Individual decision-making model construction method and system based on contract theory in health crowd intelligence perception scenario

Abstract

The present invention discloses a method and system for constructing an individual decision model based on contract theory in a health crowd intelligence perception scenario. The method is based on a health crowd intelligence perception system. The method includes: constructing an individual utility function based on the intrinsic motivation of the individual to maintain good health, the cost of using technology, the risk borne, the personalized medical guidance service provided by the platform, and the influence of compatibility and complexity; extracting the relationship between the intrinsic motivation of the individual to maintain good health, the cost of using technology, the risk borne, the personalized medical guidance service provided by the platform and the disease level in the utility function, and establishing constraints; constructing an individual decision model based on the constructed individual utility function and constraints. The present invention constructs a decision model based on key factors such as the intrinsic motivation of the individual to maintain good health and the cost of using technology, and can evaluate the individual's acceptance of HCS technology, and when the decision factors change, the individual decision results are different.

Description

Technical Field

[0001] The present invention belongs to the field of health and medical technology, and in particular relates to a method and system for constructing an individual decision-making model based on contract theory in a health crowd intelligence perception scenario. Background Art

[0002] While the digital age has brought numerous conveniences to people's lives, it has also frequently placed people in the decision-making process of whether to accept and use technology. How individuals can make decisions tailored to their needs when faced with technology and enjoy the benefits it brings has become a hot topic of research. In the healthcare sector, over 70% of health care system expenditures and 80% of mortality rates are related to chronic diseases. Health crowd sensing (HCS) technology, as a service paradigm for providing various healthcare services to individuals or organizations, has made healthcare affordable and accessible to the general public, even in resource-scarce settings, due to its rapid, efficient, and low-cost features. Since 2010, research on health crowd sensing technology has made a series of advances, including HealthKit (2019) and Google Fit (2019), which were used to create services related to the Apple and Google ecosystems, respectively. Researchers have also built HCS platforms such as Allergymap, self-management interventions for schizophrenia users, digital diabetes coaches, and obesity prevention. These platforms monitor and improve patients' quality of life and stratify patients, enabling them to adhere to medication, regulate their mood, sleep well, and socialize.

[0003] Despite this, studies have shown that HCS is still not widely used in the healthcare field. Therefore, whether individuals are willing to accept and use this technology has become an urgent issue to be addressed. Summary of the Invention

[0004] Aiming at the application of HCS technology in the field of health care, the present invention proposes a method and system for constructing an individual decision-making model based on contract theory in the scenario of health crowd intelligence perception.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In one aspect, the present invention proposes a method for constructing an individual decision-making model based on contract theory in a health crowd sensing scenario. The method is based on a health crowd sensing system (HCS). The health crowd sensing system includes individuals, platforms, and users. Individuals refer to individuals who have not yet used health crowd sensing technology, i.e., HCS technology; platforms refer to technology providers that provide personalized medical guidance services to users who use HCS technology; and users refer to individuals who have already used HCS technology. The method includes:

[0007] Construct an individual utility function based on the individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, and the impact of compatibility and complexity;

[0008] Extract the relationship between the intrinsic motivation of individuals to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform and the disease level in the utility function, and establish constraints;

[0009] An individual decision-making model is constructed based on the constructed individual utility function and constraints.

[0010] Furthermore, the utility function is:

[0011] VDec=cS+wrm-com (7)

[0012] stVDec>0

[0013] Among them, VDec represents the utility function used by individuals to decide whether to use HCS technology, w represents the intrinsic motivation of individuals to maintain good health, m represents the cost of using HCS technology, r represents the risk that individuals bear when using HCS technology, cS represents the personalized medical guidance service provided by the platform, and com represents the negative impact of compatibility attributes and complexity attributes on individuals' use of HCS technology.

[0014] Furthermore, the constraints are:

[0015] w=λ w l if l1 <l2<...<l L then w1 <w2<...<w L (8)

[0016]

[0017]

[0018] m=λ m l if l1 <l2<...<l L then m1 <m2<...<m L (11)

[0019] Moreover, in the era of digital transformation, health crowd intelligence perception as a technology has the same degree of negative impact on all individuals in terms of compatibility attributes and complexity attributes.

[0020] Among them, λ w represents the conversion coefficient between intrinsic motivation and disease level, L represents the number of disease levels, and λ cS represents the conversion coefficient between an individual's access to services and their disease level, λr is the conversion coefficient between the privacy risk borne by an individual and his / her disease level, λ m It is the conversion coefficient between the cost borne by an individual and his / her disease level.

[0021] Furthermore, the decision model is:

[0022]

[0023] Furthermore, after building the individual decision model, it also includes:

[0024] The decision model is optimized as follows:

[0025]

[0026] Furthermore, after building the individual decision model, it also includes:

[0027] Individuals decide whether to accept and use HCS technology based on the decision model. If the decision result is to accept and use the technology, they proceed to the next step; otherwise, they do not use the technology.

[0028] Individuals sign the disease-level service contract published by the platform, become platform users, and begin using HCS technology;

[0029] Users collect in-depth data related to their personal health through the smart terminals they carry, and upload the collected data to the platform;

[0030] The platform provides users with personalized diagnosis and treatment guidance services based on the contracts they sign and the data they upload.

