KR102569393B1 - System and method for competitive data trading according to non-cooperative competition - Google Patents

Abstract

Various embodiments provide a competitive data trading system and method according to uncooperative competition. According to various embodiments, a service providing device allocates a budget to each of a plurality of data broker devices, and each of the data broker devices bids on personal data for sale to the service providing device based on the allocated budget; , The service providing device may be configured to purchase personal data sold from at least one of the data broker devices using the allocated budget.

Description

Competitive data trading system and method according to non-cooperative competition {SYSTEM AND METHOD FOR COMPETITIVE DATA TRADING ACCORDING TO NON-COOPERATIVE COMPETITION}

Various embodiments relate to systems and methods for trading competitive data based on uncooperative competition.

With the newly emerging big data concept, data has become a new value asset of the 4th industrial revolution. As a result, a system for trading data in the big data market was established. Generally, the system is composed of data providers, data brokers (or data vendors), and service consumers, and the data broker relays transactions of personal data between the data providers and service consumers. Here, the personal data may include mydata that is actively managed and controlled by the individual as the subject of information and can be actively utilized. That is, data brokers collect personal data from data providers and sell personal data for service consumers. Through this, service demanders use services related to personal data.

However, in the above system, the data broker is configured as a single unit. Due to this, there is a problem in that the operating effect is low when the system as described above relays the transaction of personal data.

Various embodiments provide a competitive data trading system and method that enables effective trading of personal data according to non-cooperative competition.

Various embodiments provide a competition data trading system and method capable of stably trading personal data while maximizing profits on personal data according to non-cooperative competition.

A non-cooperative competitive data transaction method according to various embodiments includes allocating, by a service providing device, a budget to each of a plurality of data broker devices, each of the data broker devices based on the allocated budget, the service providing device. The method may include bidding on personal data to be sold, and purchasing, by the service providing device, personal data sold from at least one of the data broker devices using the allocated budget.

A non-cooperative competitive data trading system according to various embodiments includes a plurality of data broker devices configured to compete to each sell at least one personal data among a plurality of data providing devices, and a budget for each of the data broker devices. and a service providing device configured to allocate and purchase personal data sold from at least one of the data broker devices using the allocated budget, wherein each of the data broker devices corresponds to the allocated budget. Based on the data, it may be configured to bid on the personal data to be sold to the service providing device.

According to various embodiments, as the competitive data trading system is implemented with a plurality of data broker devices and a single service providing device, the data broker devices and the service providing device can effectively trade personal data according to uncooperative competition. Specifically, data broker devices sell personal data in consideration of willingness-to-sell (WTS) of data providing devices, and service providing devices consider WTP (willingness-to-pay) of service demand devices, Data can be purchased. At this time, based on the competition between the data broker devices and the competition between the data broker devices and the service providing device, the competing data trading system can effectively trade personal data according to uncooperative competition. Accordingly, within the competitive data trading system, the data broker devices and the service providing device can stably trade personal data while maximizing profits from trading personal data.

1 is a diagram illustrating a competition data trading system according to various embodiments.
2 is a diagram illustrating a signal flow in a competitive data trading system according to various embodiments.
3 is a diagram conceptually illustrating a competition data trading system according to various embodiments.
4 is a diagram illustrating a service providing device according to various embodiments.
5 is a diagram illustrating an operating method of a service providing apparatus according to various embodiments.
FIG. 6 is a diagram illustrating the personal data purchasing step of FIG. 5 .
7 is a diagram for explaining an operation algorithm of a service providing apparatus according to various embodiments.
8 is a diagram illustrating a data broker device according to various embodiments.
9 is a diagram illustrating an operating method of a data broker device according to various embodiments.

Hereinafter, various embodiments of this document will be described with reference to the accompanying drawings.

1 is a diagram illustrating a competition data trading system 100 according to various embodiments.

Referring to FIG. 1, a competitive data trading system 100 according to various embodiments includes a plurality of data providing devices (data provider; data providing device) 110, a plurality of service consumer; service consuming device ), multiple data broker devices, and a single data providing device (data provider).

Each of the data providing devices 110 may provide personal data for at least one user to at least one of the data broker devices 120 . To this end, each of the data providing devices 110 may collect personal data in relation to the user. Also, each of the data providing devices 110 may store personal data. Here, the personal data may include mydata that is actively managed and controlled by the individual as the subject of information and can be actively utilized. Personal data may be of various types, and may include, for example, at least one of physical conditions, physical activities, communication details, information use records, creative data, schedules, payment details, or visit records. The data providing devices 110 may include various types of electronic devices. For example, electronic devices include smart phones, mobile phones, navigation devices, computers, laptop computers, digital broadcasting terminals, personal digital assistants (PDAs), portable multimedia players (PMPs), tablet PCs, game consoles, It may include at least one of a wearable device, an internet of things (IoT) device, a medical device, a home appliance, a camera, or a robot. According to one embodiment, each of the data providing devices 110 may sell personal data to at least one of the data broker devices 120 .

Each of the data broker devices 120 may sell personal data of at least one of the data providing devices 110 to the service providing device 120 . To this end, at least two of the data broker devices 120 may compete with each other. Each of the data broker devices 120 may collect personal data from at least one of the data providing devices 110 . According to one embodiment, each of the data broker devices 120 may purchase personal data from at least one of the data providing devices 110 . Here, each of the data broker devices 120 may collect at least one predetermined type of personal data. The data broker devices 120 may collect different types of personal data, and at least two of the data broker devices 120 may collect the same type of personal data. The data broker devices 120 may collect personal data from different data providing devices 110, and at least two of the data broker devices 120 may collect personal data from the same data providing device 110. . For example, each of the data broker devices 120 may be a server having a database.

The service providing device 130 may purchase personal data from at least one of the data broker devices 120 . Here, the service providing device 130 may purchase at least one predetermined type of personal data. At this time, under competition between at least two of the data broker devices 120, the service providing device 130 may select at least one of the data broker devices 120 and purchase personal data from the selected data broker device 120. there is. In addition, the service providing device 130 may provide a service related to personal data to at least one of the service requesting devices 140 . According to an embodiment, the service providing device 130 may sell the service to at least one of the service demand devices 140 . For example, the service providing device 130 may be a server having a database.

