WO2021229973A1 - Dispositif de traitement d'informations, programme, et procédé de traitement d'informations - Google Patents

Dispositif de traitement d'informations, programme, et procédé de traitement d'informations Download PDF

Info

Publication number
WO2021229973A1
WO2021229973A1 PCT/JP2021/015183 JP2021015183W WO2021229973A1 WO 2021229973 A1 WO2021229973 A1 WO 2021229973A1 JP 2021015183 W JP2021015183 W JP 2021015183W WO 2021229973 A1 WO2021229973 A1 WO 2021229973A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
encrypted
structure data
function
molecular structure
Prior art date
Application number
PCT/JP2021/015183
Other languages
English (en)
Japanese (ja)
Inventor
哲哉 加川
Original Assignee
コニカミノルタ株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by コニカミノルタ株式会社 filed Critical コニカミノルタ株式会社
Priority to JP2022522566A priority Critical patent/JPWO2021229973A1/ja
Publication of WO2021229973A1 publication Critical patent/WO2021229973A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09C—CIPHERING OR DECIPHERING APPARATUS FOR CRYPTOGRAPHIC OR OTHER PURPOSES INVOLVING THE NEED FOR SECRECY
    • G09C1/00—Apparatus or methods whereby a given sequence of signs, e.g. an intelligible text, is transformed into an unintelligible sequence of signs by transposing the signs or groups of signs or by replacing them by others according to a predetermined system

Definitions

  • the present invention relates to an information processing device, a program, and an information processing method.
  • the deductive prediction model is a prediction model that predicts the function of a compound from known principles and regularities of the compound.
  • the learning model is, for example, a prediction model that represents the correlation between the explanatory variable and the objective variable, which is obtained as a result of inductive learning using the descriptor related to the structure of the compound as the explanatory variable and the function exhibited by the compound as the objective variable.
  • MI material informatics
  • Information on the target compound whose function is predicted using such a prediction model and information on the compound used to generate a learning model can be obtained from, for example, a public database.
  • the compounds that can obtain the necessary information from the public database are limited, and the number of compound candidates for functional prediction can be increased, or the number of compounds used to generate learning models can be increased to further improve the prediction accuracy. In order to do so, it is necessary to obtain compound information from a private database as well.
  • acquiring information on the structure of a compound from a private database may lead to leakage of confidential information related to the structure of the compound.
  • An object of the present invention is to provide an information processing device, a program, and an information processing method capable of enhancing the safety of confidential information related to the structure of a compound.
  • the invention of the information processing apparatus is An information providing unit that provides encryption algorithm information for executing encryption according to a predetermined encryption algorithm to a first external device, and an information providing unit.
  • a first data acquisition unit that acquires encrypted structure data of a function prediction target encrypted according to the encryption algorithm from the first external device.
  • a prediction unit that predicts the function of the compound corresponding to the encrypted structure data to be predicted based on a predetermined prediction model, and a prediction unit. Equipped with The prediction model represents the correlation between the encrypted structure data obtained by encrypting the structural data related to the structure of the compound according to the encryption algorithm and the functional data related to the function of the compound.
  • the invention according to claim 2 is the information processing apparatus according to claim 1.
  • An encryption unit that encrypts structural data related to the structure of a compound according to the encryption algorithm to generate encrypted structure data
  • a learning model generation unit that generates a learning model as the prediction model based on the encrypted structure data and the functional data related to the function of the compound. To prepare for.
  • the invention according to claim 3 is the information processing apparatus according to claim 2.
  • the first data acquisition unit functions from the first external device to the encrypted structure data of the learning target encrypted according to the encryption algorithm and the function of the compound corresponding to the encrypted structure data of the learning target. Acquire the functional data of the relevant learning target and
  • the learning model generation unit generates the learning model by using at least the encrypted structure data of the learning target and the functional data of the learning target acquired by the first data acquisition unit.
  • the invention of the information processing apparatus is An encryption unit that encrypts structural data related to the structure of a compound according to a predetermined encryption algorithm to generate encrypted structure data, A learning model generation unit that generates a learning model that represents a correlation between the encrypted structure data and the functional data based on the encrypted structure data and the functional data related to the function of the compound.
