KR101628924B1 - Method and system for calculating k-th element score - Google Patents

Abstract

The method includes the steps of: (a) receiving encrypted first attribute data of a first item of a plurality of items with a public key of a first device; (b) (C) transmitting data based on the encrypted first attribute data, the encrypted second attribute data, and the estimate to the first device, 2 attribute relates to a score calculation method and a calculation system of a k-th item that represent different attributes in the first item.
By using the present invention, it is possible to find the k-th item with the same number of algorithm execution times as that of the known technique, which is safe for open-access attacks and multi-party.

Description

METHOD AND SYSTEM FOR CALCULATING K-TH ELEMENT SCORE "

More particularly, the present invention relates to a score calculation method and a calculation system for a k-th item, which enables finding a top k-th item while protecting data privacy of each other in a multi- .

Data mining is a technique for extracting meaningful information from large amounts of data. The extracted information may reveal the relationship between the data or imply the contents of the entire data. Because of this feature, it is used in various fields of real life.

Finding the kth item among the data mining algorithms is to estimate the upper kth item from the total numerical data of the data miners. These algorithms are mainly applied to vertically divided data. Vertically segmented data refers to data sets in which each data miner has the same data set, but minors have different attributes in the set.

For example, in the case of the Korean language subject in the first grade and the person in charge of the mathematics course, the subject has the data of his / her subject, but since he does not have the data of the other subject, have. Finding the kth item in the vertically divided data means finding the item having the highest kth score in the total sum data of the Korean and mathematical scores, or finding the kth score. In this environment, privacy mismatches can occur when data mining is applied.

For example, when a company conducts an employee review for employees, it can develop into a personal privacy issue if the employee's score is revealed. Data mining techniques to solve this problem are privacy-protected data mining. Privacy Protection Data mining is to change the distribution of original data used in existing data mining, or to protect the sensitive information of data by using cryptographic techniques, and to make the same as the result of mining of original data.

For the first time, use Yao's method of comparison in the paper (J. Vaidya, and CW Clifton, K-th Element Score over Vertically Partitioned Data, Knowledge and Data Engineering, vol. 21, No. 2, pp. 253-258, We propose a method to find the kth item of privacy guarantee. However, there were disadvantages in that the technique only had two possible minors.

Therefore, if data mining is performed by several people searching for the kth item at the same time, there is a disadvantage that the algorithm must be performed several times and there is a drawback that it is vulnerable to open call attacks.

Therefore, there is a need for a score calculation method and a calculation system of the k-th item that is safe for open access attacks and that allows the k-th item to be found with the same algorithm execution count as that of the existing known technique.

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above problems, and it is an object of the present invention to provide a score calculation method and a calculation method of a kth item, The purpose of the system is to provide.

According to another aspect of the present invention, there is provided a method of calculating a score of a k-th item, the method comprising the steps of: (a) receiving encrypted first attribute data of a first item among a plurality of items using a public key of the first device; (C) encrypting the data based on the encrypted first attribute data, the encrypted second attribute data, and the estimated value with the first device and the second device, , Wherein the first attribute and the second attribute represent different attributes in the first item.

To achieve the above object, a score calculation system of a k-th item receives encrypted first attribute data of a first item of a plurality of items with a public key of a first device, And a score calculation device that encrypts data based on the second attribute data and the estimated value corresponding to the first attribute data, the encrypted second attribute data, and the encrypted second attribute data, and transmits the data based on the estimated value to the first device, 1 attribute and the second attribute indicate different attributes in the first item.

The score calculation method and the calculation system of the kth item according to the present invention are effective in the open call attack and have the effect of finding the kth item with the same algorithm execution count as that of the known technique even in a multi-party.

1 is a diagram illustrating an exemplary system diagram of a kth item score calculation system.
FIG. 2 is a diagram showing a data flow between devices collaborating to calculate the k-th item.
Fig. 3 is a diagram showing a subsequent data flow between cooperating apparatuses for calculation of the k-th item; Fig.