Each of the service requesting devices 140 may use a service provided by the service providing device 130 . According to an embodiment, each of the service demand devices 140 may purchase a service from the service providing device 130 . The service demand devices 140 may include at least one of various types of electronic devices or servers. For example, electronic devices may include at least one of a smartphone, a mobile phone, a navigation device, a computer, a laptop computer, a digital broadcast terminal, a PDA, a PMP, a tablet PC, a game console, a wearable device, an IoT device, a medical device, a home appliance, a camera, or a robot. may contain one.

According to various embodiments, within the competitive data trading system 100, non-cooperative competitive trading for personal data may be made. Each of the data broker devices 120 may sell personal data of at least one of the data providing devices 110 to the service providing device 130 . At this time, each of the data broker devices 120 may sell personal data to obtain an optimal profit. To this end, at least two of the data broker devices 120 may compete to sell personal data. Through this, the service providing device 130 can purchase personal data sold by at least one of the data broker devices 120 and provide a service related to the personal data to at least one of the service demand devices 140. . At this time, the service providing device 130 may purchase personal data so as to obtain an optimal profit. Here, at least one of the data broker devices 120 and the service providing device 130 may compete to trade, that is, to sell and purchase personal data.

2 is a diagram illustrating a signal flow in a competitive data trading system 100 according to various embodiments.

Referring to FIG. 2, competitive data transactions are made between at least one data providing device 110, at least one data broker device 120, a service providing device 130, and at least one service demand device 140. can To this end, each data broker device 120 identifies the number of at least one data providing device 110 that provides personal data for sale, and the service providing device 130 provides at least one The number of service demand devices 140 may be known. In addition, a protocol for non-cooperative competitive transactions may be established between the data broker device 120 and the service providing device 130.

First, the service providing device 130 may predict expected revenue for the service to be provided to the service demand device 140 in step 210 . In this case, the service providing device 130 may predict expected revenue based on at least one of the number of service demand devices 140 for providing the service, the price of the service, and the expected quality of the service. Thereafter, the service providing device 130 may determine a budget for purchasing personal data related to the service based on the expected revenue in step 220 . Accordingly, the service providing device 130 may allocate a budget to the data broker device 120 in step 230 . For example, for a plurality of data broker devices 120, the service providing device 130 divides the budget into individual budgets corresponding to the number of data broker devices 120, and the data broker devices 120 individual budgets can be allocated to each.

Subsequently, the data broker device 120 may predict expected revenue for selling personal data to the service providing device 130 in step 240 . In this case, the data broker device 120 may predict expected revenue based on at least one of a budget allocated from the service providing device 130 and a unit price of personal data. Here, the unit price may be determined based on at least one of the number of data providing devices 110 providing personal data or privacy sensitivity of personal data. Thereafter, the data broker device 120 may trade personal data with the data providing device 110 in step 250 . At this time, based on the unit price of the personal data, the data providing device 110 may sell the personal data and the data broker device 120 may purchase the personal data. That is, the data broker device 120 may pay the data providing device 110 a unit price for personal data and purchase personal data from the data providing device 110 .

Continuing, the data broker device 120 may bid for personal data with the service providing device 130 in step 260 . At this time, the data broker device 120 may determine a bid unit price of the personal data to be sold based on the expected profit. In addition, the data broker device 120 may bid for personal data while submitting a bid unit price to the service providing device 130 . Accordingly, the service providing device 130 may trade personal data with the data broker device 120 in step 270 . At this time, based on the budget, the data broker device 120 may sell personal data and the service providing device 130 may purchase personal data. That is, the service providing device 130 may pay a budget to the data broker device 120 and purchase personal data from the data broker device 120 .

Finally, the service providing device 130 may trade services with the service demand device 140 in step 280 . At this time, based on the price for the service, the service providing device 130 may provide the service and the service demand device 140 may use the service. That is, the service demand device 140 may pay a price for the service and purchase the service from the service providing device 130 .

3 is a diagram conceptually illustrating a competition data trading system 100 according to various embodiments.

Referring to FIG. 3 , a competition data trading system 100 according to various embodiments is a component that operates for the profit of a child for K types of personal data, Potential data providing devices 110 divided into groups, R data broker devices ( ) 120, a single service providing device 130, and M potential service demand devices 140. The data providing device 110 and the data broker device 120 are interested in selling personal data under privacy considerations for personal data, the service providing device 130 purchases personal data, and the service demand device ( 140) may be interested in providing services related to personal data.

The service providing device 130 has a predetermined quality (from at least one of all the data broker devices 120). ) of personal data as data sets of a predetermined quality Q(q). At this time, the service providing device 130 is based on at least one of the price p _s for the service, the number of at least one service demand device 140 for providing the service, or the expected service quality S for the service. Thus, expected returns can be predicted. And, since the service providing device 130 will consume the budget B as a cost for personal data, it can determine the optimal budget B to maximize its profit. Thereafter, the service providing device 130 may request a bid from the data broker devices 120 to purchase personal data from at least one of the data broker devices 120 .

Through this, competition may occur between the data broker devices 120 . That is, at least two of the data broker devices 120 may compete to sell personal data. At this time, at least one of the data broker devices 120 may purchase personal data from at least one data providing device 110 and bid to sell it to the service providing device 130 . Here, the data broker devices 120 may bid without knowing about each other's bid. Through this, at least one of the data broker devices 120 may sell personal data to the service providing device 130 .

Meanwhile, competition may also occur between the data broker devices 120 and the service providing device 130 . The service providing device 130 may maximize the quality of service S obtained from the quality Q(q) of data sets using a limited budget B to purchase personal data of a predetermined quality q. there is. Meanwhile, at least one of the data broker devices 120 may receive an individual budget bi for personal data from the service providing device 130 by selling personal data of quality qi as a data set. The individual budget (bi) may be the actual revenue of the data broker device 120.

