory computer-readable medium of claim 3, further comprising: measuring results associated with the utilization subset of users, the measured results providing information regarding effectiveness of the at least one user interaction policy assigned to each user of the utilization subset of users.

7. The non-transitory computer-readable medium of claim 6 , wherein at least one of the context models is retrained using the measured results.

8. The non-transitory computer-readable medium of claim 1 , wherein at least one of the contextual models is a contextual multi-armed bandit model.

9. collecting context data for a first subset of a plurality of users; generating a first set of context profiles for the first subset of the plurality of users based on the collected context data; determining that context data for a second subset of the plurality of users is incomplete or unavailable; training one or more imputation models based on the context data of the first subset of the plurality of users to generate the context data of the second subset of the plurality of users; generating the context data for the second subset of the plurality of users using the one or more imputation models; and generating a second set of context profiles for the second subset of the plurality of users based on the generated context data for the second subset of the plurality of users.

10. the contextual data of the first subset of the plurality of users corresponds to psychographic data of the first subset of the plurality of users; the first set of contextual profiles of the first subset of the plurality of users corresponds to a first set of psychographic profiles of the first subset of the plurality of users; the contextual data of the second subset of the plurality of users corresponds to psychographic data of the second subset of the plurality of users; The method of claim 9 , wherein the second set of contextual profiles for the second subset of the plurality of users corresponds to a second set of psychographic profiles for the second subset of the plurality of users.

11. The method further includes performing an exploration-exploitation step, the exploration-exploitation step comprising: Partitioning the plurality of users into an exploration subset of users and an utilization subset of users; randomly assigning at least one user interaction policy to each of said search subsets of users; and determining at least one user interaction policy for each of the utilization subsets of users utilizing one or more context models trained using context data corresponding to the utilization subsets of users, the context data including at least some of the first set of context profiles and the second set of context profiles.

12. The exploration-exploitation phase includes:

12. The method of claim 11, further performed by receiving user feedback telemetry from the active subset of users, the feedback telemetry providing information regarding effectiveness of the at least one user interaction policy assigned to each user of the active subset of users.

13. The method of claim 12 , wherein at least one of the one or more imputation models or at least one of the context models is retrained using the user feedback telemetry.

14. The exploration-exploitation phase includes:

12. The method of claim 11, further performed by measuring results associated with the active subset of users, the measured results providing information regarding the effectiveness of the at least one user interaction policy assigned to each user of the active subset of users.

15. The method of claim 14 , wherein at least one of the context models is retrained using the measured results.

16. The method of claim 9 , wherein at least one of the contextual models is a contextual multi-armed bandit model.

17. 1. A computing system comprising: one or more memories containing executable instructions; one or more processors in data communication with the one or more memories and configured to execute the instructions, collecting context data for a first subset of a plurality of users; generating a first set of context profiles for the first subset of the plurality of users based on the collected context data; determining that context data for a second subset of the plurality of users is incomplete or unavailable; training one or more imputation models based on the context data of the first subset of the plurality of users to generate the context data of the second subset of the plurality of users; generating the context data for the second subset of the plurality of users using the one or more imputation models; and generating a second set of context profiles for the second subset of the plurality of users based on the generated context data for the second subset of the plurality of users.

18. the contextual data of the first subset of the plurality of users corresponds to psychographic data of the first subset of the plurality of users; the first set of contextual profiles of the first subset of the plurality of users corresponds to a first set of psychographic profiles of the first subset of the plurality of users; the contextual data of the second subset of the plurality of users corresponds to psychographic data of the second subset of the plurality of users; 20. The computing system of claim 17, wherein the second set of contextual profiles for the second subset of the plurality of users corresponds to a second set of psychographic profiles for the second subset of the plurality of users.

19. The processor may further be configured to perform an exploration-exploitation step, the processor configured to perform the exploration-exploitation step may include the processor, Dividing the plurality of users into an exploration subset of users and an utilization subset of users; randomly assigning at least one user interaction policy to each of said search subsets of users; 20. The computing system of claim 17, configured to determine at least one user interaction policy for each of the utilization subsets of users using one or more context models trained using context data corresponding to the utilization subsets of users, the context data including at least some of the first set of context profiles and the second set of context profiles.

20. The processor may further be configured to perform an exploration-exploitation step, the processor configured to perform the exploration-exploitation step may include the processor, 4. The computing system of claim 3, further configured to receive user feedback telemetry from the active subset of users, the feedback telemetry providing information regarding effectiveness of the at least one user interaction policy assigned to each user of the active subset of users.

Images