  • An information providing unit that provides encryption algorithm information for executing encryption according to the encryption algorithm to the first external device, and an information providing unit. The first to acquire the encrypted structure data of the learning target encrypted according to the encryption algorithm and the functional data of the learning target related to the function of the compound corresponding to the encrypted structure data from the first external device.
  • Data acquisition department and Equipped with The learning model generation unit generates the learning model by using at least the encrypted structure data of the learning target and the functional data of the learning target acquired by the first data acquisition unit.
  • the invention according to claim 5 is the information processing apparatus according to claim 4.
  • the first data acquisition unit acquires the encryption structure data of the function prediction target encrypted according to the encryption algorithm from the first external device.
  • the information processing apparatus includes a prediction unit that predicts the function of the compound corresponding to the encrypted structure data of the function prediction target acquired by the first data acquisition unit based on the learning model.
  • the invention according to claim 6 is the information processing apparatus according to any one of claims 2, 3 and 5. It is provided with a structural data generation unit that generates the structural data.
  • the encryption unit encrypts the structural data generated by the structural data generation unit to generate the encrypted structure data to be functionally predicted.
  • the prediction unit predicts the function of the compound corresponding to the encrypted structure data of the function prediction target generated by the encryption unit based on the learning model.
  • the invention according to claim 7 is the information processing apparatus according to any one of claims 2, 3 and 5. It is provided with a second data acquisition unit that acquires structural data related to the structure of the compound from a second external device that discloses the structure of the compound.
  • the encryption unit encrypts the structural data acquired by the second data acquisition unit to generate the encrypted structure data to be functionally predicted.
  • the prediction unit predicts the function of the compound corresponding to the encrypted structure data of the function prediction target generated by the encryption unit based on the learning model.
  • the invention according to claim 8 is the information processing apparatus according to any one of claims 2 to 7. It is provided with a third data acquisition unit that acquires the structural data and the functional data from an external predetermined database.
  • the encryption unit generates the encrypted structure data based on the structure data acquired by the third data acquisition unit.
  • the learning model generation unit generates the learning model using at least the encrypted structure data and the functional data acquired by the third data acquisition unit.
  • the invention according to claim 9 is the information processing apparatus according to any one of claims 1 to 8.
  • the encryption algorithm cannot be back-converted to the structural data before encryption.
  • the invention of the program according to claim 10 is The computer installed in the information processing device An information providing means for providing encryption algorithm information for executing encryption according to a predetermined encryption algorithm to a first external device, A data acquisition means for acquiring encryption structure data of a function prediction target encrypted according to the encryption algorithm from the first external device. A prediction means that predicts the function of a compound corresponding to the encrypted structure data to be predicted based on a predetermined prediction model. To function as The prediction model represents the correlation between the encrypted structure data obtained by encrypting the structural data related to the structure of the compound according to the encryption algorithm and the functional data related to the function of the compound.
  • the invention of the program according to claim 11 is The computer installed in the information processing device
  • An encryption means that generates encrypted structure data by encrypting structural data related to the structure of a compound according to a predetermined encryption algorithm.
  • a learning model generation means for generating a learning model representing the correlation between the encrypted structure data and the functional data based on the encrypted structure data and the functional data related to the function of the compound.
  • An information providing means for providing encryption algorithm information for executing encryption according to the encryption algorithm to a first external device, Data acquisition means for acquiring the encrypted structure data of the learning target encrypted according to the encryption algorithm and the functional data of the learning target related to the function of the compound corresponding to the encrypted structure data from the first external device.
  • the learning model generation means generates the learning model by using at least the encrypted structure data of the learning target and the functional data of the learning target acquired by the data acquisition means.
  • An encryption step that encrypts structural data related to the structure of a compound according to a predetermined encryption algorithm to generate encrypted structural data
  • a learning model generation step for generating a learning model representing the correlation between the encrypted structure data and the functional data based on the encrypted structure data and the functional data related to the function of the compound.