According to various embodiments, there may be multiple data broker devices 120 and a single service provider device 130 in the competitive data trading system 100 . Data broker devices 120 will have to sell as much personal data as possible. Meanwhile, the service providing device 130 will have to find the most efficient way to use the limited budget B. At this time, each of the data broker devices 120 considers both the individual budget (bi) from the service providing device 130 and the unit price of personal data in at least one data providing device 110, thereby selling quality personal data. You can optimize your profit to do it. Here, each of the data broker devices 120 takes into account the willingness-to-sell (WTS) of the data providing device 110 to obtain the necessary personal data with a predetermined quality (qi) at a total cost ( ) can be focused on minimizing

The service providing device 130 may allocate the budget B to the data broker device 120 . That is, the service providing device 130 divides the budget B into individual budgets bi corresponding to the number of data broker devices 120, and assigns individual budgets bi to the data broker devices 120, respectively. can be assigned For example, individual budgets (bi) may be defined as in [Equation 1] below. That is, the sum of all individual budgets (bi) cannot exceed the budget (B).

Each of the data broker devices 120 may predict an expected profit Ui. In addition, the data broker devices 120 may compete as each of the data broker devices 120 bids for the service providing device 130 based on the expected profit Ui. The data broker device 120 may determine a bid unit price of personal data to be sold based on the expected profit Ui. In addition, the data broker device 120 may bid for personal data while submitting a bid unit price to the service providing device 130 .

As described below, an expectation function (Ui) may be defined for data broker apparatuses 120 and uncooperative competition between data broker apparatuses 120 according to a Nash equilibrium analysis (NE).

Definition 1 : The expected revenue (Ui) of each of the data broker devices 120 may be defined as in [Equation 2] below.

Here, δi is the unit price of personal data of a predetermined quality (qi ∈ [0, ∞)).

As shown in FIG. 3 , the data broker devices 120 compete with each other to obtain individual budget bi from the service providing device 130 by selling personal data to the service providing device 130 . This can be seen as a non-cooperative game between those defined in Definitions 2 and 3 described later. Non-cooperative competition between the data broker devices 120 proposed together with [Equation 1] and [Equation 2] requires minimum information about other participants, so it is appropriate for participants to dynamically participate in the market. . Here, the participant only needs information on the budget (B) and unit price (δ), and does not need to be concerned with the detailed strategies of other participants.

Definition 2 : Non-cooperative competition of data broker devices 120 is a non-cooperative strategy game model, i.e. is expressed as here, Is a domain of all data broker devices 120, and Uk may correspond to the expected revenue of [Equation 2] above.

Definition 3 : NE of G is It is a profile of strategies (q*) that satisfy . now, The existence and uniqueness of NE(q) to maximize can be proved.

Lemma 1 can show the feasibility of the proposed game model.

Lemma 1 : For a vector q, if at least two components of q are positive and the vector q satisfies the following [Equation 3], is the NE of the game maximizing , that is, q*.

Proof: You can prove by showing that the necessary and sufficient conditions hold. These conditions are the partial derivatives with respect to qi [i.e. ] can be derived by taking Necessary conditions can be proven to contradict zero and only one participant. A sufficient condition can be shown by considering q to be at least 2 (i.e., at least two participants to the market).

Transformed utility function like Lemma 2 Convert the original game into a modified game model with .

Lemma 2 : The optimization problem presented in [Equation 4] below is considered. Here, the following [Equation 5] is referred to. If so, [Equation 4] has a unique solution as shown in [Equation 6] below.

Here, υ is is the actual value that satisfies

Proof: Show the first partial derivative of the modified utility function, then It can be proved by applying the Karush-Kuhn-Tucker (KKT) condition to find Then, [Equation 6] can be obtained. While applying the KKT condition conditions can also be obtained.

The bidding of the data broker devices 120 is understood to be a non-negative number by [Equation 2] above. A data broker device 120 having a bid of 0 (ie, qi = 0) means that the corresponding data broker device 120 does not participate in the transaction. Therefore, it can be assumed that all participants (ie, data broker devices 120) actually participate in bidding with bi > 0 and qi > 0 for all i ∈ R. Then, the result of Lelem 2 can be simplified as shown in [Equation 7] below.

Moreover, the above [Equation 7] and the condition As B, υ is can be obtained by evaluating in other words, am.

Lemma 3 : G is the solution of Lemma 2 given optimal budget allocation that satisfies has a unique NE q* corresponding to

Proof: The modified game model of [Equation 4] above has a unique NE point with a condition from the first derivative, and then it can be seen that the modified game model is actually the same as Lelem 1. Here is an outline of the proof. First, by showing that Lemma 2 has the same eigensolution, the other side become; the other side become; , it can be seen that there exists a unique b* and a scalar υ. then vector is NE, and it can be seen that it is an eigensolution using lemma 1. finally that optimal budget allocation that satisfies It can be concluded that there exists a unique NE q* with .

In addition, the unique NE q* in Definition 2 can be obtained through the following theorem.

Theorem 1 : Uncooperative competition of data broker devices 120 has a unique NE q*

Proof: From lemma 2 and 3 above, NE solution silver is given as here, am. Then, it can be applied to [Equation 7] and [Equation 8] above.

Due to the unique NE of [Equation 8] above, all participants can easily predict the strategy of others in the market in a closed form. The service providing device 130 may determine the amount of the data set using its own budget for purchasing personal data and a seller's unit price. On the other hand, from the point of view of the data broker device 120, competitiveness in a market directly related to profits can be measured.

As described above, since positive (+) bidding of the data broker device 120 is premised, NE q* in [Equation 8] cannot but be positive. Therefore, the competition data trading system 100 must satisfy the following two conditions of [Equation 9].

The first condition is related to the feasibility of the proposed game model. That is, the proposed game model requires at least two data broker devices 120 (i.e., R ≥ 2), which is also confirmed in lemma 1. The second condition is related to the operation of each data broker device 120 and can be interpreted as shown in [Equation 10] below.

The first case (δ _i > 0) represents a basic condition for the unit price of personal data provided by each data broker device 120 . The unit price must be greater than zero. second case indicates that the individual data provided to each data broker device 120, i.e., the unit price of the data set, must be competitive enough to participate in the proposed game model. That is, when the unit price of one data broker device 120 is extremely high compared to other participants, the data broker device 120 has no opportunity to bid from the service providing device 130 .

4 is a diagram illustrating an electronic device (eg, a service providing device 130) 400 according to various embodiments.