[0031] Another aspect of the present invention proposes a system for constructing an individual decision-making model based on contract theory in a health crowd sensing scenario. The system is based on a health crowd sensing system (HCS). The health crowd sensing system includes individuals, platforms, and users. Individuals refer to individuals who have not yet used health crowd sensing technology, i.e., HCS technology; platforms refer to technology providers that provide personalized medical guidance services to users who use HCS technology; and users refer to individuals who have already used HCS technology. The system includes:

[0032] A utility function building module is used to construct an individual utility function based on the individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, and the impact of compatibility and complexity;

[0033] The constraint construction module is used to extract the relationship between the intrinsic motivation of individuals to maintain good health, the cost of using technology, the risks they bear, the personalized medical guidance services provided by the platform and the disease level in the utility function, and establish constraint conditions;

[0034] The decision model building module is used to build an individual decision model based on the constructed individual utility function and constraints.

[0035] Furthermore, the utility function is:

[0036] VDec=cS+wrm-com (7)

[0037] stVDec>0

[0038] Where VDec represents the utility function used by individuals to decide whether to use HCS technology, w represents the intrinsic motivation of individuals to maintain good health, m represents the cost of using HCS technology, r represents the risk that individuals bear when using HCS technology, cS represents the personalized medical guidance service provided by the platform, and com represents the negative impact of compatibility attributes and complexity attributes on individuals' use of HCS technology.

[0039] The constraints are:

[0040] w=λ w l if l1 <l2<...<l L then w1 <w2<...<w L (8)

[0041]

[0042]

[0043] m=λ m l if l1 <l2<...<l L then m1 <m2<...<m L (11)

[0044] Moreover, in the era of digital transformation, health crowd intelligence perception as a technology has the same degree of negative impact on all individuals in terms of compatibility attributes and complexity attributes.

[0045] Among them, λ w represents the conversion coefficient between intrinsic motivation and disease level, L represents the number of disease levels, and λ cS represents the conversion coefficient between an individual's access to services and their disease level, λ r is the conversion coefficient between the privacy risk borne by an individual and his / her disease level, λ m The conversion factor between the cost borne by an individual and the level of his or her disease;

[0046] The decision model is:

[0047]

[0048] Furthermore, it also includes:

[0049] The model optimization module is used to optimize the decision model in the following ways:

[0050]

[0051] Furthermore, it also includes:

[0052] The decision module is used by individuals to decide whether to accept and use HCS technology based on the decision model. If the decision result is to accept and use the technology, the next step is taken; otherwise, the technology is not used;

[0053] The contract signing module is used for individuals to sign the disease level service contract issued by the platform, become a platform user, and start using HCS technology;

[0054] The data collection and upload module is used by users to collect in-depth data related to their individual health through the smart terminal they carry, and upload the collected data to the platform;

[0055] The service provision module is used by the platform to provide personalized diagnosis and treatment guidance services to users based on the contracts signed and the data uploaded.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] Digital transformation has brought about tremendous social changes, and individuals are constantly faced with the challenge of accepting and using emerging new technologies (HCS technologies). This invention combines innovation diffusion theory and contract theory for the first time to construct a decision-making model to meet the above challenges. The decision-making model is constructed based on key factors such as the individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, compatibility, and the impact of complexity. By analyzing the cost-effectiveness of all parties in health crowd intelligence perception technology, the question of whether individuals accept and use health crowd intelligence perception technology is converted into a cost-effectiveness question. The experimental link uses numerical analysis methods to verify the effectiveness of the decision-making model. The decision-making model proposed in this invention enriches the individual decision-making process of the general public without prior experience in the existing individual decision-making theory system, and provides a theoretical basis and practical support for further research on the impact of digital transformation on whether individuals use HCS technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a schematic diagram of the architecture of a health crowd intelligence perception system based on a method for constructing an individual decision-making model based on contract theory in a health crowd intelligence perception scenario according to an embodiment of the present invention;

[0059] Figure 2This is a basic flow chart of a method for constructing an individual decision-making model based on contract theory in a health crowd intelligence perception scenario according to an embodiment of the present invention;

[0060] Figure 3 is the utility function curve of individuals with different disease levels when λ1>0,λ2>0 and λ1>λ2;

[0061] Figure 4 is the utility function change curve of individuals with different disease levels when λ1>0,λ2>0 and λ1≤λ2;

[0062] Figure 5 When λ1>0,λ2<0 ​​and VDec max =λ1+λ2-com>0 when the utility function change curve of users with different disease levels;

[0063] Figure 6 is the utility function change curve of users with different disease levels when λ1<0,λ2>0;

[0064] Figure 7 is the utility function change curve of users with different disease levels when λ1=0orλ2=0;

[0065] Figure 8 A schematic diagram of a system architecture for building a decision model for individuals to adopt new technologies according to an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments:

[0067] A method for constructing an individual decision model based on contract theory in a health crowd intelligence perception scenario is based on the health crowd intelligence perception system HCS (such as Figure 1 As shown), the health crowd intelligence perception system includes individuals, platforms and users, where individuals refer to individuals who have not yet used the health crowd intelligence perception technology, i.e., HCS technology; platforms refer to technology providers that provide personalized medical guidance services to users who use HCS technology; and users refer to individuals who have already used HCS technology;

[0068] In the health crowd-sensing system, while providing personalized diagnosis and treatment guidance services to users, the platform also extracts group intelligence from in-depth individual data collected from users. This extracted intelligence is used for common disease diagnosis and treatment research and to better provide personalized services to users. Therefore, it is the data uploaded by users that creates value for the platform. Users upload data according to the signed contract to obtain personalized diagnosis and treatment guidance services, with the goal of obtaining the services provided by the platform.