  • An information providing step for providing encryption algorithm information for executing encryption according to the encryption algorithm to a first external device
  • a data acquisition step of acquiring the encrypted structure data of the learning target encrypted according to the encryption algorithm and the functional data of the learning target related to the function of the compound corresponding to the encrypted structure data from the first external device.
  • the learning model is generated by using at least the encrypted structure data of the learning target and the functional data of the learning target acquired in the data acquisition step.
  • FIG. 1 is a diagram showing a schematic configuration of a compound information processing system 100.
  • the compound information processing system 100 includes an MI server 1 (information processing apparatus), a public database server 2 (hereinafter referred to as “public DB server 2”) (predetermined database), and a reagent database server 3 (hereinafter referred to as “reagent DB”). It includes a server 3) (second external device) and a client server 4 (first external device).
  • the MI server 1, the public DB server 2, the reagent DB server 3, and the client server 4 are connected to each other so as to be able to communicate with each other via the communication network N.
  • the communication network N is, for example, the Internet, but is not limited to this.
  • the MI server 1 is a device owned by a provider of an information providing service related to material informatics (MI), and performs various information processing related to MI. That is, the MI server 1 generates a learning model (also referred to as a "learned model") that predicts the function of the compound based on the information related to the compound by machine learning, and uses the learning model to develop materials by MI. Acquires or generates useful information and sends it to the client server 4. More specifically, the MI server 1 acquires the target value of the function of the compound from the client server 4, searches for the compound exhibiting the function of the target value, and obtains the information related to the structure of the specified compound to the client server. Send to 4.
  • MI material informatics
  • the MI server 1 uses a large number of combinations of encrypted molecular structure data (encrypted structure data) related to the structure of the compound and functional data related to the function of the compound to obtain encrypted molecular structure data and functional data.
  • a learning model representing the correlation of is generated by machine learning by a recursive approach.
  • the encrypted molecular structure data corresponds to the explanatory variables of machine learning
  • the functional data corresponds to the objective variable of machine learning.
  • the encrypted molecular structure data is data obtained by encrypting molecular structure data (structural data) related to the structure of a compound according to a predetermined encryption algorithm.
  • the number of datasets consisting of encrypted molecular structure data and functional data used to generate a learning model is, for example, tens of thousands or more.
  • One way to improve the prediction accuracy of the training model is to increase the number of this data set.
  • the encrypted molecular structure data and the functional data used for generating the learning model are also referred to as “encrypted molecular structure data to be learned” and “functional data to be learned”, respectively.
  • the learning model generated by MI server 1 is one of the prediction models for predicting the function of the compound.
  • the molecular structure data that is the source of the encrypted molecular structure data is not particularly limited as long as the composition of the molecule, that is, the elements constituting the molecule and the bonding mode thereof can be specified.
  • an encryption algorithm for generating encrypted structure data from molecular structure data for example, an algorithm that extracts and quantifies the characteristics of the molecular structure of a compound according to a predetermined rule can be used.
  • FIG. 2 is a diagram illustrating an example of an encryption algorithm for generating encrypted molecular structure data.
  • the structural formula shown in the upper part of the figure is converted into the code shown in the lower part of the figure according to its characteristics.
  • Each digit of the code is 0 or 1.
  • the conversion rule by the encryption algorithm of FIG. 2 can be, for example, as follows. That is, first, each atom constituting the molecule is numbered by the Morgan method. Next, atomic information is added by Daylight rule, and fragment information contained in the molecule is added. Then remove the duplicate fragments. Finally, the obtained fragment is assigned to a predetermined digit by a hash function. For example, when a specific fragment is contained in a molecule, the predetermined digit of the code is 1.
  • the encrypted molecular structure data generated by such an encryption algorithm is a kind of descriptor representing the characteristics of the molecular structure. That is, from the encrypted molecular structure data, the characteristics of the molecular structure can be multifacetedly specified from the position of the digit whose value is 1.
  • the hash function is a one-way function, the inverse conversion from the encrypted molecular structure data to the molecular structure data is impossible. That is, in this embodiment, an irreversible encryption algorithm is used.