Referring to FIG. 4 , an electronic device 400 according to various embodiments may be the service providing device 130 of FIGS. 1 and 2 . The electronic device 400 may include at least one of a communication module 410 , an input module 420 , an output module 430 , a memory 440 , and a processor 450 . In some embodiments, at least one of the components of the electronic device 400 may be omitted and at least one other component may be added. In some embodiments, at least two of the components of the electronic device 400 may be implemented as a single integrated circuit.

The communication module 410 may perform communication with external devices (eg, data broker devices 120 and service demand devices 140) in the electronic device 400. The communication module 410 may establish a communication channel between the electronic device 400 and an external device, and perform communication with the external device through the communication channel. The communication module 410 may include at least one of a wired communication module and a wireless communication module. The wired communication module may be connected to an external device through wired communication. The wireless communication module may communicate with an external device through a network.

The input module 420 may input a signal to be used in at least one component of the electronic device 400 . For example, the input module 420 may include at least one of a microphone, mouse, or keyboard.

The output module 430 may output information of the electronic device 400 . The output module 430 may include at least one of a display module and an audio output module. The display module may include, for example, at least one of a display, a hologram device, and a projector. The audio output module may include, for example, at least one of a speaker and a receiver.

The memory 440 may store various data used by at least one component of the electronic device 400 . For example, the memory 440 may include at least one of volatile memory and non-volatile memory. The data may include at least one program and related input data or output data. The program may be stored as software including at least one instruction in the memory 440, and may include, for example, at least one of an operating system, middleware, and applications.

The processor 450 may control at least one component of the electronic device 400 by executing a program of the memory 440 . Through this, the processor 450 may perform data processing or calculation. At this time, the processor 450 may execute instructions stored in the memory 440 . Processor 450 may purchase personal data from at least one of data broker devices 120 . Here, the processor 450 may purchase at least one predetermined type of personal data. At this time, under competition between at least two of the data broker devices 120, the processor 450 may select at least one of the data broker devices 120 and purchase personal data from the selected data broker device 120. Also, the processor 450 may provide a service related to personal data to at least one of the service requesting devices 140 . According to an embodiment, the processor 450 may sell the service to at least one of the service requesting devices 140 .

5 is a diagram illustrating an operating method of an electronic device 400 according to various embodiments.

Referring to FIG. 5 , in step 510, the electronic device 400 may predict expected revenue for a service to be provided to at least one of the service demand devices 140. The processor 450 may predict expected revenue based on at least one of the number of service demand devices 140 for providing the service, the price of the service, and the expected quality of the service. At this time, the processor 450 may predict the expected profit as the maximized profit.

In step 520, the electronic device 400 may determine a budget for purchasing personal data related to the service. Processor 450 may determine a budget based on expected revenue for the service. At this time, the processor 450 may identify at least one predetermined type of personal data related to the service.

The electronic device 400 may purchase personal data from at least one of the data broker devices 120 in step 530 . The processor 450 may purchase personal data using a budget. That is, the processor 450 may pay a budget to at least one of the data broker devices 120 and purchase personal data. At this time, the processor 450 may purchase personal data from at least one of the data broker devices 120 based on the bidding of at least one of the data broker devices 120 . According to one embodiment, the processor 450 requests a bid from all data broker devices 120, which allows the data broker devices 120 to bid. According to another embodiment, the processor 450 requests bids from data broker devices 120 that sell the identified type of personal data, so that the detected data broker devices 120 can bid. This will be described later with reference to FIG. 6 .

FIG. 6 is a diagram illustrating the personal data purchasing step of FIG. 5 .

Referring to FIG. 6 , the electronic device 400 may allocate budgets to the data broker devices 120 in step 631 . The processor 450 may divide the budget into individual budgets corresponding to the number of data broker devices 120 and allocate the individual budgets to the data broker devices 120, respectively. At this time, each of the individual budgets may be determined based on the quality of personal data to be sold in each of the data broker devices 120 . In addition, the processor 450 may request bids from the data broker devices 120 while allocating individual budgets to the data broker devices 120, respectively.

The electronic device 400 may acquire a bid unit price from each of the data broker devices 120 in step 633 . As the data broker devices 120 bid, the processor 450 may obtain a bid unit price from each of the data broker devices 120 . In this case, each of the data broker devices 120 may predict expected revenue based on at least one of a budget allocated from the service providing device 130 and a unit price of personal data. Here, the unit price may be determined based on at least one of the number of data providing devices 110 providing personal data or privacy sensitivity of personal data. Also, each of the data broker devices 120 may determine a bid unit price of personal data to be sold based on the expected profit. Through this, each of the data broker devices 120 may bid for personal data while submitting a bid unit price to the service providing device 130 .

The electronic device 400 may select at least one of the data broker devices 120 in step 635 . The processor 450 may select at least one of the data broker devices 120 based on the budget and the bidding unit price of each of the data broker devices 120 . Here, the sum of bid unit prices of the selected data broker device 120 must not exceed the budget.

The electronic device 400 may purchase personal data from the selected data broker device 120 in step 637 . The processor 450 may purchase personal data using a budget. That is, the processor 450 may pay an individual budget or bid unit price allocated to the selected data broker device 120 and purchase personal data sold in the selected data broker device 120 . Thereafter, the electronic device 400 may return to FIG. 5 and proceed to step 540 .

Referring back to FIG. 5 , the electronic device 400 may provide a service to at least one of the service requesting devices 140 in step 540 . At this time, based on the price for the service, the processor 450 may provide the service, and the service demand device 140 may use the service. That is, the service demand device 140 may pay a price for the service and purchase the service from the electronic device 400 .

Optimization of the service providing device 130 may be performed using the NE result. Introduce the problem. The expected revenue P of the service providing device 130 may be predicted by considering the expected quality S of the service obtained by the personal data collected with the expected quality Q, that is, the data set. The service providing device 130 may predict expected revenue by considering the expected service quality (S) and the willingness-to-pay (WTP) (Ψ) of the service demand device 140 having the expected quality (S). And, the budget (B) may be determined as an expected cost for achieving the expected quality (S). The proposed model for personal data, that is, the expected quality (Q) of a data set and the expected service quality (S) for a service, is a quantification method for predicting the amount of personal data that the service providing device 130 will purchase to maximize revenue. , which can be used as part of the criteria for determining market participation.

