Systems and methods for predicting vaccine uptake

Abstract

Certain aspects described herein include systems and methods for predicting vaccine uptake. A method includes presenting, to a user via a user interface, a set of stimuli; and receiving, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rating in the set of ratings corresponds with a respective stimuli in the set of stimuli. The method further includes determining a set of judgment variables based on the set of ratings; and generating, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users.

Description

SYSTEMS AND METHODS FOR PREDICTING VACCINE UPTAKECROSS REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to U.S. Provisional Application Serial No. 63 / 449,460, filed March 02, 2023, the entire contents of which are incorporated herein by reference.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under N000142112216 awarded by the Office of Naval Research. The government has certain rights in the invention.TECHNICAL FIELD

[0003] The present disclosure relates to computer systems and methods that are used for public health purposes, and more specifically, to machine learning models that are particularly trained for predicting results of potential public health interventions.BACKGROUND

[0004] Vaccination is a method of administration of vaccines to prevent or reduce infectious or malignant diseases. Vaccination (also called immunization) may induce immunity in a subject by presenting an agent representing a disease to the subject’s immune system for recognition as a threat and destruction. In the future, when presented with the disease, the immune system may recognize the disease and destroy the microorganism associated with the disease. Vaccination is also part of public health measures, for example through herd immunity. Because some subjects may not be able to be vaccinated, for example due to underlying health conditions, these subjects rely on herd immunity for protection. When a large portion of a population has been vaccinated for a particular disease, it is more difficult for that disease to circulate among the population. Thus, those who are not vaccinated may be protected through herd immunity because they are less likely to be exposed to the disease.

[0005] Vaccination has been used to control and / or eliminate many diseases, for example, smallpox, polio, diphtheria, measles, rubella, pertussis, and coronavirus. However, some of these diseases persist and may reemerge in under-vaccinated communities. Further, new diseases mayalso emerge for which vaccines may not yet be developed, or only recently developed but not widely distributed.

[0006] Vaccination hesitancy may result in delay or refusal of vaccination. Reduction in vaccination may result in individual harm, for example, though contraction of a vaccine- preventable disease. Reduction in vaccination may also reduce herd immunity in a population and result in harm to the population-at-large. Vaccination hesitancy may be due a mix of factors, including cultural, spiritual, political, and safety factors. For example, vaccine hesitancy may be related to safety concerns, efficacy concerns, complacency regarding risk, and accessibility limitations.SUMMARY

[0007] In one embodiment, a computer-implemented method of predicting vaccine uptake is provided, the method comprising: presenting, to a user via a user interface, a set of stimuli; receiving, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rating in the set of ratings corresponds with a respective stimuli in the set of stimuli; determining a set of judgment variables based on the set of ratings; generating, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users; generating a vaccination mechanism recommendation based on the vaccine uptake prediction; and transmitting, to the user, the vaccination mechanism recommendation.

[0008] In another embodiment, a computer-implemented method of training a vaccine uptake prediction model, the method comprising: generating training data based on a sets of ratings, comprising: presenting, to one or more users via a user interface, a set of stimuli; receiving, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli; and determining a set of judgment variables based on the sets of ratings; and training the vaccine uptake prediction model with the training data to generate a vaccine uptake prediction.

[0009] In another embodiment, a processing system, comprising: a memory comprising computer-executable instructions; and a processor configured to execute the computer-executable instructions and cause the processing system to: present, to a user via a user interface, a set of stimuli; receive, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rat in the set of ratings corresponds with a respective stimuli inthe set of stimuli; determine a set of judgment variables based on the set of ratings; generate, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users; generate a vaccination mechanism recommendation based on the vaccine uptake prediction; and transmit, to the user, the vaccination mechanism recommendation.

[0010] In another embodiment, a processing system, comprising: a memory comprising computer-executable instructions; and a processor configured to execute the computer-executable instructions and cause the processing system to: generate training data based on a sets of ratings, comprising: present, to one or more users via a user interface, a set of stimuli; receive, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli; and determine a set of judgment variables based on the sets of ratings; and train the vaccine uptake prediction model with the training data to generate a vaccine uptake prediction.

[0011] In another embodiment, a non-transitory computer-readable medium comprising computer-executable instructions that, when executed by a processor of a processing system, cause the non-transitory computer-readable medium to: present, to a user via a user interface, a set of stimuli; receive, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rat in the set of ratings corresponds with a respective stimuli in the set of stimuli; determine a set of judgment variables based on the set of ratings; generate, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users; generate a vaccination mechanism recommendation based on the vaccine uptake prediction; and transmit, to the user, the vaccination mechanism recommendation.

[0012] In another embodiment, a non-transitory computer-readable medium comprising computer-executable instructions that, when executed by a processor of a processing system, cause the non-transitory computer-readable medium to generate training data based on a sets of ratings, comprising: present, to one or more users via a user interface, a set of stimuli; receive, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli; and determine a set of judgment variables based on the sets of ratings; and train the vaccine uptake prediction model with the training data to generate a vaccine uptake prediction.

[0013] Additional features and advantages of the technology described in this disclosure will be set forth in the detailed description which follows, and in part will be readily apparent to those skilled in the art from the description or recognized by practicing the technology asdescribed in this disclosure, including the detailed description which follows, the claims, as well as the appended drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The embodiments set forth in the drawings are illustrative and exemplary in nature and not intended to limit the disclosure. The following detailed description of the illustrative embodiments can be understood when read in conjunction with the following drawings, where like structure is indicated with like reference numerals and in which:

[0015] FIG. 1 is a representation of an example system for predicting vaccine uptake of a user, in accordance with various disclosed aspects herein.

[0016] FIG. 2 is a representation of an example processing device configured to generate a vaccine uptake prediction, in accordance with various disclosed aspects herein.

[0017] FIG. 3 is a representation of an example workflow for generating a prediction, for example, a vaccine uptake prediction, in accordance with various disclosed aspects herein.

[0018] FIG. 4 is a representation of an example workflow for training a machine learning model for generating a prediction, in accordance with various disclosed aspects herein.

[0019] FIG. 5A is a representation of an example rating task, in accordance with various disclosed aspects herein.

[0020] FIGS. 5B is a representation of example relative preference graphs for determination of one or more judgment variables, in accordance with various disclosed aspects herein.

[0021] FIG. 5C is a representation of example relative preference graphs for determination of one or more judgment variables, in accordance with various disclosed aspects herein.

[0022] FIG. 5D is a representation of example relative preference graphs for determination of one or more judgment variables, in accordance with various disclosed aspects herein.

[0023] FIG. 5E is a representation of various judgment variables, in accordance with various disclosed aspects herein.

[0024] FIG. 6 is a representation of an example method for generating a prediction, for example, a vaccine uptake prediction, in accordance with various disclosed aspects herein.

[0025] FIG. 7 is a representation of an example method for training a machine learning model for generating a prediction, in accordance with various disclosed aspects herein.

[0026] Reference will now be made in greater detail to various embodiments of the present disclosure, some embodiments of which are illustrated in the accompanying drawings. Whenever possible, the same reference numerals will be used throughout the drawings to refer to the same or similar parts.DETAILED DESCRIPTION

[0027] Aspects of the present disclosure provide apparatuses, methods, processing systems, and computer-readable mediums for predicting vaccine uptake.

[0028] As described herein, vaccination has both individual and population-wide implications. Under-vaccination in a population may result in the reemergence or delayed control of vaccine preventable diseases. However, vaccination is an individual choice. Vaccination hesitancy may be due to a variety of factors. For example, vaccine hesitancy may be related to safety concerns, efficacy concerns, complacency regarding risk, and accessibility limitations.

[0029] Vaccination hesitancy results in under-vaccinated populations, in which vaccine preventable diseases may continue to spread, resulting in disease, disability, and even death. Vaccination hesitancy is a growing global problem. However, tackling this problem is challenging because vaccine hesitancy in one population may be due to different factors than vaccine hesitancy in another population. Further, individuals within a population may each have different concerns resulting in vaccine hesitancy.

[0030] Additionally, in many cases, it may be difficult to determine whether a particular individual will become vaccinated or not. Thus, it may be further difficult to determine at a population level predicted vaccination rates.

[0031] Moreover, health care and public health services may inefficiently and incorrectly allocate vaccination and healthcare resources because of these difficulties. For example, in an under-vaccinated population, increased health care resources may be needed due to increased infection and disease spread, while fewer vaccination resources may be used. This may result inover-provision of vaccination supplies and under-provision of health care supplies. In some cases, vaccination may be higher than expected, and people desiring vaccination may be left vulnerable due to insufficient supplies.

[0032] Embodiments described herein provide technical solutions to the aforementioned problems. In particular, embodiments described herein provide for systems and methods for generating predictions of vaccine uptake. In certain embodiments, a predicted vaccine uptake may indicate a vaccination status for a subject, for example, whether a user has or will become vaccinated. A vaccination status may include a positive vaccination status where the subject has or will become vaccinated. A vaccination status may include a negative vaccination status where the subject has not or will not become vaccinated. In particular, embodiments described herein provide systems and methods for generating vaccine uptake predictions with a machine learning model based on judgment variables that quantify judgment and relative preference. In embodiments, judgment variables may be determined based on a rating task completed by a subject. A rating task may include a subject rating or indicating an approach or avoidance (e.g., positive or negative) to various stimuli. For example, a subject may be presented with sets of pictures of various categories and the subject assigns a rating to each picture. Based on the ratings, a set of judgment variables may be determined. The set of judgment variables may then be processed with a machine learning model trained to predict vaccine uptake of the subject.

[0033] Certain embodiments described herein further include generating and training the machine learning model to predict vaccine uptake of a subject based on judgment variables derived from their ratings.