The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings, in which: It can be easily carried out. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating an exemplary system diagram of a kth item score calculation system.

According to FIG. 1, the score calculation system includes a plurality of devices 100 and preferably includes three or more devices 100. Each device 100 is connected to each other via a network and can transmit and receive data for calculating a score of a k-th item through a network.

Referring to each of the apparatuses 100 of FIG. 1, each of the plurality of apparatuses 100 is configured to collaborate with each other to calculate a score of a k-th (k is one or more) elements. Each device 100 may be referred to by other terms depending on the function provided. The device 100 providing the public key to be used for encryption may be referred to as a 'public key provision device' and may be referred to as a 'cryptographic device', which is a device 100 that encrypts an attribute of an item possessed by the device 100, Lt; / RTI > Further, the apparatus 100 for calculating the score of the k-th item may be referred to as a " score calculation apparatus ".

Each of the devices 100 may provide the public key according to the function given in cooperation with the score calculation device 100 or may encrypt the data according to the public key to provide the encrypted data to the other device 100 or the score calculation device 100 . The score calculation of the kth item may be made over several rounds and the function assigned to each device 100 in each round is changeable. Therefore, the public key providing apparatus 100 in the first round can function as the encrypting apparatus 100 or the score calculating apparatus 100 in the second round.

Each of the devices 100 includes an input interface, an output interface, a memory, a mass storage medium, a communication interface, and a processor. Briefly looking at the components of each device 100, the input interface is configured to receive user input. For example, the input interface may include a mouse, a keyboard, a microphone, etc. to receive user input. The device 100 may receive a calculation request of the kth item via the input interface. The output interface is configured to output a processing result according to a user input or a processing result by a program. For example, the output interface includes a display, a speaker, an LED, and the like to output a processing result.

The memory of device 100 may include volatile memory and may further include non-volatile memory. The memory can temporarily store or permanently store various data and program codes. The mass storage medium stores various data and programs. The mass storage medium stores at least the calculation program for calculating the score of the kth item according to the present invention and the attribute values to be used for calculating the score of the kth item. The calculation program cooperates with the other apparatus 100 to calculate the score of the k-th item and outputs the calculation result.

The attribute value stored in the mass storage medium is stored for each item together with the identifier of the item. The attribute value may be, for example, a score value of a specific subject, or a score of a particular field of a particular management subject. The property (e.g., Korean word score) corresponding to the attribute value stored in one device 100 is different from the attribute (e.g., English score) of the attribute value stored in the other device 100, Thereby constructing vertical division data. The data set managed by the device 100 is composed of a plurality of items. The data set is configured to identify at least an identifier of the item and an attribute value of the corresponding item for each item. The dataset may be organized in the form of a list. The identifier of an item may identify students, for example, or identify personnel, subjects, and the like. For the sake of a brief understanding of the explanation, it is assumed that the identifier of the item is represented by the following number (i) and has the index (the index of 1 to 2000 when the number of persons is 2000) as much as the number of persons to be managed.

The communication interface of the device 100 is connected to a network to send and receive network packets. The communication interface may be connected to another device 100 via a wireless LAN or a wired LAN, for example, and may exchange network packets of data with each other.

The processor of the apparatus 100 loads programs and data stored in a mass storage medium according to a user input through an input interface or a request through a communication interface, and controls the blocks by executing a program. For example, the processor may calculate the score of the kth item by loading the calculation program into a volatile memory or the like, and controlling the blocks according to the program code of the calculation program. k is used to find the kth largest (or smallest) sum value in the summed sort set of attribute values for each item scattered in the devices 100 of the data set (e.g., sorted in descending order) . k may be specified according to user input through the input interface. The processor may be a so-called CPU, MPU, or the like. The control flow of the processor will be described in detail with reference to FIG. 2 and FIG.

The device 100 may be, for example, a personal computer, a notebook, a tablet PC, a mobile phone, a smart phone, a server, and the like. A plurality of devices 100 need not all be the same type of device 100. For example, one device 100 may be a personal computer, the other device 100 may be a notebook, and the other device 100 may be a smart phone, a server, or the like.