A. Service quality (S) of the service providing device 130

Before defining the expected service quality (S), the entire personal data from all data broker devices 120, i.e. the data sets ( The expected quality (Q) of the service providing device 130 obtained by ) is defined. To define expected quality (Q), three principles apply. The first is the amount of data (i.e. size or volume) associated with inherent data qualities such as completeness, accuracy, etc. The second is the nature of the data. That is, the combination of data sets increases the chance of obtaining identifiable information. For example, in online environments, cookie synchronization techniques are widely used to identify and track user behavior for better service quality and personalization. The third is the law of diminishing marginal utility: marginal utility decreases as supply increases, and the rate of decrease is inversely proportional to the amount of data set. And, the expected quality (Q) is defined as in [Equation 11] below.

Definition 4 :

where ηij is the correlation between the data sets qi and qj.

To define the expected quality (Q), the geometric mean (root term) becomes the most popular and manageable function that satisfies the first and second principles. That is, (1) the quality of the data set (having the service providing device 130) is high when the data set is closely correlated, (2) the amount of the data set is large, and (3) the Data sets are balanced in volume. When ηij = 1 (i = j) and ηij = 0 (i ≠ j), equal to the linear sum of the data set qualities of each data broker device 120 form Furthermore, the natural log function is used to apply the third principle, which is usually proposed to model the law of diminishing marginal utility (the function is inversely proportional to the amount of the data set, i.e. du = x ^-1 dx).

Along with the expected quality Q, the expected service quality S for the service providing device 130 is proposed with predictive accuracy using the collected data set. The expected service quality (S) is defined as in [Equation 12] below.

Definition 5 :

Here, α ∈ (0, 1] and β ∈ (0, 1] are parameters, and S(Q) ∈ [0, 1).

B. WTP of Service Demand Device 140

The service providing device 130 should predict the number of service demand devices 140 to provide the service based on the service quality (S) to maximize profits and determine a budget size for purchasing a data set in the market. Accordingly, the WTP of the service requesting devices 140 must be presented.

Each service demand device 140 has its own criterion for paying the service price (ie, the service demand device 140 determines whether to pay the service price). Therefore, it is difficult to separately formulate each person's behavior from the viewpoint of the service providing device 130 . That is, the WTP should be modeled at the macro level by considering the cumulative distribution as the percentage of paid consumers of all service demand devices 140 .

To formulate the WTP, we consider the basic economic principles for service demand: 1) the WTP decreases when the offer price increases and 2) the service demand device 140 prefers to purchase a service with a higher quality. The WTP of the service demand device 140 is defined in Definition 6 described below.

Definition 6 : WTP (Ψ) of the service demand device 140 is defined as the following [Equation 13] based on the price ( _ps ) and service quality (S) presented by the cumulative distribution function (CDF).

Here, p _s ≥ 0, and 0 ≤ S < 1.

Based on the proposed model for service quality (S) and WTP (Ψ), the expected profit (P) of the service providing device 130 may be modeled as in [Equation 14] below.

Here, M is the number of service demand devices 140, p _s is the price of the service, and B is a budget for purchasing a data set from all data broker devices 120 in the market.

Means the number of expected service demand devices 140 having the proposed service price ( _ps ) and the proposed service quality (Q). thus, Means the expected revenue from the service demand devices 140. On the other hand, budget (B) is the estimated cost to purchase the data set. Based on this, the profit maximization problem can be formulated as in [Equation 15] below.

Problem 1 :

here, am. q* with a fixed budget (B) maximizing service quality (Q) can be determined by NE of Theorem 1 above.

C. Maximizing the profit of the service providing device 130

An optimal strategy of the service providing apparatus 130 maximizing an expected profit (P( _ps , B)) is solved (problem 1). for teeth, is first obtained from Theorem 2 described later together with the fixed budget B (Theorem 2), and the optimal budget B* of the service providing apparatus 130 is finally obtained from Theorem 3 described later.

Theorem 2 : The optimal price is the one that maximizes the profit function of the service providing device 130. As, as shown in [Equation 16] below.

here, ego, am.

Proof: As shown in [Equation 17] below, the concavity of P with respect to p _s is confirmed by considering whether or not the second derivative of p is 0 or less. S ∈ [0, 1).

Since the expected profit (P) is a concave function, the maximum point obtained by checking the first derivative with p _s is 0 as shown in [Equation 18] below.

From this Theorem 2, the optimal solution to Problem 1, which is the optimal strategy of the service providing apparatus 130, is finally obtained as Theorem 3 described later.

Theorem 3: Optimal strategy of the service providing device 130 maximizing profits based on Theorem 2 (Problem 1) is given as the following [Equation 19].

here, ego, is, is given in Theorem 2.

Proof: best price for service provider 130 is the budget (B) from Theorem 2. Now, p _s = The expected return (P(ps, |B)) under the condition , in other words When expressed as, it is as shown in [Equation 20] below.

Similarly, as shown in [Equation 21], the expected return (changed by taking the second derivative) ) to check the concavity of If is concave, it has a unique maximum (B ^* ).

Changed expected return ( ) is a concave function, so the overall return ( ) exists as shown in [Equation 22] below.

Since the budget B* required to maximize the profit of the service providing device 130 is obtained as a closed solution, it can be easily applied to a dynamic market. Moreover, the main part of the result ((α/β)Me ⁻¹ ) depends only on its own characteristics (ie the quality of service S and the number of service demanding devices 140). That is, the service providing device 130 can estimate a necessary budget without information about other data broker devices 120 in the market.

7 is a diagram for explaining an operating algorithm of an electronic device 400 according to various embodiments.

Referring to Figure 7, based on the analysis of the service providing device 130, for obtaining the main results (ie, data set quality (qi ^* ) and budget (bi ^* ) allocated to each data broker device 120) An algorithm (algorithm 1) is proposed. The algorithm is a basic input parameter, the number of service demand devices 140 (M), service quality (S) parameters (α, β, η), the number of data broker devices 120 (R), each data Unit price of the broker device 120 ( ) is taken. Based on the input parameters, the algorithm initializes common variables (X and Y (theorem 3)) and then calculates a total budget (B*) maximizing the profit of the service providing device 130 . Total budget (B*) and unit price of each data broker device (120) Based on , the data set quality required for each data broker device 120 (qi ^* (Theorem 1)) and the allocated individual budget (bi ^* (Equation 1)) are calculated. Through this, the service providing device 130 provides the data set quality required for all data broker devices 120 ( ) and allocated individual budgets ( ) can be obtained.