[0034] Certain embodiments described herein utilize a rating task to determine the set of judgment variables. This rating task may be readily deployed to personal computing devices of users, for example, smart devices or computers, enabling convenient completion. Beneficially, the rating task may be completed remotely, for example, to reduce infection spread. Further, the rating task may be deployed to a population, for example, to enable prediction of vaccine uptake at a population level, which, as described herein, facilitates improve public health administration.

[0035] Furthermore, the rating task limits biases involved in vaccination choice. The rating task involves presenting sets of stimuli, such as pictures, to users and receiving user ratings. The rating task, including the stimuli, does not have a perceivable relation to vaccination. Thus, the machine learning model does not incorporate such bias in generating the vaccine uptake prediction.

[0036] In certain embodiments, the vaccine uptake prediction of a subject may beneficially facilitate improved health outcomes for the subject. For example, vaccine uptake prediction may facilitate improved targeted messaging to the subject regarding the subject’s health and safety, such as infection risk or vaccination risk. As another example, the vaccine uptake prediction indicates the subject will be vaccinated, a vaccination appointment may be scheduled or a vaccination dose may allocated.

[0037] Moreover, aggregation of vaccine uptake prediction, for example, for a locality or a population may facilitate improved public health measures. For example, aggregated vaccine uptake prediction may be used to aid vaccine supply chain and administration logistics by indicating areas that may utilize increased or decreased vaccine supply. As another example, aggregated vaccine uptake prediction may be used to aid health care systems by indicating areas of higher or lower infection, based on lower or higher population vaccination rates. As yet another example, aggregated vaccine uptake prediction may facilitate improved targeted messaging regarding vaccination, disease or infection, and other public health messaging in an effort to improve future vaccine uptake.

[0038] Vaccine uptake prediction may facilitate local messaging and enable health care institutions to prepare for vaccine roll out (e.g., administration) and / or vast infection. For example, health care institutions with a lower vaccine uptake prediction may have a higher rate of infection and health care resources may be reallocated from vaccine administration to infection care. Similarly, health care institutions with a higher vaccine uptake prediction may have a lower rate of infection, and health care resources may be reallocated from infection care to vaccine administration.

[0039] Vaccine uptake prediction may further facilitate regional or national public health administration by improving vaccine uptake, and limiting infection and health care overload, for example, through improved messaging and improve vaccine administration. As an example, vaccine uptake predictions may be used to improve communication by tailoring or targeting messaging; defining the demographic and behavioral variables to facilitate the tailoring or targeting of messaging; design treatment or application protocols; informing and supporting health care infrastructure regarding preparing areas with predicted high infection rates; inform and support pharmaceutical efforts regarding research, clinical trials and recruitment; informing and supporting governmental services and policy making for any medical process, public health need, or defensive challenge; informing and supporting pharmaceutical companies regarding vaccineroll-outs or other supply chain issues related or unrelated to the field of medicine; defining risk attitudes and habits that prevent healthy behavior and engagement in products that improve health or happiness; improving communication by understanding individual preference profiles (e.g., risk aversion, loss aversion, etc.) and tailoring or targeting messaging for marketing, financial services, social media, insurance, health care services, travel services, and other commercial endeavors; aide and facilitating emergency services during novel medical crises (e.g., pandemics, ecological disasters, industrial accidents / exposures, military engagements, etc.); and facilitate public health surveillance for mitigation of negative health outcomes;

[0040] Reference will now be made in detail to embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings. It is to be understood that other embodiments may be utilized, and structural and functional changes may be made without departing from the scope of the present disclosure. Moreover, features of the embodiments may be combined, switched, or altered without departing from the scope of the present disclosure, e.g., features of each disclosed embodiment may be combined, switched, or replaced with features of the other disclosed embodiments. As such, the following description is presented by way of illustration and does not limit the various alternatives and modifications that may be made to the illustrated embodiments and still be within the spirit and scope of the present disclosure.

[0041] As used herein, the words "example" and "exemplary" mean an instance, or illustration. The words "example" or "exemplary" do not indicate a key or preferred aspect or embodiment. The word "or" is intended to be inclusive rather an exclusive, unless context suggests otherwise. As an example, the phrase "A employs B or C," includes any inclusive permutation (e.g., A employs B; A employs C; or A employs both B and C). As another matter, the articles "a" and "an" are generally intended to mean "one or more" unless context suggest otherwise.Example Prediction System

[0042] FIG. 1 depicts an example system 100 for predicting vaccine uptake of a user. In some embodiments, system 100 is configured to predict uptake of an identified vaccine, for example, a coronavirus vaccine, a flu vaccine, or a human papillomavirus vaccine. In some embodiments, system 100 is configured to predict uptake of any vaccine. In some embodiments, system 100 is configured to predict acceptance of a medical treatment. For example, a medical treatment related to cardiovascular health, pulmonary health, nutrition, prenatal and obstetrics care, addiction, ormental health. In some embodiments, system 100 is configured to predict acceptance of public health and safety measures, for example, public safety, medical and public health.

[0043] System 100 may include a user 102 interacting with a computing device 108 to complete a rating task. Computing device 108 comprises a user interface 118. The user interface 118 may run on a variety of computing devices, including personal computers, tablet computers, smart devices, and others. The user interface 118 may be a graphical user interface for a website, application, software program, and the like. The user interface 118 is configured to be displayed on a monitor, television, touchscreen, and the like, of computing device 108.

[0044] Vaccine administration component 120 comprises vaccine prediction component 110, vaccine supply component 122 and vaccine mechanism component 124. Vaccine prediction component 110 comprises ratings component 112 and machine learning component 114. In the depicted example, vaccine administration component 120 is depicted separately from computing device 108, however in other examples, vaccine administration component 120 may be configured to run on computing device 108.

[0045] In FIG. 1, a user 102 interacts with one or more pictures 106 presented on the user interface 118 of a computing device 108. The user interface 118 may be rendered by computing device 108, such as a monitor of a touch screen of computing device 108. Computing device 108 may control the user interface 118 to present the one or more pictures 106. In some embodiments, the computing device 108 is configured to present one or more stimuli to the user 102, in this example, one or more pictures 106. Other example stimuli may include one or more sounds or one or more videos. The one or more pictures 106 may comprise one or more subsets, wherein all the pictures in a subset are associated with a category (e.g., based on content, type, effect, etc.).

[0046] Further, the computing device 108 is configured to receive one or more ratings 104 from the user 102. For example, the user 102 may select or input a rating, for example through a rating user interface element displayed on user interface 118. In some embodiments, the user 102 may select one or more rating options displayed on a user interface element of user interface 118. In some embodiments, the user 102 may enter, for example through a touchscreen of user interface, through one or more control elements of user interface, or through one or more control elements of computing device 108 (e.g., keys, mouse, buttons, etc.), text comprising a rating. Each of the one or more ratings may correspond to a respective picture in the one or more pictures 106 presented on the user interface of computing device 108. For example, one picture of the one or more pictures 106 may be presented to the user 102 and one rating of the one or more ratings 104may be received from user 102. The one rating corresponds to the one picture presented, for example, rating scale 504 in FIG. 5 A depicts an example discrete rating scale associated with an example picture 502. A subset of ratings may be determined based on each rating corresponding to a picture in a given subset. For example, a subset of pictures may be associated with the category of nature, e.g., each picture in the subset depicts nature. Each subset may be identified by association to a category, for example, a nature category. Each rating entered by the user 102 for a nature picture in the nature category subset of pictures may be determined to be part of a nature category subset of ratings. Thus, each subset of ratings may correspond to a subset of pictures, wherein each rating in the subset of ratings corresponds to one of the pictures in the subset of pictures.

[0047] A rating may be based on the user’s preferences, emotions, and / or attention, for example, a user may rate the picture 106 based on an initial response. In some embodiments, a rating may comprise a continuous scale. For example, a continuous scale may be a numerical rating, such as from 0 to 5, from -3 to +3, etc. As another example, a continuous scale may be based on a sliding scale, such as a rating between “dislike very much” to “like very much” endpoints. In some embodiments, a rating may comprise a discrete scale, or one or more categories. For example, a user may select a rating from a set of rating options, such as one or more emoticons (e.g., smiley face, sad face, angry face, etc.), one or more terms (e.g., “dislike very much”, “dislike”, “neutral”, “like”, “like very much”, etc.), one or more emotions (e.g., “happy”, “sad”, “angry”, etc.).

[0048] The computing device 108 is further configured to interface with a vaccine prediction component 110, for example to send the ratings 104 received from user 102. Vaccine prediction component 110 may be run on computing device 108, or an external component (e.g., accessed through an application programming interface (API), such as in a microservices-type deployment). Vaccine prediction component 110 is configured to generate a vaccine uptake prediction 116 associated with user 102. As described in further detail, for example, with respect to workflow 300 of FIG. 3 and method 600 of FIG. 6, the vaccine update prediction 116 may indicate a vaccination status of the user 102, for example, a positive vaccination status may indicate a user 102 will become vaccinated, while a negative vaccination status may indicate the user 102 will not become vaccinated. Vaccine prediction component 110 comprises a ratings component 112 and a machine learning component 114.

[0049] Ratings component 112 is configured to determine a set of judgment variables based on the one or more ratings 104 from the user 102. The set of judgment variables may be a quantification of aspects of judgments made by the user 102 in rating the pictures. Each judgment variable in the set of judgment variables may describe a quantitative component of a user’s approach, avoidance, or judgment behavior. The set of judgment variables may include, for example, loss aversion, risk aversion, loss resilience, ante, insurance, peak positive risk, peak negative risk, reward tipping point, aversion tipping point, total reward risk, total aversion risk, reward aversion tradeoff, tradeoff range, reward aversion consistency, and consistency range. For example, FIG. 5E depicts example judgment variables and associated abbreviations.

