Information processing device, information processing system, and information processing method
The information processing device optimizes privacy budget allocation by evaluating noise influence and sample sizes to ensure effective privacy protection in multi-party data aggregation, addressing the challenge of noise impact on aggregated results.
Patent Information
- Application Number
- PCT/JP2024/017834
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-18
- Filing Date
- 2024-05-14
- Publication Date
- 2024-11-21
AI Technical Summary
When aggregating statistical data from multiple companies, determining the optimal allocation of privacy budget for each aggregation step is challenging due to the unknown influence of noise on the aggregated results, which complicates protecting user privacy.
An information processing device and method that acquires basic tabulation information, estimates sample sizes, evaluates the influence of noise, and determines privacy budget allocation for each aggregation step based on the evaluation results, ensuring privacy protection within a predetermined budget.
Optimizes privacy budget allocation across multiple aggregations, ensuring uniform noise influence and effective privacy protection, even when using pre-executed data or known values, thereby enhancing data security and accuracy in multi-party data sharing scenarios.
Smart Images

Figure JP2024017834_21112024_PF_FP_ABST
Abstract
Description
Information processing device, information system, and information processing method
[0001] The present disclosure relates to an information processing device, an information system, and an information processing method.
[0002] Leveraging data held by multiple companies promises to create value that cannot be obtained with data held by a single company. One method for achieving data utilization across multiple companies is to output aggregated results without disclosing the data held by multiple companies to each other. In this case, adding noise based on differential privacy standards to the aggregated results can protect the privacy information of the output data. However, unlike data utilization within a single company, when utilizing statistical data across multiple companies, the degree of influence of noise on the aggregated results is unknown. Therefore, when performing multiple aggregations on the same dataset, it is difficult to appropriately allocate the privacy budget for each aggregation. Therefore, there is a need for a method for appropriately allocating the privacy budget for each aggregation, taking into account the influence of the noise.
[0003] In relation to privacy budget allocation, the following non-patent document 1 discloses a technology for privacy budget allocation in a differential privacy use case among multiple parties, and shows an example in which a classical model of differential privacy is extended to define a maximization objective function that guarantees optimal privacy budget allocation among multiple parties.
[0004] "Optimal distribution of privacy budget in differential privacy," by Bkakria Anis et al., published October 16, 2018
[0005] However, Non-Patent Document 1 does not disclose a technique for appropriately determining the allocation of the privacy budget for each tally, taking into consideration the degree of influence of noise on the tally results.
[0006] Therefore, an object of the present disclosure is to optimize the allocation of privacy budgets when performing multiple aggregations on the same data set in the utilization of statistical data between multiple companies.
[0007] The information processing device of the present disclosure comprises a basic aggregation information acquisition unit that acquires basic aggregation information of the user's own device and the other device, each of which stores user data including a user ID and attribute information related to the user; an estimation unit that estimates a sample size of the aggregation result based on the basic aggregation information of the user's own device and the other device acquired by the basic aggregation information acquisition unit and a basic value used for estimating the sample size of the aggregation result; an evaluation unit that quantitatively evaluates the impact of noise for each combination included in an aggregation pattern that includes multiple combinations of data items included in the attribute information based on the sample size of the aggregation result estimated by the estimation unit; and an allocation determination unit that determines the allocation of a privacy budget to each combination related to noise for aggregation targeting the aggregation pattern within a predetermined privacy budget based on the evaluation results for each combination obtained by the evaluation unit.
[0008] In the information processing device described above, the basic tally information acquisition unit acquires basic tally information of the local device and the remote device, and the estimation unit estimates a sample size of the tally results based on the basic tally information of the local device and the remote device acquired by the basic tally information acquisition unit and a basic value used to estimate the sample size of the tally results. The evaluation unit quantitatively evaluates the influence of noise on each combination included in a tally pattern, which includes multiple combinations of data items included in the attribute information, based on the sample size of the tally results estimated by the estimation unit. The allocation determination unit determines, within a predetermined privacy budget range, the allocation of the privacy budget to each combination related to noise for tallying the tally patterns based on the evaluation results of each combination obtained by the evaluation unit. Since the influence of noise on each combination included in the tally pattern is quantitatively evaluated and the allocation of the privacy budget to each combination is determined based on the evaluation results of each combination obtained by the evaluation, the allocation of the privacy budget can be optimized.
[0009] According to the present disclosure, in the utilization of statistical data among multiple companies, it is possible to optimize the allocation of privacy budgets when performing multiple aggregations on the same data set.
[0010] 2. FIG. 1 is a configuration diagram of an information system according to a first embodiment. FIG. 2 is a flow diagram showing processing executed in the information system according to the first embodiment. FIG. 3 is a diagram for explaining example user data and a counting pattern. FIG. 4 is a diagram for explaining the processing of steps S1 to S8 of FIG. 2. FIG. 5 is a diagram for explaining data processing. FIG. 6 is a diagram for explaining irreversible ID conversion. FIG. 7 is a diagram for explaining ID encryption. (a) is a diagram for explaining encryption of attribute information, and (b) is a diagram for explaining transmission and reception of encrypted IDs. FIG. 8 is a diagram for explaining re-encryption of IDs. FIG. 9 is a diagram for explaining data matching. (b) is a diagram for explaining the processing of steps S9 to S12 of FIG. 2. FIG. 9 is a diagram for explaining the processing of steps S13 to S14 of FIG. 2. FIG. 1 is a configuration diagram of an information system according to a second embodiment. FIG. 10 is a flow diagram showing processing executed in the information system according to the second embodiment. FIG. 11 is a diagram for explaining a first example of allocation of a privacy budget. FIG. 12 is a diagram for explaining a second example of allocation of a privacy budget. FIG. 13 is a diagram for explaining the effects of the present invention. FIG. 14 is a configuration diagram of an information system according to a third embodiment. FIG. 14 is a flow diagram showing processing executed in the information system according to the third embodiment. FIG. 15 is a diagram for explaining counting method (a). FIG. 16 is a diagram for explaining counting method (b). FIG. 17 is a diagram for explaining selection of a counting method. Fig. 1 is a configuration diagram of an information system according to Modification 1. Fig. 2 is a configuration diagram of an information system according to Modification 2. Fig. 3 is a diagram illustrating an example of the hardware configuration of an information processing device.
[0011] Various embodiments according to the present disclosure will be described below with reference to the drawings. In the following, a first embodiment will be described in which a trial count is performed in advance before the actual count to calculate a base value such as a data matching rate at the time of counting, a second embodiment will be described in which a known value or an estimated value is used as the base value without performing the trial count in advance, and a third embodiment will be described in which a counting method that matches a predetermined purpose (for example, a counting method that can minimize the influence of noise) is selected.