[0069] In order to ensure the sustainability of HCS services, it is assumed that both the platform and users are rational, that is, the system will only operate when their revenue is greater than their costs, so rational constraints must be met.

[0070] Definition 1: Rational Constraints. The platform will only provide personalized diagnosis and treatment services to users when their utility is greater than zero. Users will only upload their in-depth health data when their utility is greater than zero.

[0071] Obviously, the platform's utility Vp data It is composed of revenue generated by user data minus its operating and service costs.

[0072] Right now

[0073] Vp data =θ(p)-cOpe-cSer (1)

[0074]

[0075] Where p represents data performance, i.e., the ability of all user-uploaded data to generate revenue (data of a certain size can be used to train disease diagnosis and treatment models), θ is the conversion parameter for data performance into revenue, and σ is the data performance threshold set by the platform based on actual conditions. When the platform achieves data performance greater than or equal to this value, the data is valuable and can be converted into revenue; otherwise, the data is unprofitable. cOpe represents the platform's operating cost, i.e., the cost of receiving, storing, and processing user data. cSer represents the service cost of providing services to platform users.

[0076]

[0077] p depends on the different disease levels of users l i The amount of data transmitted i The sum of , η and γ are weight factors that can be adjusted according to platform needs. N is the total number of users, N = {1, 2, ..., N}, and they belong to L disease levels according to the contracts they signed (L is the number of disease levels), L = {1, 2, ..., L}, l∈L.

[0078] cOpe=c·N (3)

[0079] For ease of representation, it is assumed that the operating cost c of each user is equal, so cOpe grows linearly with the number of users.

[0080] In actual operation, the platform may need data from users with different disease levels at different stages. The service content provided to users with different disease levels can be adjusted to attract users with corresponding disease levels to join and obtain their data. Specifically, when the platform hopes that more users with high disease levels will join, the services available to users with high disease levels can be increased; when the platform hopes that more users with low disease levels will join, the services available to users with low disease levels can be increased. As an implementable method, in this embodiment, the platform expects more users with low disease levels to join, so the services provided by the platform to users with different disease levels are negatively correlated with their disease levels, that is, the lower the disease level, the more services they obtain. The total service cost cSer of the platform is:

[0081]

[0082]

[0083] Among them, α is the value conversion coefficient between disease level and service cost. Since the platform provides the same service content to users with the same disease level, the cost is the same.

[0084] From (2), we can see that when N tends to infinity, lim N→∞ p = 1 indicates a data performance limit. That is, once the number of users reaches a certain threshold, newly uploaded data no longer contributes to platform revenue (data performance limits exist). This is consistent with the reality that when model accuracy reaches a certain threshold, newly added data no longer improves it. However, the platform's operating costs (cOpe) and the service costs (cSer) for providing services to users will continue to increase with the number of users. Therefore, to maximize profits, the platform will no longer accept new users or reduce operating and service costs.

[0085] This example assumes that in the era of digital transformation, the platform needs a large number of users to join in order to obtain sufficient data to extract swarm intelligence (training models), which is far from reaching the limit of data performance. The data performance p may be less than σ, or the revenue brought by the data performance is less than the cost, making Vp data Less than zero, the platform can achieve sustainable operation by charging advertising fees ξ and provide services to any number of users. Therefore, the platform utility function is rewritten as:

[0086]

[0087] For user i, the utility function is the service provided by the platform minus the privacy leakage risk r i and usage cost m iPrivacy leakage risk refers to the risk of leakage of personal identity, location, data and other private information when users submit data. Usage cost refers to the time, energy consumption, computing power, traffic and other costs required for users to collect and upload health-related longitudinal data. i for:

[0088] Vu i =cS i -r i -m i i∈N (6)

[0089] Among them, cS i It means that users receive services from the platform (i.e., income). Obviously, the cost of the platform to provide services to users is cSer i With cS i equivalence.

[0090] Note: Due to the strong privacy characteristics of health data, the platform does not know information such as the individual's disease level before using HCS technology. The individual only knows their own information and does not know the information of the platform and users. Therefore, when designing the contract, the platform decides the content of the services provided to individuals with different disease levels based on its own needs.

[0091] This example focuses on the decision-making process of individuals joining the system. Because HCS is a new technology, individuals are unfamiliar with its services. Furthermore, data collection and uploading incurs costs and poses privacy risks, leading to uncertainty about whether to accept and use the technology. In this section, we will extract and discuss the key factors that shape individual utility functions.

[0092] For individuals, their utility function is mainly affected by the intrinsic motivation w to maintain good health, the cost of using technology m, the risk r, and the relative advantages, compatibility, and complexity of technology as pointed out by the Diffusion of Innovation Theory (DIT).

[0093] Intrinsic motivation (w) profoundly influences individual behavior, driving their acceptance and use of HCS technology. The relative advantages of technology, reflected in the personalized medical guidance services (cS) provided by the platform, such as anomaly detection and health intervention, also contribute to individual adoption of HCS technology. Compatibility refers to the degree to which users experience similarities between accessing services through technology and traditional methods. Complexity refers to the difficulties users face in understanding and using the technology. Clearly, both compatibility and complexity negatively impact user adoption of technology, collectively expressed as com.

[0094] Therefore, if Figure 2As shown in Figure 2, a method for constructing a decision model for individuals to adopt new technologies includes:

[0095] Construct an individual utility function based on the individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, and the impact of compatibility and complexity;

[0096] Extract the relationship between the intrinsic motivation of individuals to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform and the disease level in the utility function, and establish constraints;

[0097] An individual decision-making model is constructed based on the constructed individual utility function and constraints.