[0050] Ratings component 112 is configured to determine a set of approach / avoidance variables based on the one or more ratings 104 from the user 102. The set of approach / avoidance variables may include, for example, a mean, variance, and uncertainty of the one or more ratings 104 from the user 102. In some embodiments, a set of approach / avoidance variables may be determined for each subset of ratings, e.g., each subset of ratings corresponding to a subset of pictures (e.g., based on category). Ratings component 112 is further configured to graph the approach / avoidance variables, and as described in further detail below, for example, with respect to workflow 300 in FIG. 3, and method 600 in FIG. 6, determine one or more judgment variables based on the graph(s).

[0051] Machine learning component 114 is configured to process the set of judgment variables with a machine learning model to generate a vaccine uptake prediction 116. The machine learning model may be trained to predict whether user 102 with be vaccinated. In some embodiments, machine learning component 114 comprises a classification model, for example, a binary classification model. Examples of classification models include a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine (SVM), or an artificial neural network. In some embodiments, the machine learning model 114 comprises a balanced random forest classifier machine learning model.

[0052] In some embodiments, machine learning component 114 is further configured to process biographical data in generating the vaccine uptake prediction 116. Biographical data may include biographical data associated with user 102. In embodiments, user 102 may input biographical data via a user interface of computing device 108. In some embodiments, biographical data may be obtained from a database (not pictured), for example, a health record.Biographical data may include age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data associated with a user.

[0053] In some embodiments, machine learning component 114 is further configured to process health precaution data in generating the vaccine uptake prediction 116. Health precaution data may include health precaution data associated with user 102. In some embodiments, user 102 may input health precaution data via the user interface of computing device 108 in association with the set of ratings, for example, by answering questions related to health precaution data. In some examples, health precaution data may include precaution data related to a disease associated with the vaccine. A disease may spread via direct or indirect contact, droplets, bodily fluids, fecal contamination, food and water contamination, or inset contact. Different precautionary behaviors may be utilized to protect against different diseases based on the mechanisms of spread, for example, personal protective equipment such as a masks and gloves may be worn to protect against diseases spread through direct or indirect contact, while food and water preparation techniques may be used to protect against diseases spread through food and water contamination. Health precaution data may include data related to a user’s activities and protection measures to avoid a certain disease, such as the disease associated with the vaccine. In some embodiments, health precaution data may include data related to a user’s activities and protection measures to avoid one or more diseases. In some embodiments, health precaution data associated with user 102 may include mask wearing data, hand hygiene data, social distancing data, and gathering data.

[0054] In some embodiments, vaccine uptake prediction 116 is presented to user 102, for example, on the user interface 118 of computing device 108. In some embodiments, vaccine uptake prediction 116 facilitates one or more downstream processes. For example, vaccine prediction component 110 may be further configured to send prediction 116 to one or more downstream components.

[0055] In some embodiments, vaccine uptake prediction 116 may be provided to a vaccination supply component 122. A vaccination supply component 122 may be configured to determine a vaccination supply based on the vaccine uptake prediction 116. In embodiments, vaccine uptake prediction 116 may indicate user 102 will become vaccinated (e.g., uptake, receive an administration, receive a delivery, etc.). Where vaccine uptake prediction 116 indicates an acceptance of a vaccination by user 102, the vaccination supply component 122 may determine to supply a vaccine dose to user 102. For example, the vaccination supply component 122 may transmit to a vaccination supplier to an indication to allocate a dose of vaccine to user 102. Avaccination supply may be, for example, a pharmacy, a clinic, a health care provider (e.g., a nurse, physician), a vaccine manufacturer, a supply chain manager, a public health administration. Then, vaccine doses may be efficiently allocated to receptive users.

[0056] In some embodiments, vaccine uptake prediction 116 may be provided to a vaccine mechanism component 124. A vaccination mechanism component 124 may be configured to generate a recommended vaccination mechanism based on the vaccine uptake prediction 116. In some embodiments, vaccine mechanism component 124 may recommend a vaccine mechanism based on one or more rules. In some embodiments, vaccine mechanism component 124 may recommend a vaccine mechanism based on one or more models, for example, a recommender machine learning model.

[0057] In some embodiments, vaccination mechanism component 124 may be further configured to transmit to the user 102 a communication based on the vaccine mechanism. A communication may include, in some embodiments, materials, locations of vaccine clinic, targeted messaging, etc. For example, in some embodiments, a communication may include a recommended vaccine mechanism. A recommended vaccine mechanism may include a mechanism for obtaining and delivering a vaccine, such as a location (e.g., clinic, hospital, public health center, etc.), a time (e.g., schedule an appointment), frequency (e.g., one or more additional doses, a “booster”), etc.Example Processing Device for Vaccine Uptake Predictions

[0058] FIG. 2 depicts an example processing device 200 configured to perform the methods described herein, for example, to generate a vaccine uptake prediction. In some embodiments, processing device 200 is configured to perform aspects of method 600 described with respect to FIG. 6, and method 700 in described with respect to FIG. 7. Processing device 200 may be an example of computing device 108, described with respect to FIG. 1.

[0059] Processing device 200 includes one or more processors 202. Generally processor(s) 202 may be configured to execute computer-executable instructions (e.g., software code) to perform various functions, as described herein.

[0060] Processing device 200 further includes a network interface(s) 204, which generally provides data access to any sort of data network, including personal area networks (PANs), local area networks (LANs), wide area networks (WANs), the Internet, and the like.

[0061] Processing device 200 further includes input(s) and output(s) 206, which generally provide means for providing data to and from processing device 200, such as via a connecting to computing device peripherals, including user interface peripherals.

[0062] Processing device 200 further includes a memory 210 (e.g., a non-transitory, processor readable storage medium) configured to store various types of components and data, in this example, including presenting component 220, stimuli database 240, receiving component 222, ratings database 242, determining component 224, generating component 226, training component 228, training database 248, and machine learning model 246.

[0063] Presenting component 220 is configured to present stimuli, such as a stimulus stored in stimuli database 240, for example, present a picture on a user interface, present a video on a user interface, and present a sounds via a speaker, which may be an example of an output 206. Presenting component 220 is configured to present a set of stimuli, in some embodiments, the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category. For example, presenting component 220 may be configured to present a set of stimuli to a user, such as described with respect to block 304 of FIG. 3, block 406 of FIG. 4, step 602 of FIG. 6, below.

[0064] Receiving component 222 is configured to receive ratings, for example, receive a rating via a user interface from a user and store ratings in rating database 242. For example, receiving component 222 may be configured to receive a set of ratings from a user, such as described with respect to block 306 of FIG. 3, block 406 of FIG. 4, step 604 of FIG. 6, or step 704 of FIG. 7, below. In some embodiments, receiving component 222 is further configured to receive data associated with a user, for example, biographical data and health precaution data. Biographical data may include, for example, age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data associated with a user. Health precaution data may include, for example, mask wearing data, hand hygiene data, social distancing data, and gathering data.

[0065] Determining component 224 is configured to determine a set of judgment variables, for example, based on one or more ratings in the rating database 242. For example, determining component 224 is configured to determining one or more judgment variables based on the one or more ratings in ratings database 242, such as described with respect to block 308 of FIG. 3, block 408 of FIG. 4, or step 606 of FIG. 6, below. In some embodiments, determining component 224 is configured to determine one or more subsets of ratings of the set of ratings, wherein each oneof the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli. In some embodiments, determining component 224 is configured to determine an average of each respective subset of the one or more subsets of ratings. In some embodiments, determining component 224 is configured to determine a variance of each respective subset of the one or more subsets of ratings. In some embodiments, determining component 224 is configured to determine an uncertainty of each respective subset of the one or more subsets of ratings. In some embodiments, determining component 224 is further configured to generate a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings. In some embodiments, determining component 224 is further configured to determine one or more judgment variables in the set of judgment variables based on the relative preference graph and store the one or more judgment variables in judgment variable database 244.

[0066] Generating component 226 is configured to generate a vaccine uptake prediction based on the set of judgment variables with machine learning model 246, such as described with respect to block 312 of FIG. 3, block 412 of FIG. 4, step 608 of FIG. 6. Generating component 226 is further configured to generating training data for training machine learning model 246 and store the training data in training database 248, such as described with respect to step 702 of FIG. 7.

[0067] Training component 228 is configured to train machine learning model 246 based on training data stored in training database 248, such as described with respect to step 704 of FIG.7.

[0068] Processing device 200 may be implemented in various ways. For example, processing device 200 may be implemented within on-site, remote, or cloud-based processing equipment.

[0069] Processing device 200 is just one example, and other configurations are possible. For example, in an alternative embodiment, aspects described with request to processing device 200 may be omitted, added, or substituted for alternative aspects.Example Workflow for Generating a Risk Prediction

[0070] FIG. 3 depicts example workflow 300 for generating a risk prediction, for example, a vaccine uptake prediction as described with respect to system 100 in FIG. 1.

[0071] At block 304, a set of stimuli 320 is presented to user 302, for example, via a user interface 118 of computing device 108 in FIG. 1. The set of stimuli 320 may comprise one or more stimuli, for example, one or more pictures, one or more sounds, one or more videos, or combinations thereof. For example, picture 502 in FIG. 5 depicts an example picture from a picture rating task. In some embodiments, the set of stimuli 320 may comprise one or more subsets of stimuli. The subsets may be organized by a category, for example, all stimuli in a subset are associated with the category. A category may be based on content of the stimuli, effect on the recipient (e.g., user 302), or type of stimuli (e.g., pictures, sounds, video).

[0072] In an example, a set of stimuli 320 may comprise a set of pictures. In this example, the set of pictures may comprise six subsets, wherein each subset is associated with a category. Example categories include, sports, disasters, cute animals, aggressive animals, nature, and / or food. Each subset may comprise, for example, about 8 pictures, about 6 pictures, less than 8 pictures, less than 6 pictures, greater than 8 pictures, greater than 6 pictures, etc. In some embodiments, the set of pictures may be presented in an ordered sequence or in an unordered sequence.