[0012] (First embodiment) As shown in FIG. 1, an information system 1 in the first embodiment has a "Company A's device" (hereinafter referred to as "device 10A") as a first device and a "Company B's device" (hereinafter referred to as "device 10B") as a second device (the counterpart device of the first device).
[0013] The device 10A includes a data input unit 11, a pre-execution data generation unit 12, a basic value calculation unit 13, a basic compilation information acquisition unit 14A, an estimation unit 15, an evaluation unit 16, and an allocation determination unit 17, while the device 10B includes a data input unit 11, a pre-execution data generation unit 12, and a basic compilation information provision unit 14B. The functions of each unit will be outlined below, but details will be provided later with reference to the flow diagram in Figure 2.
[0014] The data input unit 11 is a functional unit that accepts input data from outside and is provided in common to the devices 10A and 10B. However, although the input data for both Company A and Company B is user data including a user ID and attribute information related to the user, the content of the user data differs between the companies. Specific examples of user data will be described later.
[0015] The pre-execution data generation unit 12 is a functional unit that generates data for pre-execution (referred to as "pre-execution data") when a trial tabulation for calculating basic values such as the data matching rate between the two companies, which will be described later, is performed in advance before the actual tabulation, and is provided in common with the devices 10A and 10B. The pre-execution data generation unit 12 includes a data processing unit 12A that processes input data into small-scale data for calculating the basic values, an anonymization processing unit 12B that performs an anonymization process to protect the privacy of attribute information in the user data, an ID irreversible conversion unit 12C that performs an irreversible conversion process to a user ID in the user data, an encryption unit 12D that encrypts the user data, and a data transmission / reception unit 12E that transmits and receives the encrypted user data.
[0016] The base value calculation unit 13 is a functional unit provided in the device 10A that calculates a base value from the pre-execution data A generated by the pre-execution data generation unit 12 of the device 10A and the pre-execution data B generated by the pre-execution data generation unit 12 of the device 10B. The base value calculation unit 13 includes a data matching unit 13A that matches the pre-execution data A with the pre-execution data B, a counting unit 13B that performs counting based on the matching results, a concealment processing unit 13C that performs concealment processing on the counting results, and a calculation unit 13D that calculates a base value from the counting results after the concealment processing.
[0017] The basic tally information providing unit 14B is a functional unit provided in the device 10B, which calculates the sample size and itemized ratios, which are basic tally information of the user data of the device 10B, and provides them to the device 10A.
[0018] The basic summary information acquisition unit 14A is a functional unit provided in the device 10A, which calculates the sample size and item-specific ratio, which are the basic summary information of the user data of its own device 10A, and acquires the basic summary information of the devices 10A and 10B by receiving the sample size and item-specific ratio, which are the basic summary information of the user data of the device 10B, calculated by the other device (device 10B).
[0019] The estimation unit 15 is a functional unit that estimates the sample size of the aggregation results based on the basic aggregation information of the devices 10A and 10B acquired by the basic aggregation information acquisition unit 14A and the basic value calculated by the basic value calculation unit 13, and is provided in the device 10A.
[0020] The evaluation unit 16 is a functional unit that quantitatively evaluates the influence of noise on each combination included in the counting pattern based on the estimated sample size of the counting results, and is provided in the device 10A.
[0021] The allocation determination unit 17 is a functional unit that determines the allocation of the privacy budget to each combination based on the evaluation result for each combination obtained by the evaluation unit 16, and is provided in the device 10A.
[0022] Next, the processing executed in the information system 1 will be described in detail with reference to FIGS.
[0023] First, in each of devices 10A and 10B, user data to be tabulated is input to data input unit 11 (steps S1 and S2 in FIG. 2). As shown in FIG. 3, the data of Company A input here includes a user ID, an item A indicating the user's gender as part of the user's attribute information, and an item B indicating the user's age group as part of the user's attribute information. The data of Company B includes a user ID, an item C indicating the user's address (here, whether the user lives in Tokyo or outside Tokyo) as part of the user's attribute information, and an item D indicating the user's purchase history (here, whether the user has purchased a certain product) as part of the user's attribute information. Furthermore, three tabulation patterns are assumed in this embodiment: a tabulation for a combination of items A and D (hereinafter referred to as "item A x D"), a tabulation for a combination of items B and D (hereinafter referred to as "item B x D"), and a tabulation for a combination of items B, C, and D (hereinafter referred to as "item B x C x D").
[0024] Next, in each of the devices 10A and 10B, the pre-execution data generator 12 generates the pre-execution data described above (steps S3 and S4 in FIG. 2). As shown in FIG. 2, the pre-execution data generation process includes multiple processing steps, which are roughly divided into data processing (steps S3A and S4A) for converting the data into small-scale data for calculating the base value, and pre-processing (steps S3B to S3G and S4B to S4H) for data matching (step S5) in a confidential manner.
[0025] For the former, in each of the devices 10A and 10B, the data processing unit 12A processes the input data into small-scale data for calculating the base value (steps S3A and S4A). Here, for example, "item D," which is common to the totals for the three combinations of the totalization patterns (item A x D, item B x D, and item B x C x D) in this embodiment, is considered the main item, and the data is processed into small-scale data with attribute information limited to item D only. As a result, as shown in Figures 4 and 5, the data for Company A is processed into only the user ID, and the data for Company B is processed into the user ID and item D, respectively.
[0026] The latter pre-processing includes anonymization processing, irreversible ID conversion, ID encryption, etc. In the anonymization processing, the anonymization processing unit 12B in each of the devices 10A and 10B performs anonymization processing on the user data held by the respective devices as a process for protecting the privacy of attribute information (steps S3B and S4B in FIG. 2 ). Here, k-anonymization is exemplified, which converts the user data so that k or more pieces of user data with the same attribute information exist within the target user data (satisfying k-anonymity), thereby reducing the probability of identifying an individual to 1 / k or less. If there are fewer than k pieces of user data with the same attribute information in the attribute information of Company B's user data, the attribute information of user data with similar attribute information is converted so that there are k or more pieces of user data.
[0027] 6, in each of the devices 10A and 10B, the ID irreversible conversion unit 12C performs an irreversible conversion process to convert the user data stored in the device into a user ID (steps S3C and S4C). Specifically, the ID irreversible conversion unit 12C performs a hashing process on the user ID, and then discards the salt used in the hashing process.