[0098] Furthermore, the utility function is:

[0099] VDec=cS+wrm-com (7)

[0100] stVDec>0

[0101] Among them, VDec represents the utility function used by individuals to decide whether to use HCS technology, w represents the intrinsic motivation of individuals to maintain good health, m represents the cost of using HCS technology, r represents the risk that individuals bear when using HCS technology, cS represents the personalized medical guidance service provided by the platform, and com represents the negative impact of compatibility attributes and complexity attributes on individuals' use of HCS technology.

[0102] In this embodiment, decision-making refers to an individual's personal judgment or choice regarding whether to accept and use HCS technology. According to continuous finite comparison decision theory, individuals select a satisfactory solution from a selection of available options based on the various factors they can grasp in the decision-making environment. Therefore, based on the individual's utility function, the core influencing factors common to each factor are extracted to construct a decision-making model and perform optimization.

[0103] Furthermore, the constraints are:

[0104] w=λ w l if l1 <l2<...<l L then w1 <w2<...<w L (8)

[0105]

[0106]

[0107] m=λ m l if l1 <l2<...<l L then m1 <m2<...<mL (11)

[0108] Moreover, in the era of digital transformation, health crowd intelligence perception as a technology has the same degree of negative impact on all individuals in terms of compatibility attributes and complexity attributes.

[0109] Among them, λ w represents the conversion coefficient between intrinsic motivation and disease level, L represents the number of disease levels, and λ cS represents the conversion coefficient between an individual's access to services and their disease level, λ r is the conversion coefficient between the privacy risk borne by an individual and his / her disease level, λ m It is the conversion coefficient between the cost borne by an individual and his / her disease level.

[0110] Specifically, according to the individual utility function equation (7), the individual's perceived intrinsic motivation w, the cost of using technology m, the risk assumed r, and the service provided by the platform cS are all closely related to the individual's needs, namely the disease level, which is consistent with the key factor of the platform's contract provision, the disease level. Therefore, the specific relationship between the key factors in the utility function and the disease level is extracted, the following constraints are established, and an individual decision-making model is constructed.

[0111] Definition 2: There is a linear increasing relationship between an individual's disease level and their intrinsic motivation to accept and use HCS technology. That is, in order to improve their quality of life, health and life expectancy, individuals with higher disease levels have stronger intrinsic motivation to improve their health through monitoring and management.

[0112] w=λ w l if l1 <l2<...<l L then w1 <w2<...<w L (8)

[0113] where λ w It is the conversion coefficient between intrinsic motivation and disease level (also used as the value coefficient of intrinsic motivation).

[0114] Definition 3: There is a linear decreasing relationship between the services provided by the platform to individuals with different disease levels and their disease levels, that is, the service content provided by the platform decreases as the disease level increases (for example, individuals with low disease levels can enjoy other services in addition to necessary diagnosis and treatment guidance).

[0115]

[0116] where λ cS It is the conversion coefficient between the service received by an individual and his / her disease level (also used as the value coefficient of the service received by an individual). m >cS n , This is because individuals with low disease levels receive more services than individuals with high disease levels. However, since individuals with different disease levels require different personalized diagnosis and treatment guidance, if individuals with high disease levels choose services with low disease levels, their medical guidance needs cannot be met. Therefore, individuals will only choose to sign contracts that match their disease levels.

[0117] Theorem 4: There is a linear decreasing relationship between the concerns of individuals with different disease levels about privacy risks and their disease levels. That is, the higher the disease level of an individual, the higher the privacy risk he or she is willing to bear in order to obtain the necessary personalized diagnosis and treatment services, and the less concerned he or she is about privacy risks.

[0118]

[0119] where λ r It is the conversion coefficient between the privacy risk borne by an individual and his / her disease level (also used as the value coefficient of the privacy risk borne by an individual).

[0120] Theorem 5: There is a linearly increasing relationship between an individual's disease severity and the cost of using HCS technology. That is, the higher the individual's disease severity, the greater the cost of collecting and uploading data. This is because as the disease severity increases, the individual needs to submit more detailed longitudinal data, which incurs higher costs.

[0121] m=λ m l if l1 <l2<...<l L then m1 <m2<...<m L (11)

[0122] where λ m It is the conversion coefficient between the cost of use borne by an individual and his or her disease level (also used as the value coefficient of the individual cost of use).

[0123] Theorem 6: In the era of digital transformation, health crowd intelligence perception as a technology has the same degree of negative impact on all individuals in terms of compatibility attributes and complexity attributes.

[0124] Theorem 7: As a rational individual, when its utility is greater than zero, it will choose to accept and use health crowd intelligence perception services.

[0125] Furthermore, according to equation (7) and the above constraints, the individual decision model is:

[0126]

[0127] Obviously, the service content and individual risk in the decision function are inversely proportional to the individual disease level, while the intrinsic motivation and usage cost are directly proportional to the individual disease level. Therefore, in order to facilitate the deduction of their impact on the decision-making results, the service content and risk are combined into λ1, and the usage cost and individual disease level are combined into λ2.

[0128] Then the decision model is optimized as (13).

[0129]

[0130] Furthermore, after building the individual decision model, it also includes:

[0131] Individuals decide whether to accept and use HCS technology based on the decision model. If the decision result is to accept and use the technology, they proceed to the next step; otherwise, they do not use the technology.