[0073] Each stimuli in the set of stimuli 320 may be presented individually, for example, one picture or video on a user interface, or one sound played by the computing device. One or more stimuli in the set of stimuli 320 may be presented together, for example, multiple pictures on a user interface, a picture presented on the user interface and a sound played by the computing device simultaneous, and the like.

[0074] At block 306, a set of ratings 322 is receive from user 302, for example, through the user interface. In some embodiments, a rating may comprise a continuous scale. For example, a continuous scale may be a numerical rating, such as from 0 to 5, from -3 to +3, etc. As another example, a continuous scale may be based on a sliding scale, such as a rating between “dislike very much” to “like very much” endpoints. In some embodiments, a rating may comprise a discrete scale, or one or more categories. For example, a user may select a rating from a set of rating options, such as one or more emoticons (e.g., smiley face, sad face, angry face, etc.), one or more terms (e.g., “dislike very much”, “dislike”, “neutral”, “like”, “like very much”, etc.), one or more emotions (e.g., “happy”, “sad”, “angry”, etc.). For example, rating scale 504 in FIG. 5A depicts an example discrete rating scale.

[0075] A rating may be based on the user’s preferences, emotions, and / or attention, for example, a user may rate a stimuli based on an initial response.

[0076] Each rating in the set of ratings 322 may correspond to one rating in the set of stimuli 320. In certain embodiments, user 302 may be presented with one stimuli in the set of stimuli 320 (e.g., at block 304), and user 302 inputs one rating associated with the presented stimuli. Once the rating is received (e.g., at block 306), an additional stimuli in the set of stimuli 320 is presented to user 302 and user 302 inputs a rating associated with the presented additional stimuli. The presentation of stimuli and input of ratings may continue until each stimuli in the set of stimuli 320 has been presented and a rating associated with each stimuli has been received.

[0077] In some embodiments, blocks 304-306 may together be referred to as a “rating task”.

[0078] At block 308, a set of judgment variables 324 is determined based on the set of ratings 322. In some embodiments, the set of judgment variables may be determined based on a Relative Preference Theory (RPT) framework. The RPT framework may quantify and capture judgments about the valence of judgment, for example, positive vs. negative, approach vs. avoidance, etc., as well as magnitude of judgment (e.g., intensity of rating) to describe a user’s preferences. As described herein, and at block 312, such judgment variables may be used to predict risk, including risk behavior of a user.

[0079] The set of judgment variables 324 may include, for example, loss aversion, risk aversion, loss resilience, ante, insurance, peak positive risk, peak negative risk, reward tipping point, aversion tipping point, total reward risk, total aversion risk, reward aversion tradeoff, tradeoff range, reward aversion consistency, and consistency range. For example, FIG. 5E depicts an example judgment variables and associated abbreviations. The set of judgment variables 324 may be determined based on approach / avoidance variables in the pattern of ratings in the set of ratings 322.

[0080] These approach / avoidance variables included the mean magnitude (X), variance (e.g., standard deviation) (tr), and the uncertainty of the pattern of ratings (e.g., Shannon entropy ( / / )) related to a user’s preference behavior. K reflects the average (mean) of positive ratings a subject made (K+) or negative ratings (K_) within each stimulus category. Variance in positive ratings (<T+) and variance in negative ratings (<T_ ) within each stimulus category may be determined. The Shannon entropy of positive ratings (H+) or negative ratings (H_) for stimuli within each category may also be determined. The Shannon entropy may characterize uncertainty across the set of ratings, for example, by quantifying the pattern of judgments in the set of ratings. It may also be considered a memory variable.

[0081] In some embodiments, one or more subsets of ratings may be identified from the sets of ratings based on the approach / avoidance variables. For example, a subset of ratings may be identified as a category of ratings for a given user where K=Q, for example where the user rated all stimuli in the category as neural. The Shannon entropy H, cannot be computed where K=Q because, the / / computation results in evaluating logio(^), which is undefined which is undefined. In these cases, the Shannon entropy was set to H=0 for categories in which the user rated “0” for all the stimuli.

[0082] At block 310, one or more graphs of the approach / avoidance variables may be generated. In some embodiments, one or more of the quantified variables may be plotted against one or more of the other. For example, (K, H), (K.a),graphs may be generated.(K+, H+) and (K+, <T+) are plotted separately from (K_, H_) and (K_, cr_) thereby calibrating approach / avoidance, K, to the pattern of prior judgments, H, and their variance, a. A cure may be fit to each graph. For example, FIGS. 5B-D depict various example graphs.

[0083] One example graph, graph 506 in FIG. 5B, the (A, H) curve may comprise (K+, H+) on the positive x-axis and (K_, H_) on the negative x-axis. One or more of the set of judgment variables 324 may be determined based on the (K, II) curve, including Risk Aversion, Loss Resilience, Loss Aversion, Ante, and Insurance. Loss aversion, risk aversion, loss resilience, ante, and insurance are derived from the logarithmic or power-law fit of mean keypresses (K) versus the entropy of keypresses ( / / ); this is referred to as the value function, for example depicted as graph 508 in FIG. 5B.

[0084] Loss Aversion (LA) may be the absolute value of the ratio of the linear regression slope of (log K_ , log H_~) to the linear regression slope of (log K+, log H+~) . LA may measure the degree to which an individual person overweighs losses to gains, for example due to a cognitive bias.

[0085] Risk Aversion (RA) may be determined as the ratio of the second derivative of the H+, K+~) curve to its first derivative, which also produces a curve. RA may measure the degree to which an individual prefers a likely reward in comparison to a better more uncertain reward.

[0086] Loss Resilience (LR) may be the absolute value of the ratio of the second derivative of the (K_, H_) curve to its first derivative, which also produces a curve. LR may be the degree to which an individual prefers to lose a small defined amount in comparison to losing a greater amount with more uncertainty associated with this loss.

[0087] Ante may be the value of K+when setting H+= 0. This intuitively measures the ante one needs to engage in a game of chance and models the amount of a bid an individual is willing to make to enter a game of chance (e.g., poker).

[0088] Insurance may be the value of K_ when setting H_ = 0. Insurance may measure how much insurance an individual might need against bad outcomes. Insurance mirrors the ante, but in the framework of potential losses.

[0089] Another example graph, graph 510 in FIG. 5C, the (A, <T) curve comprises (K+, <T+) on the positive x-axis and (K_, cr_) on the negative x-axis. Preference magnitude (e.g., K) may be compared with respect to variance in rewards and sanctions. For example, the (K, <T) curve may model the following question: “Would an individual prefer a dollar with probability one, or value drawn from a normal distribution with mean of two and variance of two?” One or more of the set of judgment variables 324 may be determined based on the (K, <T) curve including Peak Positive Risk, Peak Negative Risk, Reward Tipping Points, Aversion Tipping Point, Total Reward Risk, and Total Aversion Risk. Peak positive risk, peak negative risk, reward tipping point, aversion tipping point, total reward risk, and total aversion risk are derived from the quadratic fit of K versus the standard deviation of keypresses (<T); this is referred to as the limit function, for example depicted as graph 512 in FIG. 5C.

[0090] Peak Positive Risk (Peak PR) may be the value of <T+for the derivative — = 0. Peak dK+PR may represent the maximum variance for approach behavior. <T+models where increases in positive value transition from a relationship with increases in risk, to a relationship with decreases in risk. The positive apex models when variance changes from weighing against a decision to facilitating a decision.

[0091] Peak Negative Risk (Peak NR) may be the value of <T_ where the derivative = 0. dK_Peak NR may represent the maximum variance for avoidance behavior. Fike with the Peak PR, this transition point may relate to avoidance decisions.

[0092] Reward Tipping Point (Reward TP) may be the value of K+when the derivative = dK+0. Reward TP represents the rating intensity with maximum variance for approach behavior, potentially when an individual decides to approach a goal-object.

[0093] Aversion Tipping Point (Aversion TP) may be the value of K_ where the derivative= 0. Aversion TP represents the rating intensity with maximum variance for approach dK_ behavior, potentially when an individual decides to avoid a goal-object.

[0094] Total Reward Risk (Total RR) may be the area under the curve (AUC) of the first quadrant of the (K+, <T+) curve. Total RR represents the relationship between K+and <T+and may be a quantity that measures the amount of value an individual associates to positive stimuli.

[0095] Total Aversion Risk (Total AR) may be the area under the curve on the negative quadrant of the graph ofTotal AR represents the relationship between K_ and <T_ and may be a quantity that measures the amount of overall value a person associates to a negative stimulus.

[0096] As another example graph, graph 514 in FIG. 5D, the (H+, H_) curve may compare patterns in approach and avoidance judgments. One or more of the set of judgment variables 324 may be determined based on the curve (H+, H_) including Reward Aversion Tradeoff, Tradeoff Range, Reward Aversion Consistency, and Consistency Range. Risk aversion tradeoff, tradeoff range, risk aversion consistency, and consistency range are derived from the radial fit of the pattern of avoidance judgments ( / £) versus the pattern of approach judgments ( / / +); this is referred to as the tradeoff function for example depicted as graph 516 in FIG. 5D.

[0097] Reward Aversion Tradeoff (RA Tradeoff) may be the mean of the polar angles of the points in the (H+, H_) plane. RA Tradeoff may be the mean ratio of entropies or patterns in approach to avoidance behavior.

[0098] Tradeoff Range may be the standard deviation of the polar angles of the points in the (H+, H_) plane. Tradeoff Range may measure the standard deviation in the patterns of approach to avoidance behavior. This variance represents the spread for positive preferences and negative preferences across a set of potential goal-objects and is one measure of the breadth of an individual’s (or group’s) preferences.