[0028] 7, in device 10A, encryption unit 12D encrypts the non-identifiable hash (the portion of the user data corresponding to the user ID) with a pre-prepared private key a to obtain ID-encrypted data for company A (step S3D). Similarly, in device 10B, encryption unit 12D encrypts the non-identifiable hash (the portion of the user data corresponding to the user ID) with a pre-prepared private key b to obtain ID-encrypted data for company B (step S4D).
[0029] 8(a), in device 10B, encryption unit 12D encrypts attribute information in the ID-encrypted data of company B obtained in step S4D with secret key B prepared in advance by a homomorphic encryption method that enables aggregation processing, thereby generating encrypted user data of company B (step S4E). Note that each piece of attribute information in the generated encrypted user data is configured using binary values in a predetermined format to enable aggregation processing in an encrypted state.
[0030] Next, as shown in FIG. 8(b), in device 10A, data transmitter / receiver 12E transmits the encrypted ID of Company A contained in the encrypted ID data of Company A obtained in step S3D to data transmitter / receiver 12E of device 10B (step S3E), and in device 10B, data transmitter / receiver 12E transmits the encrypted user data of Company B obtained in step S4E (i.e., data including the encrypted ID and encrypted attribute information of Company B) to data transmitter / receiver 12E of device 10A (step S4F).
[0031] 9, in device 10A, encryption unit 12D re-encrypts the encrypted ID of company B included in the encrypted user data of company B transmitted from device 10B using private key a, thereby obtaining the encrypted ID of company B encrypted with both private keys a and b (step S3F). Similarly, in device 10B, encryption unit 12D re-encrypts the encrypted ID of company A transmitted from device 10A using private key b, thereby obtaining the encrypted ID of company A encrypted with both private keys a and b (step S4G). Then, in device 10B, data transmitter / receiver 12E transmits the "encrypted ID of company A encrypted with both private keys a and b" obtained in step S4G to device 10A (step S4H), and data transmitter / receiver 12E of device 10A receives the encrypted ID of company A (step S3G).
[0032] Next, as shown in Fig. 10 , in device 10A, the data matching unit 13A replaces the "encrypted ID encrypted with private key a" obtained by the ID encryption in step S3D with the "encrypted ID of company A encrypted with both private keys a and b" received in step S3G, thereby obtaining the encrypted ID of company A shown in the upper left of Fig. 10 . The data matching unit 13A also replaces the "encrypted ID of company B encrypted with private key b" in the "encrypted user data of company B" received in step S3E with the "encrypted ID of company B encrypted with both private keys a and b" obtained by re-encryption in step S3F, thereby obtaining the encrypted user data of company B shown in the upper right of Fig. 10 . Then, the data matching unit 13A matches the encrypted user data of company A with the encrypted user data of company B using the "encrypted ID of company A encrypted with both private keys a and b" and the "encrypted ID of company B encrypted with both private keys a and b" as keys (step S5). Here, if "Company A's encrypted ID encrypted with both private keys a and b" and "Company B's encrypted ID encrypted with both private keys a and b" match, the attribute information in Company A's encrypted user data and the encrypted attribute information in Company B's encrypted user data are combined into a single record in the encrypted matching data. After the combination, "Company A's encrypted ID encrypted with both private keys a and b" and "Company B's encrypted ID encrypted with both private keys a and b" are deleted, resulting in encrypted matching data. However, in this embodiment, the attribute information in Company A's user data is not subject to pre-execution processing. Therefore, as shown in the lower part of Figure 10, the encrypted matching data is composed of Company B's encrypted attribute information. As described above, the encrypted attribute information is composed of binary values in a predetermined format and can be aggregated in an encrypted state. Of course, because Company A's attribute information is encrypted, Company A cannot know its contents. Therefore, the encrypted matching data is generated without revealing to Company A the contents of the attribute information that should be kept confidential from Company B.
[0033] 2 and 4, the data matching unit 13A matches the pre-execution data A with the pre-execution data B in the above-described manner, and the tallying unit 13B counts the number of records for "Purchase" and "Not Purchase" in item D for the matched records to obtain tallying results for "Purchase" and "Not Purchase" (step S6). Note that, as described above, the contents of item D, which is "Company B's encrypted attribute information" in the encrypted matched data, are encrypted, and therefore the contents cannot be known by Company A's device 10A. However, as described above, the "Company B's encrypted attribute information (item D)" in the encrypted matched data is composed of a predetermined binary value format, and therefore, in the encrypted state, it is possible to calculate the total number of records in the binary value format corresponding to "Purchase" and the total number of records in the binary value format corresponding to "Not Purchase."
[0034] The anonymization processing unit 13C then performs the following anonymization processing on the above tabulation processing results (step S7). For example, as shown in FIG. 4, the tabulation processing results for item D, in which "Purchase" is 1,884 and "Not Purchased" is 4,300, are targeted. Adding noise equivalent to consuming a privacy budget of 0.1 results in the anonymized tabulation processing results, as shown on the right side of FIG. 4, of "Purchase" being 1,900 and "Not Purchased" being 4,229. Furthermore, the calculation unit 13D calculates a base value from the tabulation processing results after the anonymization processing (step S8). Here, for example, a "data matching rate" and a "data bias" are calculated as the base values. The "data matching rate" refers to the percentage of records that were matched out of the total number of records that were subject to data matching, and a data matching rate of 40% is calculated. "Data bias" refers to the difference between the item-specific percentages for a given item derived from a single company's data and the item-specific percentages for that given item derived from data that can be matched between companies. Data bias exists for each item for which an item-specific percentage is calculated. However, due to privacy budget constraints, it is difficult to calculate data bias for all items in advance. Therefore, data bias is calculated here for item D, which is the main item for aggregation. For item D, an item-specific percentage ("purchase" percentage) of 31% is calculated by basic aggregation for Company B alone in step S10 of Figure 2, which is executed in parallel with steps S3 to S8 above. Meanwhile, as shown in Figure 4, the item-specific percentage (purchase percentage) derived from data that can be matched between Companies A and B is 31%, so a data bias of "0%" is calculated as the base value.
[0035] 2, in device 10A, basic tabulation information acquisition unit 14A calculates the sample size and item-specific percentages, which are basic tabulation information, for the user data of device 10A (step S9). For example, for tabulation (1) of item A × D, the sample sizes and percentages for each of males and females for item A (gender) in Company A's data are calculated, as shown in FIG. 11. Also, for tabulation (2) of item B × D and tabulation (3) of item B × C × D, the sample sizes and percentages for each of the age groups from 20s to 80s for item B (age group) in Company A's data are calculated.