8 is a diagram illustrating an electronic device (eg, the data broker device 120) 800 according to various embodiments.

Referring to FIG. 8 , an electronic device 800 according to various embodiments may be the data broker device 120 of FIGS. 1 and 2 . The electronic device 800 may include at least one of a communication module 810 , an input module 820 , an output module 830 , a memory 840 , and a processor 850 . In some embodiments, at least one of the components of the electronic device 800 may be omitted, and at least one other component may be added. In some embodiments, at least two of the components of the electronic device 800 may be implemented as a single integrated circuit.

The communication module 810 may communicate with external devices (eg, data providing devices 110 and service providing devices 130 ) in the electronic device 800 . The communication module 810 may establish a communication channel between the electronic device 800 and an external device, and perform communication with the external device through the communication channel. The communication module 810 may include at least one of a wired communication module and a wireless communication module. The wired communication module may be connected to an external device through wired communication. The wireless communication module may communicate with an external device through a network.

The input module 820 may input a signal to be used in at least one component of the electronic device 800 . For example, the input module 820 may include at least one of a microphone, mouse, or keyboard.

The output module 830 may output information of the electronic device 800 . The output module 830 may include at least one of a display module and an audio output module. The display module may include, for example, at least one of a display, a hologram device, and a projector. The audio output module may include, for example, at least one of a speaker and a receiver.

The memory 840 may store various data used by at least one component of the electronic device 800 . For example, the memory 840 may include at least one of volatile memory and non-volatile memory. The data may include at least one program and related input data or output data. The program may be stored as software including at least one command in the memory 840, and may include, for example, at least one of an operating system, middleware, and applications.

The processor 850 may control at least one component of the electronic device 800 by executing a program of the memory 840 . Through this, the processor 850 may perform data processing or calculation. At this time, the processor 850 may execute instructions stored in the memory 840. The processor 850 may sell personal data of at least one of the data providing devices 110 to the service providing device 120 . To this end, the processor 850 may compete with at least one other data broker device 120 . The processor 850 may collect personal data from at least one of the data providing devices 110 . According to an embodiment, the processor 850 may purchase personal data from at least one of the data providing devices 110 . Here, the processor 850 may collect at least one predetermined type of personal data.

9 is a diagram illustrating an operating method of an electronic device 800 according to various embodiments.

Referring to FIG. 9 , the electronic device 800 may detect budget allocation of the service providing device 130 in step 910 . That is, the processor 850 may detect that the budget of the service providing device 130 is allocated to the electronic device 800 . At this time, an individual budget for the electronic device 800 may be allocated by the service providing device 130 . Correspondingly, the service providing device 130 may request a bid from the electronic device 800 . Accordingly, the processor 850 may prepare a bid for the service providing device 130 in response to budget allocation of the service providing device 130 .

In step 920, the electronic device 800 may predict expected revenue for the personal data to be sold. In this case, the processor 850 may predict expected revenue based on at least one of a budget allocated from the service providing device 130, that is, an individual budget for the electronic device 800 or a unit price of personal data. Here, the unit price may be determined based on at least one of the number of at least one data providing device 110 providing personal data or privacy sensitivity of personal data. Here, the higher the number of data providing devices 110, the higher the unit price may be determined. Meanwhile, the higher the privacy sensitivity, the higher the unit price may be determined.

The electronic device 800 may bid with the service providing device 130 in step 930 . To this end, the processor 850 may purchase personal data from the data providing device 110 . At this time, the processor 850 may purchase the personal data based on the unit price of the personal data. That is, the processor 850 may pay the data providing device 110 a unit price for personal data and purchase personal data sold by the data providing device 110 . Also, the processor 850 may bid the service providing device 130 for personal data. At this time, the processor 850 may determine a bid unit price of the personal data based on the expected profit. Here, the bid unit price must not exceed the individual budget for the electronic device 800 . Through this, the processor 850 may bid for personal data while submitting a bid unit price to the service providing device 130 .

The electronic device 800 may sell personal data to the service providing device 130 based on the bidding result in step 940 . That is, if the service providing device 130 succeeds in bidding, the processor 850 may sell personal data to the service providing device 130 . In this case, when the service providing device 130 pays an individual budget or bid unit price allocated to the electronic device 120, the processor 850 may provide individual data to the service providing device 130.

Actual revenue considering the quality of the data set (qi) for each data broker device 120 and the privacy sensitivity of the data providing device 110 ( ) and the WTS function of the data providing device 110, the optimization problem of the data broker device 120 is explained. Income can be obtained as a price (NE result) from the service providing device 130, and the result can be obtained as a cost (ci) of purchasing personal data from the data providing device 110 through the WTS.

A. Estimated Cost of WTS and Data Broker Device 120

The data broker device 120 must minimize cost by estimating the number of data providing devices 110 participating in the actual market. Therefore, the WTS of the data providing devices 110 should be modeled.

Since each data providing device 110 has its own WTS with different privacy concerns, it is difficult to formulate the operation of each data providing device 110 directly. Therefore, similar to the WTP, the WTS should be defined at the macro level, that is, the WTS cumulative distribution for all data providing devices 110 . Two principles apply to define WTS. The first is that the more money is provided, the more people participate. The second is the privacy sensitivity of data. People are hesitant to share or sell privacy-sensitive data (e.g., many people believe that information about credit card usage is more privacy-sensitive than information about location). Moreover, even if people deal with the same type of data, the WTS may be different depending on the characteristics of the data broker device 120 (eg, popularity, reputation, etc.).

Definition 7 : The WTS for a specific type of personal data (k ∈ K) from the data providing device 110 of the data broker device 120 is defined as in [Equation 23] below.

Here, ρi,k is a parameter for privacy sensitivity, and ci,k is a unit price provided for the k-th data type for the group of i-th data providing devices 110, respectively.