[0132] Individuals sign the disease-level service contract published by the platform, become platform users, and begin using HCS technology;

[0133] Users collect in-depth data related to their personal health through the smart terminals they carry, and upload the collected data to the platform;

[0134] The platform provides users with personalized diagnosis and treatment guidance services based on the contracts they sign and the data they upload.

[0135] Next, by deriving the first-order and second-order derivatives of VDec with respect to l, we can obtain the relationship between disease level and utility function, that is, how individuals with different disease levels make different decisions based on the utility function. The first-order derivative of VDec with respect to l is derived as

[0136]

[0137] Based on the first-order derivative, we can derive the second-order derivative with respect to l.

[0138]

[0139] Next, we discuss the values ​​of λ1 and λ2, specifically how individuals with different disease levels make decisions when the key factors influencing decision-making change. This analysis is divided into the following five parts.

[0140] 1. When λ1 > 0 and λ2 > 0, setting the first-order derivative equal to zero to derive the extreme point yields the value of l as Equation (16), and setting the second-order derivative equal to zero yields the value of l as Equation (17). Clearly, (17) is greater than zero, so the obtained extreme point is the minimum. Substituting Equation (16) into the utility function (13) yields the minimum (18). Sections 1.1 and 1.2 discuss the cases of λ1 > λ2 and λ1 ≤ λ2, respectively.

[0141]

[0142]

[0143]

[0144] 1.1 From the first-order derivative of the utility function (14), we can see that when λ1>λ2, the value of the first-order derivative changes from negative to positive, and the corresponding utility function should first decrease and then increase. Therefore, the minimum value of the utility function determines the range of disease level. Next, we discuss the case of the minimum value of the utility function.

[0145] When VDec min When it is >0, the utility is greater than zero regardless of the individual's disease level, so the individual will accept and use HCS technology.

[0146] When VDec min <0, and when the benefit value is greater than zero, the individual accepts and uses the HCS technology, and the disease level value range is (19) or (20).

[0147]

[0148]

[0149] When VDec min = 0, the disease level of the individual when receiving and using HCS technology is (21)

[0150]

[0151] 1.2 From the first-order derivative of the utility function (14), we can see that when λ1≤λ2, the first-order derivative gradually increases from zero, so the utility function is a monotonically increasing function and takes the minimum value vDec at l=1. min =λ1+λ2-com, therefore, the disease level when the user accepts and uses HCS technology is (22).

[0152]

[0153] From the above deduction, we can see that when λ1>0,λ2>0, there will always be individuals who accept and use HCS technology. cS ,λ R The relationship between λ2andλ w ,λ m From the relationship, we can see that when the value of the services provided by the platform to an individual is greater than the value of the risks he needs to bear, and the value of the individual's intrinsic motivation to accept and use HCS technology is greater than the cost of using the technology, there will always be individuals who accept and use the technology.

[0154] 2. When λ1<0,λ2<0, the utility function (13) is less than zero, so individuals will not accept HCS technology. cS ,λ R The relationship between λ2andλ w ,λ m It can be seen from the relationship that when the value of the service provided by the platform to an individual is less than the value of the risk required, and the value of the individual's intrinsic motivation to accept and use HCS technology is less than the cost of using the technology, no individual will accept and use the technology.

[0155] 3. When λ1>0,λ2<0, the first-order derivative of the utility function is less than zero, so the utility function is monotonically decreasing, and the disease level when the individual accepts and uses HCS technology is (23). In-depth analysis of λ1andλ cS ,λ R The relationship between λ2andλ w ,λ m From the relationship, we can see that when the value of the services provided by the platform to individuals is greater than the value of the risks they need to bear, and the value of the individual's intrinsic motivation to accept and use HCS technology is less than the cost of using the technology, some individuals will accept and use the technology.

[0156]

[0157] 4. When λ1<0,λ2>0, the first-order derivative of the utility function is greater than zero, so the utility function is monotonically increasing, and the disease level when the individual accepts and uses HCS technology is (24). In-depth analysis of λ1andλ cS ,λ R The relationship between λ2andλ w ,λ m From the relationship, we can see that when the value of the services provided by the platform to individuals is less than the value of the risks they need to bear, and the value of the individual's intrinsic motivation to accept and use HCS technology is greater than the cost of using the technology, some individuals will accept and use the technology.

[0158]

[0159] 5. Obviously, when λ1=0,λ2=0, the utility function is less than zero, and individuals are only negatively affected by the technical compatibility and complexity, so they will not accept HCS technology. Next, we discuss the cases of λ1=0orλ2=0 respectively.

[0160] 5.1 When λ1=0,λ2>0, the first-order derivative of the utility function is greater than zero, so the utility function is monotonically increasing, and the disease level when the individual accepts and uses HCS technology is (25).

[0161]

[0162] 5.2 When λ1=0,λ2<0, the first derivative of the utility function is less than zero, the utility function is monotonically decreasing and VDec max =λ2-com<0, so individuals will not accept health crowd-sensing services. That is, when the value of the services provided by the platform to individuals is equal to the value of the risks they need to bear, and the value of the individual's intrinsic motivation to accept and use HCS technology is less than the cost of using the technology, no individual will accept and use the technology.

[0163] 5.3 When λ1>0,λ2=0, the first-order derivative of the utility function is less than zero, so the utility function is monotonically decreasing. The disease level when the individual accepts and uses HCS technology is (26). When the value of the service provided by the platform to the individual is greater than the value of the risk required, and the value of the individual's intrinsic motivation to accept and use HCS technology is equal to the cost of using the technology, some individuals accept and use the technology.