[0036] Similarly, in device 10B, basic tabulation information provider 14B calculates the sample size and item-specific ratios, which are basic tabulation information, for the user data of device 10B and provides them to device 10A (step S10). For example, for tabulation (1) of item A × D and tabulation (2) of item B × D, the sample sizes and ratios for each of purchases and non-purchases for item D (purchase history) in Company B's data are calculated as shown in Figure 11. Also, for tabulation (3) of item B × C × D, the sample sizes and ratios for "item C (address) × item D (purchase history)" in Company B's data are calculated as follows: within Tokyo × purchases, outside Tokyo × purchases, within Tokyo × non-purchases, and outside Tokyo × non-purchases.
[0037] Then, the basic tally information acquisition unit 14A of the device 10A acquires the basic tally information of the devices 10A and 10B by receiving the basic tally information (sample size and item-specific proportions) of the device 10B provided (sent) by the basic tally information provision unit 14B of the device 10B (step S11).
[0038] Next, the estimation unit 15 estimates the sample size of the counting results based on the basic counting information of the devices 10A and 10B acquired by the basic counting information acquisition unit 14A and the basic values calculated by the basic value calculation unit 13, for example, as follows (step S12). For item A×D in the counting pattern (item A×D, item B×D, item B×C×D), as shown in FIG. 11, the sample size n iThe sample size of the aggregated results is estimated by applying formula (1): (Estimated value) = Sample size by gender in Company A's data × Purchase rate in Company B's data × Data matching rate (1). For example, for item A (male) × item D (purchased), multiplying the sample size of 8,306 for males in Company A's data by the purchase rate of 31% in Company B's data and the data matching rate of 40% yields an estimated sample size of 1,030. Similarly, for item A (male) × item D (non-purchased), item A (female) × item D (purchased), and item A (female) × item D (non-purchased), the estimated sample sizes of the aggregated results are 2,292, 522, and 1,162, respectively. The median of the four estimated values of item A × item D, 1,030, 2,292, 522, and 1,162, is obtained as 1,096, midway between the second largest value (1,030) and the third largest value (1,162). Similarly, for items BxD and BxCxD in the tabulation pattern (item AxD, item BxD, item BxCxD), 860 and 541 are obtained as the medians of the sample size estimates of the tabulation results, respectively. Note that in this embodiment, an example has been shown in which the "median" of the sample size estimates of the tabulation results is used, but the average value, minimum value, etc. may also be used. Also, in this embodiment, as mentioned above, the "data bias" of the basic values calculated in step S8 of FIG. 2 was 0%, but if there is data bias for a major item (item D in this case), the above sample size n i It is desirable to fine-tune (correct) the "purchase rate of Company B's data" in formula (1) that calculates the (estimated value) so that it approaches the purchase rate in the cross-checked data by the amount of "data bias."
[0039] 2 , in the next step S13, the evaluation unit 16 quantitatively evaluates the influence of noise on each combination included in the aggregation pattern based on the estimated sample size of the aggregation result (here, the median of the estimated sample size values) as follows: In this embodiment, the variance of the amount of change in the sample size n is used as an evaluation index for the influence of noise. An example using the following will be explained. The variance of noise z is The value varies depending on the probability distribution used to generate the noise. When the variance of the noise z is substituted into the above equation (2), the following is obtained. Here, the allocation of the privacy budget for the summaries (1) to (3) is defined as ε 1 , ε 2 , ε 3 By using the above formula (4), the following evaluation indices for the noise influence are obtained for the summaries (1) to (3) as shown in the table on the left side of FIG. 12: It is not essential to use the variance of the change in the sample size n as the evaluation index, and different indices for measuring the accuracy of data (for example, sampling error rate, statistical power, etc.) may be used depending on the purpose of aggregation.
[0040] In the next step S14, the allocation determination unit 17 determines the allocation of the privacy budget to each combination based on the evaluation result for each combination obtained by the evaluation unit 16, as follows. Here, for example, in order to equalize the influence of noise, the allocation determination unit 17 weights the remaining privacy budget ε by a function inversely proportional to the magnitude of the influence of noise. total Specifically, as shown on the right side of Figure 12, The remaining privacy budget ε total "0.9" is the budget allocation ε for each aggregation 1 , ε 2 , ε 3 By allocating to ε 1 = 0.209, ε 2 = 0.267, ε 3 = 0.424. In this case, the effect of noise z on the summaries (1) to (3) is 3.801 x 10 -5 It becomes uniform.
[0041] According to the first embodiment described above, the influence of noise on each combination included in the aggregation pattern can be quantitatively evaluated, and then the allocation of the privacy budget to each combination can be determined based on the evaluation results for each combination obtained in the evaluation, thereby optimizing the allocation of the privacy budget. Furthermore, by performing a trial aggregation in advance using the actually input user data of both companies A and B, basic values such as the data matching rate can be appropriately calculated based on the actual input data and used in the aggregation process, thereby contributing to the optimization of the aggregation process.
[0042] Second Embodiment Next, as a second embodiment, an embodiment will be described in which a known value or an estimated value is used as the base value without pre-execution as in the first embodiment. Because pre-execution is not performed, as shown in FIG. 13 , the information system 1 does not require functional units corresponding to the base value calculation unit 13 and the pre-execution data generation unit 12 in FIG. 1 . Therefore, the device 10A includes a data input unit 11, a basic tally information acquisition unit 14A, an estimation unit 15, an evaluation unit 16, and an allocation determination unit 17, while the device 10B includes a data input unit 11 and a basic tally information provision unit 14B. Since each functional unit was described in the first embodiment, a repeated description will be omitted here.
[0043] As shown in Fig. 14, the process in the second embodiment does not include the generation of pre-execution data (steps S3 and S4 in Fig. 2) and the process related to calculation of the base value (steps S5 to S8 in Fig. 2). The processes in steps S1, S2, and S9 to S11 shown in Fig. 14 are the same as those in the first embodiment, and therefore will not be described again.
[0044] In step S12, the estimation unit 15 estimates the sample size of the tabulation results using the same method as in the first embodiment, based on the "basic tabulation information" of the devices 10A and 10B acquired by the basic tabulation information acquisition unit 14A in step S11 and the "basic values" using known or estimated values. Here, for each item in the tabulation pattern (item A × D, item B × D, item B × C × D), 1096, 860, and 541 are obtained as the sample sizes of the tabulation results (here, the medians of the sample size estimates), respectively. Note that although this embodiment illustrates an example in which the "median" of the sample size estimates is used, the average, minimum, etc. may also be used.