[0099] Reward Aversion Consistency (RA Consistency) may measures the mean of the distances of the data points in the (H+, H_) curve to the origin. RA Consistency defines how individuals can have strong preferences (e.g., biases) for the same thing, reflecting conflict, or having low preferences for something, reflecting indifference . The consistency or compatibilityof approach and avoidance may be quantified, for example, how humans can both like and dislike something, or be indifferent to both its positive and negative features.

[0100] Consistency Range may measures the standard deviation of the distances of the data points in the (H+, H_ ) plane to the origin. Consistency Range measures how the points in the HH plane vary with regard to the radial distance from the origin. The variance in this radial distance will reflect how much an individual goes between having conflicting preferences and having indifferent ones.

[0101] At block 312, the set of judgment variables 324 may be provided to a machine learning model, for example machine learning component 114 in FIG. 1, to process and generate a prediction 326. In some embodiments, prediction 326 comprises a vaccine uptake prediction. In some embodiments, a vaccine update prediction may indicate a vaccination status of the user, for example, a positive vaccination status may indicate a user will become vaccinated, while a negative vaccination status may indicate a user will not become vaccinated. In some embodiments, prediction 326 comprises a medical treatment adherence prediction. For example, a positive adherence may indicate a user will accept and adhere to the medical treatment, while a negative adherence may indicate a user will not accept and / or not adhere to the medical treatment. The machine learning model may comprise, in some embodiments, a classification model, for example, a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine (SVM), or an artificial neural network.

[0102] In some embodiments, additional data may be provided to the machine learning model. For example, biographical data associated with user 302 may be utilized by the machine learning model in generating the prediction 326. Biographical data may include, for example, age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data associated with a user. As another example, health precaution data associated with user 302 may be utilized by the machine learning model in generating the prediction 326. Health precaution data may include, for example, mask wearing data, hand hygiene data, social distancing data, and gathering data.

[0103] In some embodiments, workflow 300 proceeds to block 314, wherein a recommended vaccine mechanism is generated. A recommended vaccine mechanism may include information related to vaccine health and safety, and vaccine administration. Health and safety information may include medical safety and scientific information regarding the composition, risks, effects, dosing, and / or other information related to the vaccine. In some cases, health and safetyinformation may include infection risks or precaution behaviors to reduce infection risk and / or severity. Vaccine administration information may include information related to availability, timing, or location information to obtain a vaccine. In some embodiments, a recommended vaccine information may be communicated to the user. A recommended vaccine mechanism may also include one or more mediums for communication with the user. For example, where prediction 326 indicates a user will become vaccinated, the recommended vaccine mechanism may include health and safety information related to the vaccine, and information regarding availability, such as a pharmacy location to obtain the vaccine. Furthermore, such information may be used, for example in conjunction with demographic data, to provide scientific and medical information to underserved populations, and specifically address common concerns. For example, in some populations and demographics, medical health issues may be culturally taboo, such as shame, cultural norms, and cultural biases. A recommended vaccine mechanisms may include information in a manner and of a type directed to overcome such cultural issues. As another example, where prediction 326 indicates a user will not become vaccinated, the recommended vaccine mechanism may include health and safety information related to the protective effects of the vaccine, or scientific information regarding safety and efficacy of the vaccine.

[0104] In some embodiments, a recommended vaccine mechanism may be communicated to others, for example, a physician treating the user 302, or a public health official. In some cases, the recommended vaccine mechanism may indicate information regarding the vaccine for the physical to communicate with the user 302, indicate low vaccine uptake and potential higher infection rate, or indicate vaccine uptake.

[0105] In some embodiments, a recommended vaccine mechanism may be further based on demographic information associated with a user, for example, to indicate a type of mechanism. As an example, in some cases, females may have concerns regarding a vaccine due to fertility concerns, and a recommended vaccine mechanism for a female with negative vaccination may include a communication of safety information related to the vaccine and fertility.

[0106] In some embodiments, workflow 300 proceeds to block 316, wherein a supply is determined. In some embodiments, a supply may be a dose of a vaccine. For example, a supply may be determined where the prediction 326 indicates the user 302 will become vaccinated (positive vaccination). In some embodiments, a requisition for the supply of vaccine may be transmitted to a pharmacy, a health care clinic, public health clinic and the like, for example, to obtain a vaccine dose for the user. In some embodiments, the requisition for the supply of thevaccine may be transmited to a vaccine manufacturer or vaccine distribution center, for example, to allocate a vaccine dose to the user. The requisition may include additional information, for example, demographic information of the user, a location of the user (e.g., zip code, city, or county), or health care location of the user (e.g., local hospital, pharmacy, clinic, or public health center). Such additional information may indicate a location of the supply of vaccine requisitioned for the user, for example, to facilitate administration of the vaccine to the user. For example, location data may facilitate vaccine roll out to a locality of the user, such as a local clinic, to facilitate convenient and safe vaccine administration including logistical and supply chain management, health clinic management, and ensure adherence to public health protective measures.

[0107] In some embodiments, workflow 300 proceeds to block 318, wherein a treatment protocol is generated. A treatment protocol may indicate, for example, where prediction 326 indicates negative vaccination, precaution behaviors (e.g., protective health measures), infection symptoms and treatments, and medical resources. In some cases, a treatment protocol may be communicated to the user 302. In some cases, a treatment protocol may be communicated to a health care provider, for example, to indicate higher infection rates associated with negative vaccination status or lower infection rates with positive vaccination status. A health care provider may then beneficially prepare a health care institution for potential higher or lower infection rates and allow for allocation of health care resources. Further, a health care institution may prepare for administration of vaccine supplies where positive vaccination status is indicated.

[0108] In some embodiments, the prediction 326 may indicate adherence of a user to a medical treatment. A treatment protocol may indicate, for example, treatment options and mechanisms, precautionary health behaviors, symptoms and risks, and other medical resources.

[0109] Workflow 300 provides many benefits, including, for example, prediction of vaccination uptake (e.g., positive or negative vaccination). Vaccination uptake prediction has many beneficial uses, as described herein, for example, for requisitioning of vaccine supply, improved messaging regarding vaccination health, safety, and administration, and for health care and public health system administration. Moreover, by utilizing a rating task unrelated to vaccination, workflow 300 avoids inclusion of vaccination biases in the judgment variables and the vaccination uptake prediction.

[0110] Note that FIG. 3 is just one example of a workflow, and other workflows including fewer, additional, or alternative steps are possible consistent with this disclosure.Example Workflow for Training a Machine Learning Model

[0111] FIG. 4 depicts an example workflow 400 for training a classification model for vaccine uptake prediction, for example, a classification model of machine learning component 114 in FIG. 1.

[0112] At block 404, a set of stimuli is presented to a user. As described herein, for example with respect to workflow 300, stimuli may be presented to a user via a computing device. The set of stimuli may comprise one or more stimuli, for example, one or more pictures, one or more sounds, one or more videos, or combinations thereof. In some embodiments, the set of stimuli may comprise one or more subsets of stimuli. The subsets may be organized by a category, for example, all stimuli in a subset are associated with the category. A category may be based on content of the stimuli, an intended effect on the recipient, or type of stimuli (e.g., pictures, sounds, and video).

[0113] At block 406, a set of ratings is receive from user in response to the set of stimuli, for example, ratings as described with respect to block 306 of FIG. 3. Each rating in the set of ratings may correspond to a stimuli in the set of stimuli, for example, one rating for one picture. Workflow 400 may repeat blocks 404-406 for many users.

[0114] At block 408, training data may be generated based on sets of ratings received from users at block 406. Based on the one or more sets of ratings received from one or more users a set of judgment variables may be determined. In some embodiments, the set of judgment variables may be determined such as described with respect to block 308 of FIG. 3. For example, in some embodiments, the set of judgment variables may be based on a Relative Preference Theory (RPT) framework. The RPT framework may quantify and capture judgments about the valence of judgment, for example, positive vs. negative, approach vs. avoidance, etc., as well as magnitude of judgment (e.g., intensity of rating) to describe a user’s preferences.

[0115] The set of judgment variables may include, for example, loss aversion, risk aversion, loss resilience, ante, insurance, peak positive risk, peak negative risk, reward tipping point, aversion tipping point, total reward risk, total aversion risk, reward aversion tradeoff, tradeoff range, reward aversion consistency, and consistency range. The set of judgment variables may be determined based on a set of approach / avoidance variables determined based on the sets of ratings.

[0116] As described, these approach / avoidance variables included the mean magnitude (A), variance (e.g., standard deviation) (cr), and the uncertainty of the pattern of ratings (e.g., Shannonentropy ( / / )) related to a user’s preference behavior, for example, as described with respect to block 308 in FIG. 3.

[0117] In some embodiments, one or more subsets of ratings may be identified from the sets of ratings based on the approach / avoidance variables. For example, a subset of ratings may be identified as a category of ratings for a given user where K=Q, for example where the user rated all stimuli in the category as neural. The Shannon entropy H, cannot be computed where K=Q because, the / / computation results in evaluating logio(^), which is undefined which is undefined. In these cases, the Shannon entropy was set to H=0 for categories in which the user rated “0” for all the stimuli.

[0118] In some embodiments, one or more graphs of the approach / avoidance variables may be generated. In some embodiments, one or more of the approach / avoidance variables may be plotted against one or more of the other approach / avoidance variables. For example, (K, II). (K, a), and (H+, f / _) graphs may be generated. (K+, H+) and (K+, <T+) may be plotted separately from (K_, H_) and (K_, cr_) thereby calibrating approach / avoidance, K, to the pattern of prior judgments, H, and their variance, <T. A cure may be fit to each graph.