  • the type of learning model generated by MI server 1 is not particularly limited as long as it represents the correlation between the encrypted molecular structure data and the functional data.
  • various known ones such as linear regression, principal component analysis, decision tree, random forest, support vector machine, and random forest can be used.
  • the MI server 1 applies the generated learning model to the encrypted molecular structure data related to the compound whose function is to be predicted (hereinafter, also referred to as "encrypted molecular structure data to be functionally predicted"). , Predict the function of the compound corresponding to the encrypted molecular structure data.
  • the MI server 1 performs function prediction for a large number of encrypted molecular structure data, and identifies the encrypted molecular structure data for which a prediction result matching the target value of the function received from the client server 4 is obtained. Then, the MI server 1 transmits the information related to the specified encrypted molecular structure data to the client server 4.
  • the public DB server 2 stores molecular structure data related to the molecular structure of a large number of compounds and functional data related to the functions exhibited by the compound.
  • the public DB server 2 provides these molecular structure data and functional data in response to a request from another device (MI server 1 in this embodiment).
  • the molecular structure data provided from the public DB server 2 to the MI server 1 is encrypted in the MI server 1 and converted into encrypted molecular structure data.
  • This encrypted molecular structure data can be used together with the functional data for machine learning for generating a learning model, and can also be used as the encrypted molecular structure data for which the function is predicted.
  • the reagent DB server 3 stores molecular structure data related to the molecular structures of a large number of compounds to be sold. It can also be said that the reagent DB server 3 provides a catalog of reagents (compounds) that can be purchased. The reagent DB server 3 provides molecular structure data in response to a request from another device (MI server 1 in this embodiment). In the present embodiment, it is assumed that the reagent DB server 3 does not provide functional data related to the function of the compound.
  • the molecular structure data provided from the reagent DB server 3 to the MI server 1 is encrypted in the MI server 1 and converted into encrypted molecular structure data. This encrypted molecular structure data is used as the encrypted molecular structure data for which the function is predicted.
  • the client server 4 is a device owned by a client who receives an information providing service related to MI by the MI server 1.
  • the client server 4 transmits data or the like that specifies a target value of the function of the compound desired by the client to the MI server 1, and receives information related to the structure of the compound exhibiting the function from the MI server 1. Further, the client server 4 transmits the encrypted molecular structure data of the function prediction target to the MI server 1 in order to receive the necessary information providing service related to MI, and the encrypted molecular structure data of the learning target and learning. Send the target function data.
  • the molecular structure data related to the structure of the compound stored in the client server 4 is regarded as confidential information.
  • the encrypted molecular structure data obtained by encrypting the molecular structure data in the client server 4 is transmitted to the MI server 1 to disclose the molecular structure data which is confidential information to the MI server 1. It is possible to receive the necessary information provision service. The mechanism for protecting the confidential information in the client server 4 in this way will be described in detail later.
  • FIG. 3 is a block diagram showing a main functional configuration of the MI server 1.
  • the MI server 1 includes a control unit 11, an operation unit 12, a display unit 13, a communication unit 14, and the like, and each of these units is connected by a bus 15.
  • the control unit 11 is a processor (computer) that collectively controls the operation of the MI server 1.
  • the control unit 11 has a CPU 111 (Central Processing Unit), a RAM 112 (Random Access Memory), and a storage unit 113.
  • the CPU 111 reads out various control programs 113c and setting data stored in the storage unit 113 and stores them in the RAM 112, and executes the program 113c to perform various arithmetic processes.
  • the RAM 112 provides a working memory space for the CPU 111 and stores temporary data.
  • the RAM 112 may include a non-volatile memory.
  • the storage unit 113 stores various data for performing information processing related to MI.
  • the storage unit 113 for example, an HDD (Hard Disk Drive) may be used, or a DRAM (Dynamic Random Access Memory) or the like may be used in combination.
  • the data stored in the storage unit 113 includes general data 113a, client-derived data 113b, encryption algorithm information D1, learning model data D2, and the like.