The devices 100 are connected to each other via a network, which may be a local area network or a broadband network. The local area network is composed of, for example, a wired LAN and / or a wireless LAN. A broadband network includes one or more local networks. Network packets over a transport layer can be exchanged between the devices 100 through the network.

FIG. 2 is a diagram showing a data flow between devices 100 that collaborate to calculate the k-th item. The data flow of FIG. 2 is an illustration of an example of three devices 100 collaborating. In the data flow of FIG. 2, each device 100 is preferably configured to transmit and receive data to and from the other devices 100 using the processor and communication interface of each device 100.

The data flow of FIG. 2 preferably begins with a user input via an input interface at a particular device 100. k, the devices 100 provide the functions of the devices 100 and proceed with a setup process for sharing the setup data (see (1)). The granting of the function is performed by any device (100). For example, the device 100 receiving the user input may give the functions of the devices 100, or the designated specific device 100 may give the functions of the devices 100. [ The score of the k-th item represents, for example, the sum of the k-th items of the sum of the attribute values of the items in the data set.

During the setup process, a particular device 100 is set to the public key provision device 100 (device A of FIG. 2) and the other device 100 is set to the encryption device 100 B) and another apparatus 100 is set to the score calculation apparatus 100 (apparatus C in FIG. 2).

In the setup process, each device 100 generates a public key and a private key. In particular, the public key providing apparatus 100 transmits the public key of the apparatus 100 to the other apparatuses 100. [ For example, assuming that the maximum value that the score of the kth item can have is | F | (300 points when the attribute of each device 100 is a national language score, an English score, or a math score) The public key n is 2 * | F | Is set to a larger value. The public key is set for each device 100 and is a product of the prime numbers p and q. The public key is set to be at least 2 * | F | due to the estimate used to prevent exposure of each attribute value in the kth item score calculation process.

The score calculation apparatus 100 designated in the setup process sets an upper bound (also referred to as ubound), a lower bound (also referred to as lbound) and an estimate, And transmits the upper limit value and the lower limit value to the other apparatuses 100. [ At least the score calculation device 100 transmits the estimated value to the public key providing device 100. [

)에서 난수값을 선택한다. 초기의 난수 범위는 데이터 집합의 항목들의 개수에 기초해서 결정되며 k번째 스코어 산출을 위해 반복이 예상되는 최대 라운드 개수에 기초해서 설정된다. 초기의 난수 범위는 항목들의 개수와 최대 라운드 개수에 따라 예를 들어 +/- 10으로 설정되고, 스코어 산출 장치(100)는 초기의 난수 범위에서 임의의 난수값을 선택하고 난수값을 상한값과 하한값의 중간값에 적용(합산)하여 추정값을 설정한다.The score calculation apparatus 100 first sets the upper limit value to the maximum value (| F |) and sets the lower limit value to the minimum value (for example, 0) of the sum of the attributes for the score calculation. The score calculation apparatus 100 sets an estimation value and transmits it to at least the public key distribution apparatus 100. The score calculation apparatus 100 calculates a random number range

) ≪ / RTI > The initial random number range is determined based on the number of items in the data set and is set based on the maximum number of rounds for which iteration is expected for the kth score calculation. The initial random number range is set to, for example, +/- 10 according to the number of items and the maximum number of rounds, and the score calculation device 100 selects an arbitrary random number value in the initial random number range and outputs the random number value as an upper limit value and a lower limit value (Summed) to the median value of < / RTI >

, 공개키는

, 속성값은

임.)하고 셋업과정에서 설정된 후속하는 암호화 장치(100)로 암호화된 속성 데이터를 전송(③ 참조)한다. After completion of the setup process, the public key providing apparatus 100 encrypts the attribute value of each item (i) of the data set included in the mass storage medium with its own public key

, The public key is

, The attribute value is

And transmits the encrypted attribute data to the subsequent encryption device 100 set in the setup process (see (3)).