[0045] In the next step S13, the evaluation unit 16 quantitatively evaluates the influence of noise on each combination included in the tally pattern based on the estimated sample size of the tally result, using the same method as in the first embodiment. As an evaluation index of the influence of noise, the variance of the amount of change in the sample size n is used. By using the above formula (5), the following can be used as an evaluation index for the influence of noise for the summaries (1) to (3), as shown in the table on the left side of FIG. In addition to the above, the evaluation index may be an index that measures the accuracy of different data (for example, sampling error rate, power of test, etc.) depending on the purpose of aggregation.
[0046] In the next step S14, the allocation determination unit 17 determines the allocation of the privacy budget to each combination as follows, based on the evaluation results for each combination obtained by the evaluation unit 16. Here, as a first allocation example, an example will be described in which, as in the first embodiment, each combination is weighted by a function that is in an inverse proportional relationship depending on the magnitude of the influence of noise for each combination, and the allocation of the privacy budget to each combination is determined. As a second allocation example, an example will be described in which the allocation of the privacy budget to each combination is determined so as to satisfy the level of accuracy required to suppress the influence of noise for each combination.
[0047] In the first allocation example, in order to equalize the influence of noise, the allocation determination unit 17 weights the remaining privacy budget ε by a function that is inversely proportional to the magnitude of the influence of noise. total In the second embodiment, since the "generation of pre-execution data (steps S3 and S4 in FIG. 2)" which consumes the privacy budget is not executed, the remaining privacy budget ε total is "1.0", unlike the first embodiment. Specifically, as shown on the right side of FIG. The remaining privacy budget ε total "1.0" is the total budget allocation ε for each of (1) to (3). 1 , ε 2 , ε 3 By allocating to ε 1 = 0.233, ε 2 = 0.296, ε 3 = 0.471. In this case, the influence of noise z on the summaries (1) to (3) is 3.079 x 10 -5 It becomes uniform.
[0048] In the second allocation example, the allocation determination unit 17 allocates the remaining privacy budget ε to each combination so as to satisfy the level of accuracy required to suppress the influence of noise for each combination. total The allocation of "1.0" is determined. In this case, if the required level of accuracy cannot be achieved within the privacy budget, it is possible to not perform certain aggregations among aggregations (1) to (3), or to relax the required level of accuracy. One example of a method for quantitatively determining the required level of accuracy is to set the required level of accuracy as the level at which the items to be detected (for example, purchase rates by attribute (gender or age group), differences in the number of purchasers by attribute (gender or age group)) are expected to be detectable from the aggregation results. For example, on the right side of Figure 16, the influence of noise for each combination is plotted against the required level of accuracy (for example, 5.000 x 10 -5 ) (here, in accordance with the level (i.e., to be the same as the level)). In this case, the influence of noise for each combination is determined based on the required level of accuracy (for example, 5,000 × 10 -5 ) and the remaining privacy budget is allocated to the budget allocation ε 1 , ε 2 , ε 3 and ε 1 = 0.182, ε 2 = 0.233, ε 3 = 0.370. Therefore, the total privacy budget consumed is 0.785, and the remaining privacy budget ε total Compared to "1.0", there will be a surplus in the privacy budget.
[0049] Note that the method of allocating the privacy budget is not limited to the first and second allocation examples described above, and various allocation methods other than the first and second allocation examples described above may be adopted depending on the purpose of the aggregation.
[0050] Next, the effect of using the present invention will be described in comparison with the effect of not using the present invention, with reference to Fig. 17. As shown on the left side of Fig. 17, when the present invention is not used, the tabulation results are not known, so the privacy budget ε total"1.0" is allocated evenly to aggregations (1) to (3). In this case, the influence of noise is 1.498 x 10 for each of aggregations (1) to (3). -5 , 2.434×10 -5 , 6.150×10 -5 This causes a problem in that the difference in the influence of noise between the summaries (1) to (3) becomes large.
[0051] On the other hand, in the first allocation example when the present invention is used, as shown in the upper right of FIG. 17, the budget allocations for the summaries (1) to (3) are respectively ε 1 = 0.233, ε 2 = 0.296, ε 3 = 0.471, and the effect of noise z on aggregation (1) to (3) is 3.079 × 10 -5 In this way, in the first distribution example, the privacy budget ε total Although all "1.0" is consumed, the influence of noise z is uniform in each calculation and is at a relatively low level of 3.079 x 10 -5 This has the effect of keeping the noise level to a minimum.
[0052] In addition, in the second allocation example, as shown in the lower right of FIG. 17, the budget allocations for the summaries (1) to (3) are ε 1 = 0.182, ε 2 = 0.233, ε 3 = 0.370, and the total privacy budget consumed is kept to 0.785. The influence of noise z in aggregation (1) to (3) is 5.000 × 10 -5 The required level of accuracy (5.000 x 10 -5 ) is satisfied. In this way, the second allocation example has the effect of preventing allocation of an excessive privacy budget relative to the required level, and suppressing budget consumption to the minimum necessary.
[0053] Third Embodiment Next, as a third embodiment, an embodiment in which a counting method that matches a predetermined purpose (for example, a counting method that can minimize the influence of noise) is selected will be described. As shown in Fig. 18, the configuration of the information system 1 of the third embodiment is a configuration in which a counting method selection unit 18 is added to the device 10A in the configuration of the second embodiment (Fig. 13) described above. Note that the configuration of the information system 1 of the third embodiment is not limited to the configuration of Fig. 18, and may also be a configuration in which a counting method selection unit 18 is added to the device 10A in the configuration of the first embodiment (Fig. 1) described above.
[0054] The tallying method selection unit 18 is a functional unit that executes the following operations according to each of a plurality of tallying methods that can be used when performing tallying on a tally pattern: acquisition of basic tally information by the basic tally information acquisition unit 14A; estimation of the sample size of the tally results by the estimation unit 15; evaluation of the impact of noise for each combination by the evaluation unit 16; and determination of the allocation of the privacy budget to each combination by the allocation determination unit 17; compares the results of all the tallying methods; and selects a tallying method that matches a predetermined purpose. Below, as an example, a description will be given of an example in which the tallying method selection unit 18 selects a tallying method that can minimize the impact of noise within an equivalent privacy budget.
[0055] Next, the processing of the third embodiment will be described with reference to the flow diagram of Fig. 19. After data input in steps S1 and S2, the tallying method selection unit 18 identifies possible tallying methods (step S1X). Here, it is assumed that the tallying methods identified are tallying method (a) in which each tally is performed independently, and tallying method (b) in which multiple tallying methods are performed together.