  • the general data 113a is data related to the structure and function of the compound, which is acquired without going through the client server 4, that is, the data acquired from the public DB server 2 or the reagent DB server 3, or in the MI server 1. It is the data etc. generated in. Specifically, the general data 113a includes the molecular structure data A1 to be learned, the encrypted molecular structure data A2 thereof, and the functional data A3. Further, the general data 113a includes the molecular structure data B1 to be functionally predicted and the encrypted molecular structure data B2 thereof. Of these, the molecular structure data A1 and the functional data A3 are acquired from the public DB server 2. Further, the molecular structure data B1 is acquired from the public DB server 2 or the reagent DB server 3. Further, as will be described later, the molecular structure data B1 may be generated in the MI server 1.
  • the client-derived data 113b is data acquired from the client server 4 among the data related to the structure and function of the compound.
  • the client-derived data 113b includes a labeled encrypted molecular structure data C2L to be functionally predicted, an encrypted molecular structure data C2 to be learned, and a functional data C3 to be learned.
  • FIG. 4 is a diagram showing a content example of the labeled encrypted molecular structure data C2L.
  • the labeled encrypted molecular structure data C2L is data in which a unique label (here, a natural number) is associated with each of the plurality of encrypted molecular structure data whose functions are to be predicted.
  • the encryption algorithm information D1 shown in FIG. 3 is information related to an encryption algorithm for generating encrypted molecular structure data from the molecular structure data.
  • encryption algorithm information D1 when executing a predetermined encryption program for generating encrypted molecular structure data, encryption according to a specific encryption algorithm can be performed.
  • the encryption algorithm information D1 may be the encryption program itself.
  • the encryption algorithm information D1 is used when the control unit 11 of the MI server 1 encrypts the molecular structure data A1 and B1 to generate the molecular structure data A1 and B2. Further, the encryption algorithm information D1 is transmitted to the client server 4 for the encryption process in the client server 4.
  • the learning model data D2 relates to a learning model generated by machine learning based on the encrypted molecular structure data A2 and / or the encrypted molecular structure data C2 to be learned and the functional data C3 to be learned. It is data.
  • the learning model represented by the learning model data D2 By applying the learning model represented by the learning model data D2 to the encrypted molecular structure data to be predicted, the function of the compound corresponding to the encrypted molecular structure data can be predicted.
  • generating the training model data D2 is also referred to as "generating a learning model”.
  • the encryption unit encrypts the molecular structure data A1 and B1 related to the structure of the compound according to the encryption algorithm indicated by the encryption algorithm information D1 to generate the encrypted molecular structure data A2 and B2.
  • the learning model generation unit performs machine learning based on the encrypted molecular structure data A2, the functional data A3, and / or the encrypted molecular structure data C2 to be learned, and the functional data C3 to obtain the learning model data D2. Generate.
  • the information providing unit provides the client server 4 with encryption algorithm information D1 for executing encryption according to the above encryption algorithm (transmitted by the communication unit 14).
  • the first data acquisition unit receives the labeled encrypted molecular structure data C2L of the function prediction target, the encrypted molecular structure data C2 of the learning target, and the functional data C3 of the learning target from the client server 4 via the communication unit 14. To get.
  • the second data acquisition unit acquires the molecular structure data B1 from the reagent DB server 3 via the communication unit 14.
  • the third data acquisition unit acquires the molecular structure data A1 and the functional data A3 from the public DB server 2 via the communication unit 14.
  • the prediction unit predicts the functions of the compounds corresponding to the encrypted molecular structure data A2 and B2 to be functionally predicted and the labeled encrypted molecular structure data C2L based on the learning model represented by the learning model data D2.
  • the structural data generation unit mechanically generates the molecular structure data B1 using a genetic algorithm or the like, and stores it in the storage unit 113.
  • the operation unit 12 is realized by an input device such as a keyboard and a mouse, a touch panel integrally provided with the display unit 13, and the like.
  • the operation unit 12 receives the operation input from these input devices and the touch panel, and outputs the operation signal corresponding to the operation input to the control unit 11.
  • the display unit 13 is realized by a liquid crystal display device, an organic EL display device, or the like, and displays various information under the control of the control unit 11.