)을 암호화하고 암호화된 데이터(

)와 수신된 암호화 데이터를 곱하여 전송할 속성 데이터를 생성(④ 참조)한다. 생성된 속성 데이터는 물론 공개키로 암호화되어 있고 적어도 특정 항목에 대한 암호화 장치(100)에서의 속성값과 공개키 제공 장치(100)에서의 속성값을 포함한다. 각각의 속성값은 서로 다른 속성에 대한 속성값을 나타낸다. The encryption apparatus 100 receives the encrypted attribute data and stores the attribute value of the item (i) stored in the mass storage medium in correspondence with the attribute data encrypted with the blank key of the public key providing apparatus 100

) And encrypts the encrypted data (

) And the received encrypted data to generate attribute data to be transmitted (see (4)). The created attribute data is of course encrypted with a public key and includes at least attribute values in the encryption apparatus 100 and attribute values in the public key providing apparatus 100 for a specific item. Each attribute value represents an attribute value for a different attribute.

Thereafter, the encryption apparatus 100 transmits the encrypted attribute data generated through (4) to the score calculation apparatus 100, and the score calculation apparatus 100 receives the attribute data (see (5)).

)을 암호화(

)한다. 또한 스코어 산출 장치(100)는 셋업 과정을 통해 설정된 추정값에 기초한 데이터를 암호화(

)한다. 이후 스코어 산출 장치(100)는 수신되고 암호화된 속성 데이터, 스코어 산출 장치(100)에 저장된 속성값의 암호화된 속성 데이터 및 암호화된 추정값의 데이터를 곱하여 공개키 제공 장치(100)로 전송할 데이터를 생성(⑥ 참조)한다. 추정값에 기초한 데이터('mean' 값이라고도 지칭됨)는 "- 추정값 mod n"으로 설정될 수 있고 n 은 스코어 산출 장치(100)의 공개키로 사용되는 값이다. 이 mean 값은 공개키 제공 장치(100)에서 이용되고 추정값보다 큰 값의 개수만을 파악하기 위한 용도로 이용되며 추정값과의 비교를 위해서 이용된다. 스코어 산출 장치(100)에 저장된 속성값의 속성과 수신된 속성 데이터에 포함된 속성들은 동일한 항목(i)에 대한 서로 다른 속성을 나타낸다. Upon receiving the encrypted attribute data, the score calculation apparatus 100 calculates the attribute value (i) of the item (i) stored in the mass storage medium in correspondence with the attribute data encrypted with the blank key of the public key providing apparatus 100

) To encrypt

)do. Also, the score calculation apparatus 100 may encrypt (decrypt) data based on the estimated value set through the setup process

)do. Thereafter, the score calculation device 100 multiplies the received and encrypted attribute data, the encrypted attribute data of the attribute value stored in the score calculation device 100, and the encrypted estimated value data to generate data to be transmitted to the public key providing device 100 (See ⑥). The data based on the estimate (also referred to as the 'mean' value) can be set to "-modified value mod n" and n is the value used as the public key of the score computing device 100. This mean value is used for the purpose of grasping only the number of values used in the public key providing apparatus 100 and larger than the estimated value and is used for comparison with the estimated value. The attributes of the attribute values stored in the score calculation device 100 and the attributes included in the received attribute data represent different attributes for the same item (i).

Then, the score calculation apparatus 100 transmits the generated data to the public key providing apparatus 100 (refer to (vii)).

The process of (1) through (7) is performed for each item of the data set (1 <= i <n, n is the index maximum value of the items) And is configured to receive encrypted data.

The public key providing apparatus 100 decrypts encrypted data for each item with its own private key.

Here, the encryption technique used in the present invention will be described. Encryption and decryption are performed using a cryptographic technique having a homographic property and preferably using Paillier's public key cryptography technique.

Paillier's public key cryptosystem is based on the difficulty of exponentiation, and unlike RSA, it can operate in encrypted state using the quasi-dynamic property. That is, Paillier's public key cryptosystem decodes each of the multiplied encrypted data so that the result of addition of the decrypted data can be known.