[0164]

[0165] 5.4 When λ1<0,λ2=0, the utility function (13) is rewritten as Therefore, individuals will not accept health crowd-sensing services. When the value of the services provided by the platform to individuals is less than the value of the risks they need to bear, and the value of the individual's intrinsic motivation to accept and use health crowd-sensing technology is equal to the cost of using the technology, no user will accept and use the technology.

[0166] The above derivation process details the role of key factors in the decision-making model, including relative technology advantages (service content), intrinsic motivation, cost, risk, technology compatibility, and complexity. It also demonstrates the dynamic relationship between these key factors and the decision-making outcomes of individuals with different disease levels. The following experimental phase will verify this with a case study.

[0167] To verify the variability of individual decision-making outcomes when key factors in the decision-making model are varied, this example employs numerical analysis. The different values ​​of key factors in the experiment were set based on the derivation of the decision-making model, and the user's disease level L was set to {1, 2, 3}. The software environment was Python 3.8.

[0168] To ensure visibility of the experimental results, the following figures only present the individual acceptance and use of HCS technology. Numerical experiments examining individuals' non-acceptance of HCS technology have also been experimentally verified. The horizontal axis in the figure represents the individual's disease level, L, and the vertical axis represents the individual's utility, VDec. Detailed experimental analysis is as follows.

[0169] like Figure 3 As shown in the figure, when λ1>0, λ2>0, and λ1>λ2, there are four cases for the utility function curves of individuals with different disease levels. Next, the values ​​of the key factors in the decision model are set for the four cases respectively, and the corresponding decision results of different individuals are given.

[0170] 1. Let λ cS =30,λ w =24,λ r =10,λ m =14, com=20, the first line (from top to bottom) is obtained, that is, the minimum value of the utility function is greater than zero. Therefore, regardless of the individual's disease level, they will accept and use HCS technology (that is, individuals with disease levels of 1, 2, and 3 all accept and use HCS technology).

[0171] 2. Let λ cS =36,λ w =21,λ r =12,λ m =10, com = 33, the second line (from top to bottom) is obtained, indicating that the minimum value of the utility function is less than zero, and λ1+λ2-com>0. When VDec = 0, l takes the values ​​of 1.24 and 1.76, respectively, i.e., l∈[1,1.24) and l∈(1.76,∞). Therefore, individuals with disease levels 1, 2, and 3 all accept and use HCS technology.

[0172] 3. Let λ cS =48,λ w =21,λ r =26,λ m =10, com = 33, the third line (from top to bottom) is obtained, that is, the minimum value of the utility function is less than zero, and λ1+λ2-com = 0, and when VDec = 0, l takes the values ​​of 1 and 2 respectively, that is, l∈(2,∞). Therefore, individuals with disease level 3 accept and use HCS technology.

[0173] 4. Let λ cS =36,λ w =21,λ r =16,λ m =10, com = 33, the fourth line (from top to bottom) is obtained, indicating that the minimum value of the utility function is less than zero, and λ1+λ2-com < 0. When VDec = 0, l takes values ​​of 0.84 and 2.16, respectively, i.e., l∈(2.16,∞). Therefore, individuals with disease level 3 accept and use HCS technology.

[0174] like Figure 4 As shown in Figure 2, when λ1>0,λ2>0, and λ1≤λ2, the utility function change curves of individuals with different disease levels. The specific analysis is:

[0175] 1. Let λ cS =48,λ w =25,λ r =37,λ m The first line (from top to bottom) is obtained when λ = 5 and com = 25, indicating that the minimum value of the utility function is greater than zero, and λ1 + λ2 - com > 0, i.e., l∈[1,∞). Therefore, individuals with disease levels 1, 2, and 3 all accept and use HCS technology.

[0176] 2. Let λ cS =48,λ w =25,λ r =37,λ m =10, com = 30, the second line (from top to bottom) is obtained, indicating that the minimum value of the utility function is less than zero and λ1+λ2-com < 0. When VDec = 0, l takes values ​​of 0.48 and 1.52, respectively. After discarding values ​​less than 1, l∈(1.52,∞). Therefore, individuals with disease levels 2 and 3 accept and use HCS technology.

[0177] like Figure 5 As shown, when λ1>0,λ2<0, and VDec max Utility function change curve of users with different disease levels when λ1+λ2-com>0.

[0178] Let λ cS =48,λ w =15,λ r =8,λ m =25,com=3, we get Figure 5 The line in ( ) indicates that the maximum value of the utility function is greater than zero. When VDec = 0, l takes values ​​of -2.16 and 1.86, respectively. After discarding values ​​less than 1, l∈[1,1.86]. Therefore, users with a disease level of 1 accept and use HCS technology.

[0179] like Figure 6 As shown in Figure 2, when λ1<0,λ2>0, the utility function change curves of users with different disease levels are divided into two cases.

[0180] 1. Let λ cS =15,λ w =35,λ r =25,λ m =20,com=2, and VDec min =λ1+λ2-com>0, the first line (from top to bottom) is obtained, that is, the minimum value of the utility function is greater than zero, so individuals with disease levels of 1, 2, and 3 all accept and use HCS technology.