[0056] The counting method selection unit 18 executes the subsequent steps S9 to S14 for each of the counting methods (a) and (b). For example, the counting method selection unit 18 first executes the basic counting information acquisition unit 14A and the basic counting information provision unit 14B to perform basic counting according to counting method (a) (steps S9 and S10), and then causes the basic counting information acquisition unit 14A to acquire basic counting information for the devices 10A and 10B (step S11). The unit then executes the estimation unit 15 to estimate the sample size of the counting results (step S12), the evaluation unit 16 to evaluate the impact of noise on each combination (step S13), and the allocation determination unit 17 to determine the allocation of the privacy budget to each combination (step S14). In counting method (a), as shown in FIG. 20, counting steps (1) to (3) are executed independently, and noise is added to each counting result. In the aggregation method (a), the privacy budgets of each aggregation are combined, so it is not possible to allocate a large budget to each aggregation, which increases the strength of the added noise.
[0057] Next, the counting method selection unit 18 performs basic counting using the basic count information acquisition unit 14A and the basic count information provision unit 14B (steps S9 and S10) in accordance with counting method (b), as described above, and causes the basic count information acquisition unit 14A to acquire basic count information for the devices 10A and 10B (step S11). The estimation unit 15 then estimates the sample size of the counted results (step S12), the evaluation unit 16 evaluates the impact of noise on each combination (step S13), and the allocation determination unit 17 determines the allocation of the privacy budget to each combination (step S14). In counting method (b), as shown in FIG. 21, multiple counts are performed simultaneously, and statistical information for each count is generated based on the results. In counting method (b), the reduced number of counts allows a larger privacy budget to be allocated, thereby reducing the strength of the added noise. On the other hand, because the statistical information for each individual tally is generated by adding up multiple tally results with noise added (the rightmost cells in the left-hand table in Figure 21), noise is superimposed multiple times, equal to the number of cells being summed. For example, in tally (1), for item A (male) x item D (purchase), the 14 rightmost cells (1), (3), ... (27) in the left-hand table are summed to generate the statistical information for the tally, so noise is superimposed 14 times. Similarly, in tally (2), noise is superimposed four times, and in tally (3), noise is superimposed twice.
[0058] After the processing of steps S9 to S14 is completed for all the identified counting methods (counting methods (a) and (b)), the counting method selection unit 18 compares the evaluation of the noise impact of each counting method and selects the optimal counting method (step S15). Figure 22 shows an example of the results of quantitatively calculating the noise impact of counting methods (a) and (b). Here, the privacy budget is set to 1.0 and allocated for both counting methods (a) and (b). As shown at the right end of the left table in Figure 22, for counting method (a), the noise impact is 3.079 x 10 for each count. -5 On the other hand, as shown in the right-hand table of FIG. 22, for the counting method (b), the influence of noise is 2.331×10 for counting (1) to (3), respectively. -5 , 1.082×10 -5 , 1.367×10-5 As described above, compared to aggregation method (a), aggregation method (b) is less affected by noise in all of aggregations (1) to (3), so aggregation method (b) is selected.
[0059] In addition to the above-mentioned selection method based on the degree of noise influence, other methods may be used, such as a selection method based on overall balance, or a method of comparing the degree of noise influence for each aggregation method by focusing on specific aggregations with high priority. Examples of the former method (a selection method based on overall balance) include a method of avoiding aggregations that include aggregations with an extremely large degree of noise influence, and a method of evaluating each aggregation method using the average value of the degree of noise influence for each aggregation.
[0060] According to the third embodiment described above, it is possible to select an appropriate aggregation method that matches the purpose (for example, an aggregation method that can minimize the influence of noise, an aggregation method that has the best overall balance, an aggregation method that minimizes the influence of noise in specific aggregations with high priority, etc.) from the identified aggregation methods.
[0061] In the third embodiment, a configuration in which a counting method selection unit 18 is added to the device 10A in the configuration of the second embodiment (Figure 13) has been described. However, even if a configuration in which a counting method selection unit 18 is added to the device 10A in the configuration of the first embodiment (Figure 1) has been described, the same effects as those of the third embodiment can be achieved.
[0062] Furthermore, with regard to how the privacy budget is allocated, various allocation methods such as the first and second allocation examples have been explained in the second embodiment described above, but these various allocation methods can be applied not only to the second embodiment but also to the first and third embodiments.
[0063] (Variation 1) In the first embodiment, as shown in FIG. 1 , Company A's device 10A includes functional units such as a basic value calculation unit 13, an estimation unit 15, an evaluation unit 16, and an allocation determination unit 17, while Company B's device 10B does not include these functional units. However, the present invention is not limited to this configuration. As Variation 1, both companies' devices 10A and 10B may have similar configurations that include the functional units described above. Specifically, devices 10A and 10B may be configured as shown in FIG. 23 , and the basic value calculation, estimation, evaluation, and allocation determination processes can be performed by either device. For example, in a case where device 10A of devices 10A and 10B performs the basic value calculation, estimation, evaluation, and allocation determination processes, device 10A calculates basic aggregate information from input data to device 10B and provides the obtained basic aggregate information of Company B to device 10A's basic aggregate information acquisition unit 14A. In the above case, the base value calculation unit 13, the estimation unit 15, the evaluation unit 16, and the allocation determination unit 17 of the device 10B do not execute processing (shown by dashed lines in FIG. 23). Even in this modified example 1, the same effects as those of the first embodiment can be obtained.
[0064] (Variation 2) In the second embodiment, as shown in FIG. 13 , Company A's device 10A includes functional units such as the estimation unit 15, the evaluation unit 16, and the allocation determination unit 17, while Company B's device 10B does not include these functional units. However, this configuration is not limited to this. As Variation 2, both companies' devices 10A and 10B may have similar configurations that include the above functional units. Specifically, devices 10A and 10B may be configured as shown in FIG. 24 , and the estimation, evaluation, and allocation determination processes can be performed by either device. For example, in a case where device 10A of devices 10A and 10B performs the estimation, evaluation, and allocation determination processes, device 10A calculates basic tally information from input data to device 10B and provides the obtained basic tally information of Company B to device 10A's basic tally information acquisition unit 14A. In the above case, the estimation unit 15, evaluation unit 16, and allocation determination unit 17 of the device 10B do not execute the processes (shown by dashed lines in FIG. 24). Even in such a modified example 2, the same effects as those of the second embodiment can be obtained.