)를 추정값과 비교하고 각 항목의 비교 결과(output)를 합산(⑧ 참조,

)한다. 합산된 비교 결과는 스코어 산출 장치(100)로 전송(⑨ 참조)된다. According to the decryption and the perturbation property, the public key providing apparatus 100 decrypts the decrypted data (

) Is compared with the estimated value, and the output of each item is summed up (see ⑧,

)do. The summed comparison result is transmitted to the score calculation apparatus 100 (see (9)).

In detail, the decryption of the received data of each item extracts the sum value of the attribute value and the mean value (estimated value) of each device 100, and outputs the extracted sum value as n Public key) and compares the moduler-calculated data with the estimated value. If the sum value is larger than the estimated value, the comparison result (output value) is set to 1; otherwise, it is set to zero. The comparison results of the respective items are summed and transmitted to the score calculation apparatus 100. The result of the summed comparison is to know how many items are at least larger than the estimated value.

Upon receipt of the summed comparison result, the score calculation device 100 compares the summed comparison result with k (set in the setup process) and calculates the score of the kth item, or thereafter calculates the score of the kth item (See 10) and proceeds to a subsequent round (hereinafter referred to as "comparison result calculation round" or shortly referred to as "round") (see FIG. 3).

)을 k와 비교한다. 그 비교 결과에 따라 결과값이 k 보다 작으면 상한값을 mean 값(또는 추정값)으로 변경하고 상한값과 하한값의 차가 1이면 mean 값(또는 추정값)을 k번째 항목의 스코어로 결정한다. 만일 비교 결과의 결과값이 k보다 크면 하한값을 mean 값(또는 추정값)으로 변경한다. 만일 비교 결과 결과값이 k이면 mean 값(또는 추정값)을 k번째 항목의 스코어로 결정한다. 비교 결과의 결과값이 k보다 크거나 작은 경우(즉 k번째 항목의 스코어가 결정되지 못한 경우) 추정값이 또한 변경된다. Specifically, the score calculation apparatus 100 calculates the result value of the received summed comparison result (

) Is compared with k. If the result is smaller than k, the upper limit value is changed to the mean value (or the estimated value). If the difference between the upper limit value and the lower limit value is 1, the mean value (or estimated value) is determined as the score of the kth item. If the result of the comparison is greater than k, change the lower bound to mean (or estimate). If the result of the comparison is k, the mean value (or estimated value) is determined as the score of the kth item. If the result of the comparison is greater than or less than k (i.e., the score of the kth item is not determined), the estimate is also changed.

The modified estimate is used in a subsequent comparison result calculation round, and the estimated value to be changed is set by applying a random value. The score calculation device 100 first sets a random number range for a subsequent round of comparison result calculation. This random number range is narrower than the random number range used in the preceding round of comparison. For example, if you were in the range of +/- 10 in the previous round, the random number range is determined in the range (+/- 5), which is 1/2 in the following rounds. As described above, the random number range used in each comparison result calculation round is narrower than the random number range of the previous round and is always narrower than the range of the score value (the range of the upper limit value and the lower limit value), thereby facilitating the score calculation.

The score calculation apparatus 100 selects an arbitrary random number value within the newly set random number range and applies the selected random number value at the intermediate value between the changed upper and lower limit values to change the estimation value for the subsequent round. The changed upper limit value, the lower limit value, and the estimated value, as well as the random number range, can be transmitted to other devices 100. [

If the score of the k-th item is calculated, the score calculation device 100 transmits the score of the calculated k-th item to the other devices 100, and the flow of FIG. 2 does not proceed any more And is terminated. The score transmitted here makes it possible to distinguish at least the kth item from the (k + 1) th item.

The devices 100 (in particular, the device 100 requested to search for the kth item) can output the corresponding score through the output interface upon receipt of the kth item.