[0119] One example graph, the (K, H) curve may comprise (K+, H+) on the positive x-axis and (K_, H_) on the negative x-axis. One or more of the set of judgment variables 324 may be determined based on the (K, H) curve, including LA, RA, LR, Ante, and Insurance.

[0120] Another example graph, the (A, <T) curve comprises (K+, <T+) on the positive x-axis and (K_, a_) on the negative x-axis. Preference magnitude (e.g., K) may be compared with respect to variance in rewards and sanctions. For example, the (A, <T) curve may model the following question: “Would an individual prefer a dollar with probability one, or value drawn from a normal distribution with mean of two and variance of two?” One or more of the set of judgment variables 324 may be determined based on the (A, <T) curve including Peak PR, Peak NR, Reward TP, Aversion TP, Total RR, and Total AR.

[0121] As another example graph, the (H+, H_) curve may compare patterns in approach and avoidance judgments. One or more of the set of judgment variables 324 may be determined based on the curve (H+, H_ ) including RA Tradeoff, Tradeoff Range, RA Consistency, and Consistency Range.

[0122] In some embodiments, training data may include additional data associated with training users, such as biographical data and / or health precaution data. Biographical data may include, for example, age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data associated with a user. Health precaution data may include, for example, mask wearing data, hand hygiene data, social distancing data, and gathering data.

[0123] At block 412, the training data may be used to train a machine learning model to generate a predicted risk. In some embodiments, the predicted risk comprises a vaccine uptake prediction. In some embodiments, the predicted risk comprises a medical treatment adherence prediction.

[0124] The machine learning model may be trained in a supervised manner. Supervised learning may be used to generate a model of relationships between one or more input features (e.g., a feature vector) and a target output, in this example, a risk prediction. Training data is labeled including input data and the desired output. Models may be trained with supervised learning to perform tasks including classification and regression. A classification task is a task to predict discrete values. A regression task is to predict continuous values. Example models trained through supervised learning include a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine (SVM), or an artificial neural network.

[0125] In some embodiments, the machine learning model may comprise a balanced random forest. A balanced random forest (bRF) is a machine learning classification model. A random forest is an ensemble machine learning model that may be used for classification tasks. A random forest ensembles a set of decision trees, in particular, bagging each decision tree in the set. Bagging involves sampling with replacement of the training data to obtain a bootstrapped training set where a given training data instance may appear more than one in the training set. Each decision tree in the set may then be trained on a different bootstrapped training set sampled from common training data. Outputs from each decision tree in the set of decision trees may be aggregated across the entire set to obtain, for example, through averaging. A random forest may have improved performance over any individual decision tree because it ensembles the outputs of the individual trees. Furthermore, the random forest may reduce variance due to bagging training techniques.

[0126] Imbalanced data may have skewed class distribution. In examples described herein, vaccination status, for example, whether an individual has been or will be vaccinated for particular vaccine, may result in imbalanced data. For example, a fully vaccinated class may be a majority class while a not fully vaccinated class may be a minority class. In other examples, a fullyvaccinate class may be a minority class, such as in particular populations, for a particular vaccination, or for a particular disease.

[0127] Where a severe class imbalance is present in training data, regular bagging may result in underperformance of the model because each bootstrapped training set (e.g., used to train an individual decision tree in the random forest) may not represent the skewed class distribution of the imbalanced training data. The minority class may be underrepresented in the bootstrapped training sets, and thus may be underrepresented or otherwise skew a prediction.

[0128] A balanced random forest model overcomes this problem and provides improved performance through under-sampling of bootstrapped training sets. In embodiments, a bootstrapped training set may be generated by randomly under-sampling the majority class of training data. Then, decision trees in the random forest are trained on the balanced bootstrapped training sets to overcome the imbalance.

[0129] In some embodiments, a number of decision trees in the balanced random forest is set to 100. In some embodiments, a number of decision trees in the balanced random forest is set to 300. In some embodiments, a number of splits is set to 10. In some embodiments, a number of splits is set to 15. In some embodiments, a number of repeats is set to 100, random state is set to 42, and a number of jobs was set to -1.

[0130] In some embodiments, one or more features may be identified to improve performance of the model. One or more inputs (e.g., features) of the machine learning model may be determined based on the importance of the feature. In examples, important features may be utilized earlier in a tree, for example, at a base or top node. Splits at base or top nodes may lead to increased information gain, as compared to splits at end nodes. Thus, model performance may be improved when important features utilized as based nodes.

[0131] Features may be determined as important based on how much a given feature decreases the impurity, or the quality of a split. Impurity may be determined based on Gini importance. For example, one or more features may include age data, income data, marital status data, employment data, ethnicity data, education level data, sex data, mask wearing data, hand hygiene data, social distancing data, or large group gathering data.

[0132] Workflow 400 provides many benefits, including, for example, for generating a model for prediction of vaccination uptake (e.g., positive or negative vaccination). Vaccination uptake prediction has many beneficial uses, as described herein, for example, for requisitioning of vaccinesupply, improved messaging regarding vaccination health, safety, and administration, and for health care and public health system administration. Moreover, by utilizing a rating task unrelated to vaccination, workflow 400 avoids inclusion of vaccination biases in the judgment variables and the vaccination uptake prediction.

[0133] Note that FIG. 4 is just one example of a workflow, and other workflows including fewer, additional, or alternative steps are possible consistent with this disclosure.Example Method for Predicting Vaccine Uptake with a Machine Learning Model

[0134] FIG. 1 depicts an example method 600 for predicting vaccine uptake, for example, as described with respect to FIG. 3, for a user.

[0135] Initially, method 600 begins at step 602 with presenting, to a user via a user interface, a set of stimuli, for example, as described with respect to block 304 of FIG. 3.

[0136] In some embodiments, the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category. In some embodiments, a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video. For example, a stimuli in the set of stimuli may be picture 106 in FIG. 1.

[0137] Method 600 proceeds to step 604 with receiving, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rating in the set of ratings corresponds with a respective stimuli in the set of stimuli, for example, ratings 104 in FIG. 1.

[0138] Method 600 then proceeds to step 606 with determining a set of judgment variables based on the set of ratings.

[0139] In some embodiments, determining the set of judgment variables based on the set of ratings comprises: determining one or more subsets of ratings of the set of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determining an average of each respective subset of the one or more subsets of ratings; determining a variance of each respective subset of the one or more subsets of ratings; and determining an uncertainty of each respective subset of the one or more subsets of ratings. In some embodiments, generating a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings, for example, as described with respect to block 308 in FIG. 3.

[0140] In some embodiments, determining one or more judgment variables in the set of judgment variables based on the relative preference graph, for example as described with respect to block 310 in FIG. 3. In some embodiments, the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

[0141] Method 600 then proceeds to step 608 with generating, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users, for example as described with respect to block 312 in FIG. 3. In some embodiments, the machine learning model comprises a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine, or an artificial neural network. In some embodiments, the machine learning model comprises a balanced random forest classifier machine learning model.

[0142] In some embodiments, the vaccine uptake prediction is further based on biographical data associated with the user. In some embodiments, the biographical data associated with the user comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

[0143] In some embodiments, the vaccine uptake prediction is further based on health precaution data associated with the user. In some embodiments, the health precaution data associated with the user comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gathering data.

[0144] In some embodiments, the vaccine uptake prediction comprises a prediction of coronavirus vaccination uptake associated with the user.

[0145] Method 600 then proceeds to step 610 with generating a vaccination mechanism recommendation based on the vaccine uptake prediction, for example as described with respect to block 314 in FIG. 3.

[0146] Method 600 then proceeds to step 612 with transmitting, to the user, the vaccination mechanism recommendation.

[0147] In some embodiments, the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of a location of a vaccine dose.

[0148] In some embodiments, the vaccine uptake prediction comprises an indication of no vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of health and safety information associated with the vaccine.

[0149] In some embodiments, the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the method 600 further comprises determining a vaccination supply based on the indication of vaccination, for example as described with respect to block 316 in FIG. 3.

[0150] In some embodiments, method 600 further comprises transmitting, to a pharmacy, a requisition for the vaccination supply.

[0151] Method 600 provides many benefits, including, for example, prediction of vaccination uptake (e.g., positive or negative vaccination). Vaccination uptake prediction has many beneficial uses, as described herein, for example, for requisitioning of vaccine supply, improved messaging regarding vaccination health, safety, and administration, and for health care and public health system administration. Moreover, by utilizing a rating task unrelated to vaccination, method 600 avoids inclusion of vaccination biases in the judgment variables and the vaccination uptake prediction. The rating task may also be completely remotely to reduce infection risk, for example, during a disease surge.

[0152] Note that FIG. 6 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.Example Method for Training a Machine Learning Model

[0153] FIG. 7 depicts an example method 700 for training a classification mode for vaccination uptake prediction, for example, as described with respect to workflow 400 of FIG. 4.

[0154] Initially, method 700 begins at step 702 with generating training data based on a sets of ratings. Step 702 comprises, presenting, to one or more users via a user interface, a set of stimuli, for example, as described with respect to block 404 in FIG. 4. In some embodiments, a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video. In some embodiments, the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

[0155] Step 702 further comprises receiving, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli for example, as described with respect to block 404 in FIG. 4.

[0156] Step 702 further comprises determining a set of judgment variables based on the sets of ratings for example, as described with respect to block 408 in FIG. 4. In some embodiments, the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

[0157] In some embodiments, method 700 further comprises generating a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

[0158] In some embodiments, the training data further comprises biographical data associated with the one or more users. In some embodiments, the biographical data associated with the one or more users comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

[0159] In some embodiments, the training data further comprises health precaution data associated with the one or more users. In some embodiments, the health precaution data associated with the one or more users comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gathering data.