[0181] 2. Let λ cS =15,λ w =25,λ r =25,λ m =15,com=6, and VDec min =λ1+λ2-com<0, the second line (from top to bottom) is obtained, indicating that the minimum value of the utility function is less than zero. When VDec=0, l takes the values ​​of -0.74 and 1.34, respectively. After discarding values ​​less than 1, l∈(1.34,∞). Therefore, individuals with disease levels 2 and 3 accept and use HCS technology.

[0182] like Figure 7 As shown in Figure 2, when λ1 = 0 or λ2 = 0, there are two cases for the utility function change curves of users with different disease levels.

[0183] 1. Let λ cS =15,λ w =25,λ r =15,λ m =12,com=10, and VDec min When =λ2-com>0, an upward curve is obtained, that is, the minimum value of the utility function is greater than zero, so individuals with disease levels of 1, 2, and 3 all accept and use HCS technology.

[0184] 2. Let λ cS =48,λ w =15,λ r =18,λ m =15,com=10, and VDec max =λ1-com>0, a descending curve is obtained, that is, the maximum value of the utility function is greater than zero. When VDec=0, l takes the value 3, and l∈[1,3). Therefore, individuals with disease levels 1 and 2 accept and use HCS technology.

[0185] Numerical experimental analysis further demonstrates that when key factors influencing decision-making vary, individuals' decisions regarding technology adoption and use can differ. Furthermore, the platform can influence individual decisions by adjusting the services provided in contracts for individuals with different disease levels, thereby obtaining data from different individuals. This decision-making model provides a theoretical basis for further research on the impact of digital transformation on individual technology adoption.

[0186] Based on the above embodiments, Figure 8As shown, the present invention also proposes a system for constructing an individual decision model based on contract theory in a health crowd intelligence perception scenario. The system is based on the health crowd intelligence perception system HCS. The health crowd intelligence perception system includes individuals, platforms, and users. Individuals refer to individuals who have not yet used health crowd intelligence perception technology, i.e., HCS technology; platforms refer to technology providers that provide personalized medical guidance services to users who use HCS technology; and users refer to individuals who have already used HCS technology. The system includes:

[0187] A utility function building module is used to construct an individual utility function based on the individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, and the impact of compatibility and complexity;

[0188] The constraint construction module is used to extract the relationship between the intrinsic motivation of individuals to maintain good health, the cost of using technology, the risks they bear, the personalized medical guidance services provided by the platform and the disease level in the utility function, and establish constraint conditions;

[0189] The decision model building module is used to build an individual decision model based on the constructed individual utility function and constraints.

[0190] Furthermore, the utility function is:

[0191] VDec=cS+wrm-com (7)

[0192] stVDec>0

[0193] Where VDec represents the utility function used by individuals to decide whether to use HCS technology, w represents the intrinsic motivation of individuals to maintain good health, m represents the cost of using HCS technology, r represents the risk that individuals bear when using HCS technology, cS represents the personalized medical guidance service provided by the platform, and com represents the negative impact of compatibility attributes and complexity attributes on individuals' use of HCS technology.

[0194] The constraints are:

[0195] w=λ w l if l1 <l2<...<l L then w1 <w2<...<w L (8)

[0196]

[0197]

[0198] m=λ m l if l1 <l2<...<l Lthen m1 <m2<...<m L (11)

[0199] Moreover, in the era of digital transformation, health crowd intelligence perception as a technology has the same degree of negative impact on all individuals in terms of compatibility attributes and complexity attributes.

[0200] Among them, λ w represents the conversion coefficient between intrinsic motivation and disease level, L represents the number of disease levels, and λ cS represents the conversion coefficient between an individual's access to services and their disease level, λ r is the conversion coefficient between the privacy risk borne by an individual and his / her disease level, λ m The conversion factor between the cost borne by an individual and the level of his or her disease;

[0201] The decision model is:

[0202]

[0203] Furthermore, it also includes:

[0204] The model optimization module is used to optimize the decision model in the following ways:

[0205]

[0206] Furthermore, it also includes:

[0207] The decision module is used by individuals to decide whether to accept and use HCS technology based on the decision model. If the decision result is to accept and use the technology, the next step is taken; otherwise, the technology is not used;

[0208] The contract signing module is used for individuals to sign the disease level service contract issued by the platform, become a platform user, and start using HCS technology;

[0209] The data collection and upload module is used by users to collect in-depth data related to their individual health through the smart terminal they carry, and upload the collected data to the platform;

[0210] The service provision module is used by the platform to provide personalized diagnosis and treatment guidance services to users based on the contracts signed and the data uploaded.

[0211] In summary, this invention combines innovation diffusion theory and contract theory for the first time. Based on the self-disclosure and fair payment mechanisms of contract theory, and taking into account key factors such as an individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, and the impact of compatibility and complexity, a decision-making model for individual acceptance and use of HCS technology in the digital age was established. By extracting the relationship between key factors influencing decision-making and individual core needs, and transforming individual decision-making problems into cost-effectiveness problems, a cost-effectiveness analysis was conducted to derive the dynamic relationship between individual decision-making outcomes and changes in key factors. In the experimental phase, numerical analysis was used to verify the differences in individual decision-making outcomes when different key factor values ​​were used, based on the results of model optimization. These experimental results indicate the research directions that should be pursued to further promote individual use of HCS technology.