Fig. 3 is a diagram showing the subsequent data flow between the devices 100 collaborating for calculation of the k-th item. The data flow of FIG. 3 proceeds as a subsequent comparison result calculation round after completion of the data flow of FIG. The data flow of FIG. 3 is repeatedly performed when the score of the kth item is not determined in each comparison result calculation round, and the upper limit value, the lower limit value and the estimation values used in each iteration are changed.

The data flow of FIG. 3 is almost the same as that of FIG. 2, and therefore, description will be focused on contents different from FIG. The parts omitted from this description are replaced by the description of FIG.

For the execution of the subsequent round, the devices 100 collaborate with each other to perform the setup process (see (1)) to grant the functions of the respective devices 100 and share setup data.

Each device 100 that collaborates is given a function. One device 100 is set to the public key provision device 100 and the other device 100 is set to the encryption device 100 and the other device 100 is set to the score calculation device 100 ). At least the public key providing apparatus 100 in the previous round and the public key providing apparatus 100 in the current round can be set differently. For example, the public key providing apparatus 100 in the previous round is set to the encrypting apparatus 100, and the encrypting apparatus 100 in the previous round is set to the public key providing apparatus 100. The currently set public key providing apparatus 100 transmits the public key to the other apparatuses 100. [ The public key and the private key of each device 100 may be a public key and a private key generated in the first round.

The public key providing apparatus 100 that provides a public key for each round is set to a different apparatus 100, which is intended to prevent authorization from concentrating on the same apparatus 100. [ Particularly, if the public key of the same device 100 is continuously used in each round, it becomes vulnerable to an external attack. Also, by setting different devices 100 to the public key providing device 100, it is possible to distribute the computation amount of each device 100 for each row, and furthermore, the public key used in each round is different, Do.

After setting up the setup process, the public key providing apparatus 100 encrypts the attribute value of the stored item using its own public key, and transmits the encrypted attribute data to the other encryption apparatus 100 (see (2)).

Upon receiving the encrypted attribute data, the encryption apparatus 100 encrypts its own item attribute value with the received public key, multiplies the attribute data encrypted with the attribute value of the received and encrypted attribute data, And transmitted to the device 100 (see (3)).

The score calculation apparatus 100 receives the encrypted attribute data, encrypts data based on the item attribute value and the estimated value of the item, and transmits the multiplied data to the public key providing apparatus 100 (see (4)), do. The data based on the estimated value (mean value) can be set to "- estimated value mod n ", and n is a value used as the public key of the score calculation device 100. The attribute values encrypted in each device 100 represent values for different attributes for the same item (i).

The public key providing apparatus 100 receives the encrypted data in accordance with steps 1 to 4, decrypts the received data with the private key, compares the decrypted data with the mean value, and outputs the result of comparison, which is output as 1 or 0, do. The summed comparison result is transmitted to the score calculation apparatus 100 (see Fig. 5).

The score calculation apparatus 100 may compare the summed comparison result with k and change the upper limit value, the lower limit value and / or the estimated value according to the comparison result, and proceed to the subsequent round. The subsequent rounds are the same as in Fig. Or the score calculation device 100 transmits the score of the kth item to the other devices 100 when the kth item is found. The received scores may be output via the output interface of the devices 100. The score transmitted here makes it possible to distinguish at least the kth item from the (k + 1) th item.

2 and 3, it is possible to perform calculation for calculation of the k-th item without exposing data (attribute value) between each apparatus 100. [ For example, if you want to calculate the rank of a class, the person in charge of each course can calculate the rank without sharing the data (attribute value). In the case of the person who is promoted through the personnel affairs department, the information such as the performance and the working attitude can be synthesized without sharing between the persons in charge, and accordingly, the promotion can be judged.

In FIG. 2 and FIG. 3, it is possible to effectively cope with an open call attack in which two or more devices 100 (users) can obtain an attribute value of another device 100 by adding an arbitrary random number value. For example, even when two or more apparatuses 100 conspire, the attribute value of another apparatus 100 (particularly, the score calculation apparatus 100) can not be known by a random number value, and thus it is possible to cope with an open call attack effectively.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention. The present invention is not limited to the drawings.