[0065] The gist of the present disclosure lies in the following [1] to
[10] : [1] An information processing device comprising: a basic tally information acquisition unit that acquires basic tally information of a user's own device and a counterpart device, each of which stores user data including a user ID and attribute information related to the user; an estimation unit that estimates a sample size of a tally result based on the basic tally information of the user's own device and the counterpart device acquired by the basic tally information acquisition unit and a basic value used for estimating the sample size of the tally result; an evaluation unit that quantitatively evaluates the influence of noise on each combination included in a tally pattern, the tally pattern including a plurality of combinations of data items included in the attribute information, based on the sample size of the tally result estimated by the estimation unit; and an allocation determination unit that determines, within a predetermined privacy budget range, an allocation of a privacy budget to each combination related to noise for tallying the tally pattern, based on an evaluation result of each combination obtained by the evaluation unit. [2] The information processing device according to [1], further comprising a base value calculation unit that calculates the base value from the user data of the own device and the encrypted user data of the other device, wherein the estimation unit uses the base value calculated by the base value calculation unit as the base value used to estimate the sample size of the aggregation result. [3] The information processing device according to [1], wherein the estimation unit uses a known base value or a base value previously determined by estimation as the base value used to estimate the sample size of the aggregation result. [4] The information processing device according to any one of [1] to [3], wherein the evaluation unit quantitatively evaluates the influence of noise for each combination included in the aggregation pattern using a predetermined index related to the data accuracy of the estimated aggregation result as an evaluation index for the influence of noise. [5] The information processing device according to any one of [1] to [4], wherein the allocation determination unit determines the allocation of the privacy budget to each combination based on weighting each combination using a function that depends on the magnitude of the influence of noise for each combination. [6] The information processing device according to any one of [1] to [4], wherein the allocation determination unit determines the allocation of the privacy budget to each combination based on a level of accuracy required to suppress the influence of noise for each combination.[7] The information processing device according to any one of [1] to [4], wherein the allocation determination unit determines the allocation of the privacy budget to each combination based on weighting for each combination using a function that depends on the magnitude of the influence of noise for each combination, under the condition that the combination satisfies a level of accuracy required for suppressing the influence of noise. [8] The information processing device according to any one of [1] to [7], further comprising: an aggregation method selection unit that causes the basic aggregation information acquisition unit to acquire the basic aggregation information, the estimation unit to estimate a sample size of the aggregation results, the evaluation unit to evaluate the influence of noise for each combination, and the allocation determination unit to determine the allocation of the privacy budget to each combination, in accordance with each of a plurality of aggregation methods that can be taken when performing aggregation on the aggregation pattern, and [9] An information system including: a first device including: a basic tally information acquisition unit that acquires basic tally information of a first device and a second device, each of which holds user data including a user ID and attribute information related to the user; an estimation unit that estimates a sample size of the tally results based on the basic tally information of the first device and the second device acquired by the basic tally information acquisition unit and a basic value used for estimating the sample size of the tally results; an evaluation unit that quantitatively evaluates the impact of noise for each combination included in an tally pattern, the combination including multiple combinations of data items included in the attribute information, based on the sample size of the tally results estimated by the estimation unit; and an allocation determination unit that determines the allocation of a privacy budget to each combination related to noise for aggregation targeting the tally pattern, based on the evaluation result for each combination obtained by the evaluation unit; and a second device that provides the basic tally information of the user data of the second device to the basic tally information acquisition unit.
[10] An information processing method comprising: a step in which an information processing device acquires basic aggregate information of its own device and a counterpart device, each of which holds user data including a user ID and attribute information related to the user; a step in which the information processing device estimates a sample size of the aggregated results based on the acquired basic aggregate information of its own device and the counterpart device and a basic value used for estimating the sample size of the aggregated results; a step in which the information processing device quantitatively evaluates the impact of noise for each combination included in an aggregated pattern, which includes multiple combinations of data items included in the attribute information, based on the estimated sample size of the aggregated results; and a step in which the information processing device determines, within a predetermined privacy budget range, an allocation of a privacy budget to each combination related to noise for aggregation targeting the aggregated pattern, based on the evaluation results for each combination obtained by the evaluation.
[0066] (Explanation of terms, explanation of hardware configuration (Figure 25), etc.) Note that the block diagrams used in the explanation of the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using one device that is physically or logically coupled, or may be realized using two or more devices that are physically or logically separated and connected directly or indirectly (for example, using wires, wirelessly, etc.) and these multiple devices. A functional block may be realized by combining software with the one device or the multiple devices.
[0067] Functions include, but are not limited to, judgment, determination, assessment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.
[0068] For example, an information processing device according to an embodiment of the present disclosure may function as a computer that executes the processes of the present disclosure. Fig. 25 is a diagram illustrating an example of a hardware configuration of an information processing device 10A according to an embodiment of the present disclosure. The information processing device 10A described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, etc., and the information processing device 10B may also be configured in a similar manner.
[0069] In the following description, the term "apparatus" can be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the information processing device 10A may be configured to include one or more of the apparatuses shown in the figure, or may be configured to exclude some of the apparatuses.
[0070] Each function in the information processing device 10A is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.
[0071] The processor 1001 controls the entire computer by running, for example, an operating system, and may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc.
[0072] The processor 1001 also reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. While the various processes have been described as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may be transmitted from a network via a telecommunications line.
[0073] The memory 1002 is a computer-readable recording medium and may be configured by, for example, at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for implementing a wireless communication method according to an embodiment of the present disclosure.
[0074] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.
[0075] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as, for example, a network device, a network controller, a network card, a communication module, etc. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD).
[0076] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. The input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).
[0077] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.
[0078] The information processing device 10A may also be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.
[0079] The notification of information is not limited to the aspects / embodiments described in the present disclosure and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI) and Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB) and System Information Block (SIB))), other signals, or a combination thereof. Furthermore, the RRC signaling may be referred to as an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.
[0080] Each aspect / embodiment described in the present disclosure may be implemented using any of the following standards: LTE (Long Term Evolution), LTE-Advanced (LTE-A), SUPER 3G, IMT-Advanced, 4G (4th generation mobile communication system), 5G (5th generation mobile communication system), 6th generation mobile communication system (6G), xth generation mobile communication system (xG) (xG (x is, for example, an integer or a decimal number)), FRA (Future Radio Access), NR (new Radio), New radio access (NX), Future generation radio access (FX), W-CDMA (registered trademark), GSM (registered trademark), CDMA2000, UMB (Ultra Mobile Broadband), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.17 (WiMAX (registered trademark)), IEEE 802.19 (WiMAX (registered trademark)), IEEE 802.20 (WiMAX (registered trademark)), IEEE 802.21 (Wi-Fi (registered trademark)), IEEE 802.22 (WiMAX (registered trademark)), IEEE 802.23 (WiMAX (registered trademark)), IEEE 802.24 (WiMAX (registered trademark)), IEEE 802.25 (WiMAX (registered trademark)), IEEE 802.26 (WiMAX (registered trademark)), IEEE 802.27 (WiMAX (registered trademark)), IEEE 802.28 (WiMAX (registered trademark)), IEEE 802.29 (WiMAX (registered trademark)), IEEE 802.30 (WiMAX (registered trademark)), IEEE 802.31 (Wi-Fi (registered trademark)), IEEE 802.32 (WiMAX (registered trademark)), IEEE 802.33 (WiMAX (registered trademark)), IEEE 802.34 ( The present invention may be applied to at least one of systems using 802.20, UWB (Ultra-Wide Band), Bluetooth (registered trademark), or other suitable systems, and next-generation systems that are extended, modified, created, or defined based on these systems. The present invention may also be applied to a combination of multiple systems (e.g., a combination of LTE and / or LTE-A with 5G).