[0212] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for constructing an individual decision-making model based on contract theory in a health crowd intelligence perception scenario, characterized in that: The method is based on a health crowd intelligence perception system, which includes individuals, platforms and users, wherein individuals refer to individuals who have not used health crowd intelligence perception technology, i.e., HCS technology, platforms refer to technology providers that provide personalized medical guidance services to users using HCS technology, and users refer to individuals who have used HCS technology; the method includes: Construct an individual utility function based on the individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, and the impact of compatibility and complexity; The utility function is: VDec=cS+wrm-com (7) stVDec>0 Among them, VDec represents the utility function used by individuals to decide whether to use HCS technology, w represents the intrinsic motivation of individuals to maintain good health, m represents the cost of using HCS technology, r represents the risk that individuals take when using HCS technology, cS represents the personalized medical guidance service provided by the platform, and com represents the negative impact of compatibility attributes and complexity attributes on individuals' use of HCS technology; Extract the intrinsic motivation of individuals to maintain good health, the cost of using technology, the risk taken, the relationship between the personalized medical guidance services provided by the platform and the disease level in the utility function, and establish constraints; The constraints are: w=λ w l if l1<l2<...<l L then w1<w2<...<w L (8) m=λ m l if l1<l2<...<l L then m1<m2<...<m L (11) And in the era of digital transformation, health crowd intelligence perception as a technology has the same degree of negative impact on all individuals in terms of compatibility attributes and complexity attributes. Among them, λ w represents the conversion coefficient between intrinsic motivation and disease level, L represents the number of disease levels, and λ cS represents the conversion coefficient between the individual's access to services and their disease level, λ r is the conversion coefficient between the privacy risk borne by an individual and his / her disease level, λ m The conversion factor between the cost of use borne by an individual and his / her disease level; An individual decision-making model is constructed based on the constructed individual utility function and constraints.

2. The method for constructing an individual decision-making model based on contract theory in a health crowd intelligence perception scenario according to claim 1 is characterized in that: The decision model is:

3. The method for constructing an individual decision-making model based on contract theory in a health crowd intelligence perception scenario according to claim 2 is characterized in that: After building the individual decision model, it also includes: The decision model is optimized as follows:

4. The method for constructing an individual decision-making model based on contract theory in a health crowd intelligence perception scenario according to claim 1 or 3, characterized in that: After building the individual decision model, it also includes: Individuals decide whether to accept and use HCS technology based on the decision model. If the decision result is to accept and use the technology, they proceed to the next step, otherwise they do not use the technology; Individuals sign the disease-level service contract published by the platform, become platform users, and start using HCS technology; Users collect in-depth data related to their individual health through the smart terminals they carry, and upload the collected data to the platform; The platform provides personalized diagnosis and treatment guidance services based on the contracts signed by users and the data uploaded.

5. A system for building an individual decision-making model based on contract theory in a health crowd intelligence perception scenario, characterized in that: The construction system is based on the health crowd intelligence perception system HCS, which includes individuals, platforms and users, where individuals refer to individuals who have not yet used the health crowd intelligence perception technology, namely HCS technology, platforms refer to technology providers that provide personalized medical guidance services to users using HCS technology, and users refer to individuals who have already used HCS technology; the construction system includes: A utility function building module, which is used to build an individual utility function based on the individual's intrinsic motivation to maintain good health, the cost of using technology, the risks assumed, the personalized medical guidance services provided by the platform, and the impact of compatibility and complexity; The utility function is: VDec=cS+wrm-com (7) stVDec>0 Among them, VDec represents the utility function used by individuals to decide whether to use HCS technology, w represents the intrinsic motivation of individuals to maintain good health, m represents the cost of using HCS technology, r represents the risk that individuals take when using HCS technology, cS represents the personalized medical guidance service provided by the platform, and com represents the negative impact of compatibility attributes and complexity attributes on individuals' use of HCS technology; The constraint building module is used to extract the intrinsic motivation of individuals to maintain good health, the cost of using technology, the risk they bear, the relationship between the personalized medical guidance services provided by the platform and the disease level in the utility function, and establish constraint conditions; The constraints are: w=λ w l if l1<l2<...<l L then w1<w2<...<w L (8) m=λ m l if l1<l2<...<l L then m1<m2<...<m L (11) And in the era of digital transformation, health crowd intelligence perception as a technology has the same degree of negative impact on all individuals in terms of compatibility attributes and complexity attributes. Among them, λ w represents the conversion coefficient between intrinsic motivation and disease level, L represents the number of disease levels, and λ cS represents the conversion coefficient between the individual's access to services and their disease level, λ r is the conversion coefficient between the privacy risk borne by an individual and his / her disease level, λ m The conversion factor between the cost of use borne by an individual and his / her disease level; The decision model building module is used to build an individual decision model based on the constructed individual utility function and constraints.

6. The individual decision model construction system based on contract theory in the health crowd intelligence perception scenario according to claim 5 is characterized in that: The decision model is:

7. The individual decision model construction system based on contract theory in the health crowd intelligence perception scenario according to claim 6 is characterized in that: Also includes: The model optimization module is used to optimize the decision model in the following ways:

8. The individual decision model construction system based on contract theory in the health crowd intelligence perception scenario according to claim 5 or 7 is characterized in that: Also includes: The decision module is used for individuals to decide whether to accept and use the HCS technology according to the decision model. If the decision result is to accept and use the technology, the next step is entered; otherwise, the technology is not used; The contract signing module is used for individuals to sign the disease level service contract issued by the platform, become a platform user, and start using HCS technology; The data collection and upload module is used for users to collect in-depth data related to individual health through the smart terminal they carry, and upload the collected data to the platform; The service provision module is used by the platform to provide personalized diagnosis and treatment guidance services to users based on the contracts signed and the data uploaded by them.

Images