100: Device

Claims (10)

(a) receiving an encrypted first attribute data, which is an encrypted value of an attribute value of a first item among a plurality of items stored in the mass storage device with a public key of a public key providing device;
(b) encrypting each of the second attribute data and the data based on the score estimation value obtained by encrypting the attribute value corresponding to the first attribute data with the public key of the public key providing apparatus; And
(c) transmitting, to the public key providing apparatus, data based on the encrypted first attribute data, the encrypted second attribute data, and the score estimation value,
Wherein the attribute value of the first item and the attribute value corresponding to the first attribute data are different from each other in the first item. The method according to claim 1,
Setting the score upper limit value, the score lower limit value and the score estimation value before the step (a); And
And transmitting the score estimate to the public key providing device,
Wherein the score estimation value is set by applying a random number value at an intermediate value between the score upper limit value and the lower score limit value. The method according to claim 1,
Wherein the steps (a) to (c) are performed by a score calculation device for each of a plurality of items,
The method may further include decrypting the data received through the step (c) with a private key of the public key providing apparatus,
Comparing the decrypted data with a score estimate; And
And transmitting the comparison result to the score calculation device. The method of claim 3,
Changing the score upper limit value, the score lower limit value and the score estimation value according to the reception of the comparison result; And transmitting the modified score estimate to a plurality of devices including the public key provision device,
Wherein the modified score estimate is set by applying a random number value at an intermediate value between the modified score upper limit value and the lower score value and wherein the random number value to be used in a subsequent comparison result calculation round is set for the steps (a) to (c) And calculating a score of a kth item selected from a random number range narrower than the set random number range. The method according to claim 1,
After the step (c)
The encrypted first attribute data, which is obtained by encrypting the attribute value of the first item among the plurality of items stored in the mass storage device using the public key of the public key providing device, which is another one of the plurality of devices including the public key providing device, Receiving;
Encrypting an attribute value corresponding to the first attribute data with a public key of the other public key providing apparatus, and encrypting the second attribute data and the data based on the score estimation value, respectively; And
And transmitting data based on the encrypted first attribute data, the encrypted second attribute data, and the score estimation value to the other public key providing apparatus. The method according to claim 1,
The encryption in steps (a) and (b) is performed using Paillier's public key cryptography with homographic properties,
The method of calculating the score of the kth item. Receiving encrypted first attribute data obtained by encrypting an attribute value of a first item among a plurality of items stored in the mass storage device using the public key of the public key providing device, And encrypts the encrypted second attribute data and the second encrypted data, based on the first attribute data, the encrypted second attribute data, and the score estimation value, to the public key providing apparatus And a score calculation device for transmitting the score,
Wherein the attribute value of the first item and the attribute value corresponding to the first attribute data are different from each other in the first item. 8. The method of claim 7,
The score calculation apparatus sets a score upper limit value, a score lower limit value and a score estimation value, and transmits a score estimation value to the public key provision apparatus before receiving the first attribute data,
Wherein the score estimation value is set by applying a random number value at an intermediate value between the score upper limit value and the lower score limit value. 8. The method of claim 7,
And a public key provision device connected to the score calculation device,
The public key provision apparatus decrypts the data received from the score calculation apparatus with the private key of the public key provision apparatus, compares the decrypted data with the estimated value, and transmits the comparison result to the score calculation apparatus. . 8. The method of claim 7,
Further comprising a public key providing apparatus connected to the score calculating apparatus and another public key providing apparatus,
Wherein the score calculation apparatus calculates the score based on the encrypted first attribute data, the encrypted second attribute data, and the data based on the score estimation value after the transmission to the public key provision apparatus by using the public key of the other public key provision apparatus The encrypted first attribute data, which is obtained by encrypting the attribute value of the first item among the plurality of items stored in the mass storage device, is received, and the attribute value corresponding to the first attribute data Encrypts the second attribute data and the data based on the score estimation value, encrypts the encrypted first attribute data, the encrypted second attribute data, and the data based on the score estimation value, The score calculation system of the kth item transmitted to the device.

Images