[0160] Method 700 then proceeds to step 704 with training the vaccine uptake prediction model with the training data to generate a vaccine uptake prediction for example, as described with respect to block 412 in FIG. 4. In some embodiments, the machine learning model comprises a balanced random forest classifier machine learning model.

[0161] Method 700 provides many benefits, including, for example, for generating a model for prediction of vaccination uptake (e.g., positive or negative vaccination). Vaccination uptake prediction has many beneficial uses, as described herein, for example, for requisitioning of vaccine supply, improved messaging regarding vaccination health, safety, and administration, and for health care and public health system administration.

[0162] Note that FIG. 7 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.

[0163] It may be noted that one or more of the following claims utilize the terms “where,” “wherein,” or “in which” as transitional phrases. For the purposes of defining the present technology, it may be noted that these terms are introduced in the claims as an open-ended transitional phrase that are used to introduce a recitation of a series of characteristics of the structure and should be interpreted in like manner as the more commonly used open-ended preamble term “comprising.”

[0164] Having described the subject matter of the present disclosure in detail and by reference to specific embodiments, it may be noted that the various details described in this disclosure should not be taken to imply that these details relate to elements that are components of the various embodiments described in this disclosure, even in casings where a particular element may be illustrated in each of the drawings that accompany the present description. Rather, the claims appended hereto should be taken as the sole representation of the breadth of the present disclosure and the corresponding scope of the various embodiments described in this disclosure. Further, it will be apparent that modifications and variations are possible without departing from the scope of the appended claims.Example Clauses

[0165] Implementation examples are described in the following numbered clauses:

[0166] Clause 1 : A computer-implemented method of predicting vaccine uptake, the method comprising: presenting, to a user via a user interface, a set of stimuli; receiving, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rating in the set of ratings corresponds with a respective stimuli in the set of stimuli; determining a set of judgment variables based on the set of ratings; generating, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users; generating a vaccination mechanism recommendation based on the vaccine uptake prediction; and transmitting, to the user, the vaccination mechanism recommendation.

[0167] Clause 2: The computer- implemented method of clause 1, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

[0168] Clause 3 : The computer-implemented method of clause 2, wherein determining the set of judgment variables based on the set of ratings comprises: determining one or more subsets of ratings of the set of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determining an average of each respective subset of the one or more subsets of ratings; determining a variance of each respective subset of the one or more subsets of ratings; and determining an uncertainty of each respective subset of the one or more subsets of ratings.

[0169] Clause 4: The computer-implemented method of clause 3, further comprising generating a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

[0170] Clause 5: The computer-implemented method of clause 4, further comprising determining one or more judgment variables in the set of judgment variables based on the relative preference graph.

[0171] Clause 6: The computer- implemented method of clause 5, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

[0172] Clause 7 : The computer-implemented method of any one of clauses 1 - 6, wherein the machine learning model comprises a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine, or an artificial neural network.

[0173] Clause 8: The computer- implemented method of any one of clauses 1 - 6, wherein the machine learning model comprises a balanced random forest classifier machine learning model.

[0174] Clause 9: The computer- implemented method of any one of clauses 1 - 8, wherein the vaccine uptake prediction is further based on biographical data associated with the user.

[0175] Clause 10: The computer-implemented method of clause 9, wherein the biographical data associated with the user comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

[0176] Clause 11 : The computer-implemented method of any one of clauses 1 - 10, wherein the vaccine uptake prediction is further based on health precaution data associated with the user.

[0177] Clause 12: The computer- implemented method of clause 11 wherein the health precaution data associated with the user comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gathering data.

[0178] Clause 13: The computer-implemented method of any one of clauses 1 - 12, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

[0179] Clause 14: The computer-implemented method of any one of clauses 1 - 13, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the computer-implemented method further comprises determining a vaccination supply based on the indication of vaccination.

[0180] Clause 15: The computer- implemented method of any one of clauses 1 - 14, further comprising transmitting, to a pharmacy, a requisition for the vaccination supply.

[0181] Clause 16: The computer-implemented method of any one of clauses 1 - 15, wherein the vaccine uptake prediction comprises a prediction of coronavirus vaccination uptake associated with the user.

[0182] Clause 17: The computer-implemented method of any one of clauses 1 - 16, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of a location of a vaccine dose.

[0183] Clause 18: The computer-implemented method of any one of clauses 1 - 16, wherein: the vaccine uptake prediction comprises an indication of no vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of health and safety information associated with the vaccine.

[0184] Clause 19: A computer-implemented method of training a vaccine uptake prediction model, the method comprising: generating training data based on a sets of ratings, comprising: presenting, to one or more users via a user interface, a set of stimuli; receiving, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli; and determining a set of judgment variables based on the sets of ratings; and training the vaccine uptake prediction model with the training data to generate a vaccine uptake prediction.

[0185] Clause 20: The computer-implemented method of clause 19, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

[0186] Clause 21 : The computer-implemented method of any one of clauses 19 - 20, wherein determining the set of judgment variables based on the sets of ratings comprises: determining one or more subsets of ratings of each set of the sets of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determining an average of each respective subset of the one or more subsets of ratings; determining a variance of each respective subset of the one or more subsets of ratings; and determining an uncertainty of each respective subset of the one or more subsets of ratings.

[0187] Clause 22: The computer- implemented method of clause 21, further comprising generating a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

[0188] Clause 23: The computer-implemented method of any one of clauses 21 - 22, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

[0189] Clause 24: The computer-implemented method of any one of clauses 19 - 23, wherein the machine learning model comprises a balanced random forest classifier machine learning model.

[0190] Clause 25: The computer-implemented method of any one of clauses 19 - 24, wherein the training data further comprises biographical data associated with the one or more users.

[0191] Clause 26: The computer-implemented method of clause 25, wherein the biographical data associated with the one or more users comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

[0192] Clause 27 : The computer-implemented method of any one of clauses 1 - 26, wherein the training data further comprises health precaution data associated with the one or more users.

[0193] Clause 28: The computer- implemented method of clause 27, wherein the health precaution data associated with the one or more users comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gathering data.

[0194] Clause 29: The computer-implemented method of any one of clauses 19 - 28, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

[0195] Clause 30: A processing system, comprising means for performing a method in accordance with any one of Clauses 1 - 29.

[0196] Clause 31 : A non-transitory computer-readable medium comprising computerexecutable instructions that, when executed by a processor of a processing system, cause the processing system to perform a method in accordance with any one of Clauses 1 - 29.

[0197] Clause 32: A computer program product embodied on a computer-readable storage medium comprising code for performing a method in accordance with any one of Clauses 1 - 29.

[0198] Clause 33: An apparatus, comprising: comprising at least one processor, a network interface configured to communicate, via a network, with a telematics device and a first computing device, and a memory storing computer-readable instructions that, when executed by the at least one processor, cause the apparatus to perform a method in accordance with any one of Clauses 1 - 29.

Claims

CLAIMS1. A computer-implemented method of predicting vaccine uptake, the method comprising: presenting, to a user via a user interface, a set of stimuli; receiving, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rating in the set of ratings corresponds with a respective stimuli in the set of stimuli; determining a set of judgment variables based on the set of ratings; generating, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users; generating a vaccination mechanism recommendation based on the vaccine uptake prediction; and transmitting, to the user, the vaccination mechanism recommendation.

2. The computer-implemented method of claim 1, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

3. The computer- implemented method of claim 2, wherein determining the set of judgment variables based on the set of ratings comprises: determining one or more subsets of ratings of the set of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determining an average of each respective subset of the one or more subsets of ratings; determining a variance of each respective subset of the one or more subsets of ratings; and determining an uncertainty of each respective subset of the one or more subsets of ratings.

4. The computer-implemented method of claim 3, further comprising generating a relative preference graph based on the average of each respective subset of the one or more subsetsof ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

5. The computer-implemented method of claim 4, further comprising determining one or more judgment variables in the set of judgment variables based on the relative preference graph.

6. The computer-implemented method of claim 5, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

7. The computer-implemented method of claim 1, wherein the machine learning model comprises a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine, or an artificial neural network.

8. The computer- implemented method of claim 1, wherein the machine learning model comprises a balanced random forest classifier machine learning model.

9. The computer-implemented method of claim 1, wherein the vaccine uptake prediction is further based on biographical data associated with the user.

10. The computer-implemented method of claim 9, wherein the biographical data associated with the user comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

11. The computer-implemented method of claim 1, wherein the vaccine uptake prediction is further based on health precaution data associated with the user.

12. The computer- implemented method of claim 11 wherein the health precaution data associated with the user comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gathering data.

13. The computer-implemented method of claim 1, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

14. The computer-implemented method of claim 1, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the computer-implemented method further comprises determining a vaccination supply based on the indication of vaccination.

15. The computer- implemented method of claim 14, further comprising transmitting, to a pharmacy, a requisition for the vaccination supply.

16. The computer-implemented method of claim 1, wherein the vaccine uptake prediction comprises a prediction of coronavirus vaccination uptake associated with the user.

17. The computer-implemented method of claim 1, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of a location of a vaccine dose.

18. The computer- implemented method of claim 1, wherein: the vaccine uptake prediction comprises an indication of no vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of health and safety information associated with the vaccine.

19. A computer-implemented method of training a vaccine uptake prediction model, the method comprising: generating training data based on a sets of ratings, comprising: presenting, to one or more users via a user interface, a set of stimuli; receiving, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli; and determining a set of judgment variables based on the sets of ratings; and training the vaccine uptake prediction model with the training data to generate a vaccine uptake prediction.

20. The computer-implemented method of claim 19, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

21. The computer-implemented method of claim 20, wherein determining the set of judgment variables based on the sets of ratings comprises: determining one or more subsets of ratings of each set of the sets of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determining an average of each respective subset of the one or more subsets of ratings; determining a variance of each respective subset of the one or more subsets of ratings; and determining an uncertainty of each respective subset of the one or more subsets of ratings.