The WTS affects the characteristics of the data broker device 120 as well as the characteristics of personal data. According to the WTS, the amount (ni,k) of the k-th data type collected from the group Ni of the i-th data supply devices 110 by the data broker device 120 is defined as in [Equation 24] below .

The unit price of the kth data type is ci,k/ni,k. And, the total estimated cost (Ei) for purchasing the personal data of the i-th data broker device 120 is as follows [Equation 25].

here, am.

B. Data Set Quality of Data Broker Device 120

From the perspective of the data broker device 120, a method is needed to measure and quantify the value of collected data sets to determine whether there are enough data sets to sell in the marketplace. Thus, the concept of a quality measure for a data set is defined taking into account both inherent data quality and characteristics of data similar to Definition 4.

Definition 8: The quality of the data set provided by the data broker device 120 along with the group Ni of the data supply devices 110 is as follows [Equation 26].

Here, ni,x and ni,y are expected amounts of the x-th and y-th data types that can be collected by the data broker device 120, respectively. r _xy is the degree of correlation between data types. The geometric mean is used to quantify the quality of data sets to account for big data characteristics (eg accuracy, completeness, etc.) as well as privacy data characteristics similar to Definition 4 (eg distinctiveness, etc.). As a result, the proposed model can measure not only the amount of data itself but also the synergistic effect with the degree of correlation. If r _xy = 1 (x = y) and r _xy = 0 (x ≠ y), form

C. Revenue Optimization of Data Broker Device 120

The actual revenue of the data broker device 120 ( ) is defined as in [Equation 27] below.

Since the maximum budget (bi ^* (qi ^* )) is determined between the data broker device 120 and the service providing device 130, the profit function of the data broker device 120 can be converted as shown in [Equation 28] below. there is.

Thereafter, the revenue maximization problem can be converted to cost minimization of the data broker apparatus 120 . Thus, the cost minimization problem is solved. In the definition of the cost Ei of the data broker device 120 (Equation 25 above), the optimization problem may be defined as in [Equation 29] below.

Problem 2 :

here, is the data cost vector that minimizes the cost (Ei(ci ^* )). qi ^* is obtained from [Equation 8] above. Since the optimization problem is bounded in [Equation 29] above, a limited nonlinear optimization algorithm (eg, sequential quadratic programming (SQP)) that satisfies the condition can be applied.

Since WTS is an asymptotic function, it is necessary to define a visible problem space for the proposed cost minimization problem. Regarding the WTS, the condition of [Equation 29] is obtained through the assumptions described below.

Assumption 1 (upper bound of unit price for the kth data type): To consider the visible problem space for cost minimization, the upper bound on cost per data type is 2/ρ, so that the WTS value is about 86% (i.e. i ∈ At R and k ∈ K ) to be That is, about 14% of the data providing devices 110 reject the privacy evaluation method.

1) Privacy Principled (PF): The group merely rejects any proposals regarding privacy assessments (25% of the public).

2) Privacy Pragmatists: This group values the value between privacy assessment and privacy protection (57% of the public).

3) No Privacy: Groups are less concerned about invasion or abuse of privacy (18% of the public).

Assumption 1 provides intuition for finding an expected unit price (δi) for obtaining a data set as a quality criterion of the data broker device 120 for the bidding market. Since each data broker device 120 selects a bid without being aware of the activity of others, each data broker device 120 has all possible scenarios (e.g., the service providing device 130 will provide from the data broker device 120). purchase of the entire data set available). That is, the maximum capability of each data broker device 120 should be considered. Therefore, based on Assumption 1, Assumption 2 described below is obtained.

Assumption 2: The unit price δi of the data set having the expected quality qi of the data broker device 120 can be obtained as shown in Equation 30 below.

Assumption 2 provides a unified unit price for each data broker device 120 even if it has different numbers of data supply devices 110 with different data types. These factors can be used to measure the relative competitiveness of each data broker device 120 in the market, and the service providing device 130 can predict the capability of the data broker device 120. Here, the unit price (δ) of the data set is determined by Assumption 1.

The non-cooperative competition data trading system 100 according to various embodiments includes a plurality of data broker devices 120 configured to compete to sell at least one personal data among a plurality of data providing devices 110, respectively; and a service providing device 130 configured to allocate a budget to each of the data broker devices 120 and purchase personal data sold from at least one of the data broker devices 120 using the allocated budget. can do.

According to various embodiments, each of the data broker devices 120 may be configured to bid on personal data for sale to the service providing device 130 based on an allocated budget.

According to various embodiments, each of the data broker devices 120 predicts the expected revenue for the personal data to be sold based on the allocated budget and the unit price of the personal data to be sold, and determines the predicted expected revenue. Based on this, it may be configured to bid on personal data to be sold to the service providing device 130 .

According to various embodiments, the service providing device 130 may be configured to provide a service related to purchased personal data to at least one of the plurality of service demand devices 140 .

According to various embodiments, the service providing device 130 based on at least one of the number of at least one service demand device 140 for providing the service, the price for the service, or the expected quality of the service, It may be configured to predict expected revenue for the service, determine a budget based on the predicted expected revenue, and allocate the determined budget to each of the data broker devices 120 .

According to various embodiments, the data broker devices 120 may be configured to each purchase personal data from at least one of the plurality of data providing devices 110 .

A non-cooperative competitive data transaction method according to various embodiments includes the step of allocating, by the service providing device 130, a budget to each of a plurality of data broker devices 120, the data broker device 120, each of which is allocated to the budget. Based on this, the step of bidding on personal data to be sold to the service providing device 130, and the individual sold from at least one of the data broker devices 120 using the budget allocated to the service providing device 130. It may include purchasing data.

According to various embodiments, the step of bidding on personal data to be sold may include bidding on personal data to be sold based on a budget allocated to each of the data broker devices 120 and a unit price of personal data to be sold. Predicting expected revenue, and each of the data broker devices 120 bidding on personal data to be sold to the service providing device 130 based on the predicted expected revenue.

According to various embodiments, the non-cooperative competitive data transaction method may further include providing a service related to the purchased personal data to at least one of the plurality of service demand devices 140 by the service providing device 130. there is.