  • the communication unit 14 transmits / receives data to / from the public DB server 2, the reagent DB server 3, and the client server 4 via the communication network N under the control of the control unit 11.
  • FIG. 5 is a block diagram showing a main functional configuration of the client server 4.
  • the client server 4 includes a control unit 41, an operation unit 42, a display unit 43, a communication unit, and the like, and each of these units is connected by a bus 45.
  • the control unit 41 is a processor that collectively controls the operation of the client server 4.
  • the control unit 41 has a CPU 411, a RAM 412, and a storage unit 413.
  • the CPU 411 reads out various control programs 413a and setting data stored in the storage unit 413 and stores them in the RAM 412, and executes the program 413a to perform various arithmetic processes.
  • the RAM 412 provides a working memory space for the CPU 411 and stores temporary data.
  • the RAM 412 may include a non-volatile memory.
  • the storage unit 413 stores molecular structure data C1, labeled encrypted molecular structure data C2L, encrypted molecular structure data C2, functional data C3, encryption algorithm information D1, and the like.
  • the storage unit 413 for example, an HDD may be used, or a DRAM or the like may be used in combination.
  • the encryption algorithm information D1 stored in the storage unit 413 is transmitted from the MI server 1 and is the same as the encryption algorithm information D1 stored in the storage unit 113 of the MI server 1.
  • the control unit 41 of the client server 4 can encrypt the molecular structure data C1 according to the same encryption algorithm as the MI server 1 and generate the encrypted molecular structure data C2.
  • the molecular structure data C1 is data relating to the molecular structure of the compound possessed by the client. Further, the molecular structure data C1 is managed as confidential information by the client.
  • the labeled encrypted molecular structure data C2L is data in which a label unique to a plurality of encrypted molecular structure data is associated (see FIG. 4).
  • the encrypted molecular structure data included in the labeled encrypted molecular structure data C2L is generated by the control unit 41 encrypting the molecular structure data C1 according to the encryption algorithm indicated by the encryption algorithm information D1. ..
  • the encrypted molecular structure data C2 is data generated by the control unit 41 encrypting the molecular structure data C1 according to the encryption algorithm indicated by the encryption algorithm information D1.
  • the encrypted molecular structure data C2 may include the same as the encrypted molecular structure data contained in the labeled encrypted molecular structure data C2L, or may be different from each other.
  • the functional data C3 is data relating to the function of the compound corresponding to the molecular structure data C1 (and the encrypted molecular structure data C2). It is assumed that the functional data C3 is not regarded as confidential information.
  • the configurations of the operation unit 42, the display unit 43, and the communication unit 44 are the same as the configurations of the operation unit 12, the display unit 13, and the communication unit 14 of the MI server 1, the description thereof will be omitted.
  • At least one of the following two learning target encrypted structure data is used to generate the learning model data D2 in the MI server 1.
  • (A1) Encrypted molecular structure data A2 included in general data 113a.
  • (A2) Encrypted molecular structure data C2 included in the client-derived data 113b.
  • (B1) Labeled encrypted molecular structure data C2L included in client-derived data 113b.
  • (B2) Of the encrypted molecular structure data B2 included in the general data 113a, the encrypted molecular structure data B2 obtained by encrypting the molecular structure data B1 acquired from the outside (for example, the reagent DB server 3).
  • (B3) Of the encrypted molecular structure data B2 included in the general data 113a, the encrypted molecular structure data B2 obtained by encrypting the molecular structure data B1 generated inside the MI server 1.
  • the client server 4 receives the prediction result of the compound function by MI without disclosing (transmitting) the molecular structure data C1 which is confidential information from the client server 4 to the outside. be able to.
  • FIG. 6 is a diagram illustrating a first method for predicting the function of a compound.
  • FIG. 6 shows the flow of various data processing executed by the MI server 1, the public DB server 2, and the client server 4, and the flow of data transmission / reception between each server.
  • control unit 11 (41) controls the communication unit 14 (44) and causes the communication unit 14 (44) to transmit data
  • control unit 11 (41) transmits data.