22. The computer-implemented method of claim 21, further comprising generating a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

23. The computer-implemented method of claim 21, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

24. The computer-implemented method of claim 19, wherein the vaccine uptake prediction model comprises a balanced random forest classifier machine learning model.

25. The computer- implemented method of claim 19, wherein the training data further comprises biographical data associated with the one or more users.

26. The computer- implemented method of claim 25, wherein the biographical data associated with the one or more users comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

27. The computer-implemented method of claim 19, wherein the training data further comprises health precaution data associated with the one or more users.

28. The computer- implemented method of claim 27, wherein the health precaution data associated with the one or more users comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gathering data.

29. The computer-implemented method of claim 19, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

30. A processing system, comprising: a memory comprising computer-executable instructions; and a processor configured to execute the computer-executable instructions and cause the processing system to: present, to a user via a user interface, a set of stimuli; receive, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rat in the set of ratings corresponds with a respective stimuli in the set of stimuli; determine a set of judgment variables based on the set of ratings; generate, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users; generate a vaccination mechanism recommendation based on the vaccine uptake prediction; and transmit, to the user, the vaccination mechanism recommendation.

31. The processing system of claim 30, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

32. The processing system of claim 31, wherein to determine the set of judgment variables based on the set of ratings comprises:determine one or more subsets of ratings of the set of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determine an average of each respective subset of the one or more subsets of ratings; determine a variance of each respective subset of the one or more subsets of ratings; and determine an uncertainty of each respective subset of the one or more subsets of ratings.

33. The processing system of claim 32, further comprising generate a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

34. The processing system of claim 33, further comprising determine one or more judgment variables in the set of judgment variables based on the relative preference graph.

35. The processing system of claim 34, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

36. The processing system of claim 30, wherein the machine learning model comprises a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine, or an artificial neural network.

37. The processing system of claim 30, wherein the machine learning model comprises a balanced random forest classifier machine learning model.

38. The processing system of claim 30, wherein the vaccine uptake prediction is further based on biographical data associated with the user.

39. The processing system of claim 38, wherein the biographical data associated with the user comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

40. The processing system of claim 30, wherein the vaccine uptake prediction is further based on health precaution data associated with the user.

41. The processing system of claim 40, wherein the health precaution data associated with the user comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gather data.

42. The processing system of claim 38, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

43. The processing system of claim 30, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the processing system further comprises determine a vaccination supply based on the indication of vaccination.

44. The processing system of claim 43, further comprising transmit, to a pharmacy, a requisition for the vaccination supply.

45. The processing system of claim 30, wherein the vaccine uptake prediction comprises a prediction of coronavirus vaccination uptake associated with the user.

46. The processing system of claim 30, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of a location of a vaccine dose.

47. The processing system of claim 30, wherein: the vaccine uptake prediction comprises an indication of no vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of health and safety information associated with the vaccine.

48. A processing system, comprising: a memory comprising computer-executable instructions; and a processor configured to execute the computer-executable instructions and cause the processing system to: generate training data based on a sets of ratings, comprising: present, to one or more users via a user interface, a set of stimuli; receive, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli; and determine a set of judgment variables based on the sets of ratings; and train a vaccine uptake prediction model with the training data to generate a vaccine uptake prediction.

49. The processing system of claim 48, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

50. The processing system of claim 49, wherein determine the set of judgment variables based on the sets of ratings comprises: determine one or more subsets of ratings of each set of the sets of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determine an average of each respective subset of the one or more subsets of ratings; determine a variance of each respective subset of the one or more subsets of ratings; and determine an uncertainty of each respective subset of the one or more subsets of ratings.

51. The processing system of claim 50, further comprising generate a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

52. The processing system of claim 50, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resiliencevariable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

53. The processing system of claim 48, wherein the vaccine uptake prediction model comprises a balanced random forest classifier machine learning model.

54. The processing system of claim 48, wherein the training data further comprises biographical data associated with the one or more users.

55. The processing system of claim 54, wherein the biographical data associated with the one or more users comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

56. The processing system of claim 48, wherein the training data further comprises health precaution data associated with the one or more users.

57. The processing system of claim 56, wherein the health precaution data associated with the one or more users comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gather data.

58. The processing system of claim 48, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

59. A non-transitory computer-readable medium comprising computer-executable instructions that, when executed by a processor of a processing system, cause the non-transitory computer-readable medium to: present, to a user via a user interface, a set of stimuli; receive, from the user via the user interface, a set of ratings associated with the set of stimuli, wherein each respective rat in the set of ratings corresponds with a respective stimuli in the set of stimuli; determine a set of judgment variables based on the set of ratings; generate, with a machine learning model, a vaccine uptake prediction based on the set of judgment variables, wherein the machine learning model is trained to determine a vaccination status of users; generate a vaccination mechanism recommendation based on the vaccine uptake prediction; andtransmit, to the user, the vaccination mechanism recommendation.

60. The non-transitory computer-readable medium of claim 59, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

61. The non-transitory computer-readable medium of claim 60, wherein to determine the set of judgment variables based on the set of ratings comprises: determine one or more subsets of ratings of the set of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determine an average of each respective subset of the one or more subsets of ratings; determine a variance of each respective subset of the one or more subsets of ratings; and determine an uncertainty of each respective subset of the one or more subsets of ratings.

62. The non-transitory computer-readable medium of claim 61, further comprising generate a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

63. The non-transitory computer-readable medium of claim 62, further comprising determine one or more judgment variables in the set of judgment variables based on the relative preference graph.

64. The non-transitory computer-readable medium of claim 63, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

65. The non-transitory computer-readable medium of claim 59, wherein the machine learning model comprises a linear regression, a random forest, a Gaussian Process, a Gaussian mixture model, a Support Vector Machine, or an artificial neural network.

66. The non-transitory computer-readable medium of claim 59, wherein the machine learning model comprises a balanced random forest classifier machine learning model.

67. The non-transitory computer-readable medium of claim 59, wherein the vaccine uptake prediction is further based on biographical data associated with the user.

68. The non-transitory computer-readable medium of claim 67, wherein the biographical data associated with the user comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

69. The non-transitory computer-readable medium of claim 59, wherein the vaccine uptake prediction is further based on health precaution data associated with the user.

70. The non-transitory computer-readable medium of claim 69, wherein the health precaution data associated with the user comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gather data.

71. The non-transitory computer-readable medium of claim 70, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

72. The non-transitory computer-readable medium of claim 59, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the non-transitory computer-readable medium further comprises determine a vaccination supply based on the indication of vaccination.

73. The non-transitory computer-readable medium of claim 72, further comprising transmit, to a pharmacy, a requisition for the vaccination supply.

74. The non-transitory computer-readable medium of claim 59, wherein the vaccine uptake prediction comprises a prediction of coronavirus vaccination uptake associated with the user.

75. The non-transitory computer-readable medium of claim 59, wherein: the vaccine uptake prediction comprises an indication of vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of a location of a vaccine dose.

76. The non-transitory computer-readable medium of claim 59, wherein: the vaccine uptake prediction comprises an indication of no vaccination associated with the user; and the vaccination mechanism recommendation comprises an indication of health and safety information associated with the vaccine.

77. A non-transitory computer-readable medium comprising computer-executable instructions that, when executed by a processor of a processing system, cause the non-transitory computer-readable medium to: generate training data based on a sets of ratings, comprising: present, to one or more users via a user interface, a set of stimuli; receive, from the one or more users via the user interface, one or more sets of ratings associated with the set of stimuli; and determine a set of judgment variables based on the sets of ratings; and train a vaccine uptake prediction model with the training data to generate a vaccine uptake prediction.

78. The non-transitory computer-readable medium of claim 77, wherein the set of stimuli comprises one or more subsets of stimuli, and each subset of the one or more subset of stimuli comprises one or more stimuli associated with a category.

79. The non-transitory computer-readable medium of claim 78, wherein determine the set of judgment variables based on the sets of ratings comprises: determine one or more subsets of ratings of each set of the sets of ratings, wherein each one of the one or more subsets of ratings corresponds to the respective stimuli in the one or more subsets of the set of stimuli; determine an average of each respective subset of the one or more subsets of ratings;determine a variance of each respective subset of the one or more subsets of ratings; and determine an uncertainty of each respective subset of the one or more subsets of ratings.

80. The non-transitory computer-readable medium of claim 79, further comprising generate a relative preference graph based on the average of each respective subset of the one or more subsets of ratings, the variance of each respective subset of the one or more subsets of ratings, and the uncertainty of each respective subset of the one or more subsets of ratings.

81. The non-transitory computer-readable medium of claim 79, wherein the set of judgment variables comprises one or more of: a loss aversion variable, a risk aversion variable, a loss resilience variable, an ante variable, an insurance variable, a peak positive risk variable, a peak negative risk variable, a total reward variable, a total aversion risk variable, or a tradeoff range variable.

82. The non-transitory computer-readable medium of claim 77, wherein the vaccine uptake prediction model comprises a balanced random forest classifier machine learning model.

83. The non-transitory computer-readable medium of claim 77, wherein the training data further comprises biographical data associated with the one or more users.

84. The non-transitory computer-readable medium of claim 83, wherein the biographical data associated with the one or more users comprises at least one or more of: age data, income data, marital status data, employment data, ethnicity data, education level data, or sex data.

85. The non-transitory computer-readable medium of claim 77, wherein the training data further comprises health precaution data associated with the one or more users.

86. The non-transitory computer-readable medium of claim 85, wherein the health precaution data associated with the one or more users comprises at least one or more of: mask wearing data, hand hygiene data, social distancing data, or large group gather data.

87. The non-transitory computer-readable medium of claim 77, wherein a stimuli in the set of stimuli comprises one or more of: an image, a sound, or a video.

Images