According to various embodiments, the step of allocating a budget may include among the number of at least one service demand device 140 for the service providing device 130 to provide the service, the price for the service, or the expected quality of the service. Based on at least one, predicting the expected revenue for the service, determining a budget based on the expected revenue by the service providing device 130, and setting the determined budget to each of the data broker devices 120. It may include the step of allocating to.

According to various embodiments, the uncooperative competitive data trading method may further include the step of each data broker device 120 purchasing personal data from at least one of the plurality of data providing devices 110 .

According to various embodiments, as the competitive data trading system 100 is implemented with a plurality of data broker devices 120 and a single service providing device 130, the data broker devices 120 and the service providing device 130 ) can effectively trade personal data according to uncooperative competition. Specifically, the data broker devices 120 sell personal data in consideration of the willingness-to-sell (WTS) of the data providing devices 110, and the service providing device 130 considers the WTP of the service demand devices 140. Personal data can be purchased on a willingness-to-pay basis. At this time, based on the competition between the data broker devices 120 and the competition between the data broker devices 120 and the service providing device 110, the competitive data trading system 100 effectively trades personal data according to uncooperative competition. can do. Accordingly, within the competitive data trading system 100, the data broker devices 120 and the service providing device 130 can stably trade personal data while maximizing profits from personal data transactions.

The devices described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in the embodiments include a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), and a programmable PLU (programmable logic unit). logic unit), microprocessor, or any other device capable of executing and responding to instructions. The processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. You can command the device. The software and/or data may be embodied in any tangible machine, component, physical device, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. there is. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

Methods according to various embodiments may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer readable medium. In this case, the medium may continuously store a program executable by a computer or temporarily store the program for execution or download. In addition, the medium may be various recording means or storage means in the form of a single or combined hardware, but is not limited to a medium directly connected to a certain computer system, and may be distributed on a network. Examples of the medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROM and DVD, magneto-optical media such as floptical disks, and ROM, RAM, flash memory, etc. configured to store program instructions. In addition, examples of other media include recording media or storage media managed by an app store that distributes applications, a site that supplies or distributes various other software, and a server.

Various embodiments of this document and terms used therein are not intended to limit the technology described in this document to a specific embodiment, and should be understood to include various modifications, equivalents, and/or substitutes of the embodiment. In connection with the description of the drawings, like reference numerals may be used for like elements. Singular expressions may include plural expressions unless the context clearly dictates otherwise. In this document, expressions such as "A or B", "at least one of A and/or B", "A, B or C" or "at least one of A, B and/or C" refer to all of the items listed together. Possible combinations may be included. Expressions such as "first," "second," "first," or "second" may modify the elements in any order or importance, and are used only to distinguish one element from another. The components are not limited. When a (e.g., first) element is referred to as being "(functionally or communicatively) connected" or "connected" to another (e.g., second) element, it is referred to as being "connected" to the other (e.g., second) element. It may be directly connected to the component or connected through another component (eg, a third component).

The term "module" used in this document includes a unit composed of hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit, for example. A module may be an integral part or a minimum unit or part thereof that performs one or more functions. For example, the module may be composed of an application-specific integrated circuit (ASIC).

According to various embodiments, each component (eg, module or program) of the described components may include a singular object or a plurality of entities. According to various embodiments, one or more components or steps among the aforementioned components may be omitted, or one or more other components or steps may be added. Alternatively or additionally, a plurality of components (eg modules or programs) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to integration. According to various embodiments, steps performed by a module, program, or other component are executed sequentially, in parallel, iteratively, or heuristically, or one or more of the steps are executed in a different order, omitted, or , or one or more other steps may be added.

Claims (10)

In the non-cooperative competition data trading method,
Collecting and storing personal data by a plurality of data providing devices;
Allocating, by a service providing device, a budget to each of a plurality of data broker devices;
purchasing personal data sold from at least one of the data providing devices, based on the allocated budget, by each of the data broker devices;
each of the data broker devices bidding on the personal data to be sold to the service providing device; and
Purchasing, by the service providing device, personal data sold from at least one of the data broker devices using the allocated budget
including,
at least two of the data broker devices compete to purchase the same personal data;
A method in which at least two of the data broker devices compete to purchase the same personal data from the same data providing device.
The method of claim 1, wherein bidding on personal data for sale comprises:
predicting, by each of the data broker devices, an expected profit for the personal data to be sold based on the allocated budget and the unit price of the personal data to be sold; and
and each of the data broker devices bidding for the personal data to be sold to the service providing device based on the predicted expected revenue.
According to claim 1,
The method further comprising providing, by the service providing device, a service related to the purchased personal data to at least one of a plurality of service demand devices.
The method of claim 1, wherein allocating the budget comprises:
predicting expected revenue for the service based on at least one of the number of at least one service demand device for providing the service, a price for the service, and an expected quality of the service;
determining, by the service providing device, the budget based on the predicted expected revenue; and
and allocating the determined budget to each of the data broker devices.
delete In the non-cooperative competitive data trading system,
A plurality of data providing devices that collect and store personal data;
a plurality of data broker devices configured to compete to each sell personal data of at least one of the data providing devices; and
A service providing device configured to allocate a budget to each of the data broker devices and purchase personal data sold from at least one of the data broker devices using the allocated budget;
Each of the data broker devices,
Purchasing personal data sold from at least one of the data providing devices based on the allocated budget;
configured to bid on personal data for sale to the service providing device;
at least two of the data broker devices compete to purchase the same personal data;
A system in which at least two of the data broker devices compete to purchase the same personal data from the same data providing device.
The method of claim 6, wherein each of the data broker devices,
Based on the allocated budget and the unit price of the personal data to be sold, expected revenue for the personal data to be sold is predicted;
A system configured to bid for the personal data for sale to the service providing device based on the predicted expected revenue.
The method of claim 6, wherein the service providing device,
A system configured to provide a service related to the purchased personal data to at least one of a plurality of service-demanding devices.
The method of claim 6, wherein the service providing device,
Predicting expected revenue for the service based on at least one of the number of at least one service demand device for providing the service, the price of the service, or the expected quality of the service;
Based on the predicted expected revenue, determine the budget;
A system configured to allocate the determined budget to each of the data broker devices.
delete