[0081] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.
[0082] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be transmitted to another device.
[0083] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).
[0084] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).
[0085] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.
[0086] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.
[0087] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.
[0088] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.
[0089] Note that terms described in this disclosure and terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings. For example, at least one of a channel and a symbol may be a signal (signaling). Furthermore, a signal may be a message. Furthermore, a component carrier (CC) may be called a carrier frequency, a cell, a frequency carrier, etc.
[0090] As used in this disclosure, the terms "system" and "network" are used interchangeably.
[0091] Furthermore, the information, parameters, etc. described in the present disclosure may be expressed using absolute values, may be expressed using relative values from a predetermined value, or may be expressed using other corresponding information. For example, a radio resource may be indicated by an index.
[0092] The names used for the above-described parameters are not intended to be limiting in any way. Furthermore, the mathematical expressions using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (e.g., PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not intended to be limiting in any way.
[0093] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.
[0094] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."
[0095] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.
[0096] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.
[0097] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.
[0098] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."
[0099] 1...information system, 10A, 10B...information processing device, 11...data input unit, 12...pre-execution data generation unit, 12A...data processing unit, 12B...anonymization processing unit, 12C...ID irreversible conversion unit, 12D...encryption unit, 12E...data transmission / reception unit, 13...basic value calculation unit, 13A...data matching unit, 13B...counting processing unit, 13C...confidentiality processing unit, 13D...calculation unit, 14A...basic count information acquisition unit, 14B...basic count information provision unit, 15...estimation unit, 16...evaluation unit, 17...allocation determination unit, 18...counting method selection unit, 1001...processor, 1002...memory, 1003...storage, 1004...communication device, 1005...input device, 1006...output device, 1007...bus.
Claims
1. An information processing device comprising: a basic tally information acquisition unit that acquires basic tally information of a user's own device and a counterpart device, each of which holds user data including a user ID and attribute information related to the user; an estimation unit that estimates a sample size of an tally result based on the basic tally information of the user's own device and the counterpart device acquired by the basic tally information acquisition unit and a basic value used for estimating the sample size of the tally result; an evaluation unit that quantitatively evaluates the impact of noise for each combination included in an tally pattern, which includes multiple combinations of data items included in the attribute information, based on the sample size of the tally result estimated by the estimation unit; and an allocation determination unit that determines the allocation of a privacy budget to each combination related to noise for aggregation targeting the tally pattern, within a predetermined privacy budget, based on the evaluation results of each combination obtained by the evaluation unit.
2. An information processing device as described in claim 1, further comprising a base value calculation unit that calculates the base value from the user data of its own device and the encrypted user data of the other device, wherein the estimation unit uses the base value calculated by the base value calculation unit as the base value used to estimate the sample size of the aggregation result.
3. The information processing device according to claim 1, wherein the estimation unit uses a known basic value or a basic value obtained in advance by estimation as a basic value used to estimate the sample size of the tabulation results.
4. The information processing device of claim 1, wherein the evaluation unit quantitatively evaluates the influence of noise for each combination included in the aggregation pattern by using a predetermined index related to the data accuracy of the estimated aggregation result as an evaluation index for the influence of noise.
5. The information processing device according to claim 1, wherein the allocation determination unit determines the allocation of the privacy budget to each combination based on weighting each combination by a function that depends on the magnitude of the effect of noise on each combination.
6. The information processing device according to claim 1, wherein the allocation determination unit determines the allocation of the privacy budget to each combination based on a level of accuracy required to suppress the effect of noise for each combination.
7. The information processing device of claim 1, wherein the allocation determination unit determines the allocation of the privacy budget to each combination based on weighting each combination by a function that depends on the magnitude of the influence of noise for each combination, under the condition that the level of accuracy required for suppressing the influence of noise for each combination is met.
8. The information processing device of claim 1, further comprising: an aggregation method selection unit that executes, according to each of a plurality of aggregation methods that can be taken when performing aggregation on the aggregation pattern, the acquisition of basic aggregation information by the basic aggregation information acquisition unit, the estimation unit estimating the sample size of the aggregation results, the evaluation unit evaluating the impact of noise for each combination, and the allocation determination unit determining the allocation of the privacy budget to each combination, and selects an aggregation method that matches a specified purpose based on a comparison result of all the aggregation methods.
9. An information system comprising: a first device comprising: a basic tally information acquisition unit that acquires basic tally information of a first device and a second device, each of which holds user data including a user ID and attribute information related to the user; an estimation unit that estimates a sample size of an aggregation result based on the basic tally information of the first device and the second device acquired by the basic tally information acquisition unit and a basic value used for estimating the sample size of the aggregation result; an evaluation unit that quantitatively evaluates the impact of noise for each combination included in an aggregation pattern, the combination including a plurality of combinations of data items included in the attribute information, based on the sample size of the aggregation result estimated by the estimation unit; and an allocation determination unit that determines an allocation of a privacy budget to each combination related to noise for aggregation targeting the aggregation pattern, within a predetermined privacy budget range, based on an evaluation result for each combination obtained by the evaluation unit; and a second device that provides basic tally information of the user data of the second device to the basic tally information acquisition unit.
10. An information processing method comprising: a step in which an information processing device acquires basic aggregate information of its own device and a counterpart device, each of which holds user data including a user ID and attribute information related to the user; a step in which the information processing device estimates a sample size of the aggregated results based on the acquired basic aggregate information of its own device and the counterpart device and a basic value used for estimating the sample size of the aggregated results; a step in which the information processing device quantitatively evaluates the influence of noise for each combination included in an aggregated pattern including multiple combinations of data items included in the attribute information, based on the estimated sample size of the aggregated results; and a step in which the information processing device determines an allocation of a privacy budget to each combination related to noise for aggregation targeting the aggregated pattern, within a predetermined privacy budget range, based on the evaluation results for each combination obtained by the evaluation.
Citation Information
Patent Citations
Method and apparatus for collecting key-value pair data
US20210081412A1
Data analytics privacy platform with quantified re-identification risk
WO2022061162A1