  • "(a1) encrypted molecular structure data A2 included in the general data 113a” is used as the encrypted molecular structure data to be learned, and "(a1) the encrypted molecular structure data to be functionally predicted is” ().
  • b1) The labeled encrypted molecular structure data C2L included in the client-derived data 113b ” is used.
  • the control unit 11 of the MI server 1 acquires the molecular structure data A1 to be learned and the corresponding functional data A3 from the public DB server 2 (step S101: third data acquisition step). ..
  • the control unit 11 of the MI server 1 encrypts and encrypts the acquired molecular structure data A1 and the molecular structure data A1 stored in advance in the storage unit 113 according to the encryption algorithm indicated by the encryption algorithm information D1.
  • Generate molecular structure data A2 (step S102: encryption step).
  • the control unit 11 of the MI server 1 learns by machine learning based on the generated encrypted molecular structure data A2, the functional data A3 acquired in step S101, and the functional data A3 stored in advance in the storage unit 113.
  • Model data D2 is generated (step S103: learning model generation step).
  • the control unit 11 updates the learning model data D2 with the newly generated contents.
  • For machine learning among the encrypted molecular structure data A2 of the molecular structure data A1 acquired from the public DB server 2 and the encrypted molecular structure data A2 of the molecular structure data A1 stored in advance in the storage unit 113. Only one may be used.
  • control unit 11 of the MI server 1 transmits the encryption algorithm information D1 to the client server 4 (step S104: information provision step).
  • the control unit 41 of the client server 4 that has received the encryption algorithm information D1 encrypts the molecular structure data C1 to be functionally predicted according to the encryption algorithm indicated by the encryption algorithm information D1 and assigns a label. Generates encrypted molecular structure data C2L with a label for which the function is predicted (step S105). Further, the control unit 41 transmits the labeled encrypted molecular structure data C2L to the MI server 1. In response to this, the control unit 11 of the MI server 1 receives the labeled encrypted molecular structure data C2L (step S106: first data acquisition step).
  • the control unit 11 of the MI server 1 applies the learning model represented by the learning model data D2 to each encrypted molecular structure data included in the acquired labeled encrypted molecular structure data C2L, so that each cipher Predict the function of the compound corresponding to the cryptographic structure data (step S107: prediction step).
  • the control unit 11 compares the prediction result of the function with the target value of the function received from the client server 4, and identifies the encrypted molecular structure data in which the prediction result of the function matches the target value (step S108).
  • the prediction result of the function matches the target value
  • the value of the index indicating that the function is exhibited matches the target value
  • the index is within a predetermined range, or the index is a predetermined value. The above may be the case.
  • the control unit 11 transmits the label associated with the encrypted molecular structure data specified in step S108 to the client server 4 (step S109).
  • the control unit 41 of the client server 4 identifies the encrypted molecular structure data corresponding to the received label in the labeled encrypted molecular structure data C2L, and determines the compound corresponding to the encrypted molecular structure data. , Specified as a compound exhibiting the desired function.
  • the MI server 1 includes the control unit 11, and the control unit 11 follows a predetermined encryption algorithm for the client server 4 in the above-mentioned first method.
  • the learning model data D2 for executing the encryption is provided (information providing unit), and the labeled encrypted molecular structure data C2L of the function prediction target encrypted according to the above encryption algorithm is acquired from the client server 4. (First data acquisition unit), the function of the compound corresponding to the labeled encrypted molecular structure data C2L of the function prediction target is predicted based on the learning model as a prediction model (prediction unit), and the prediction model is the compound.
  • the correlation between the encrypted structure data obtained by encrypting the structural data related to the structure according to the encryption algorithm and the functional data related to the function of the compound is shown.
  • the encrypted labeled molecular structure data C2L after encryption and performing function prediction, necessary processing is performed without receiving the molecular structure data C1 which is confidential information from the client server 4. be able to. Therefore, since the confidential information of the client is not stored inside the MI server 1, the security of the confidential information can be enhanced.
  • a prediction model (here, a learning model) that represents the correlation between encrypted molecular structure data and functional data is used, if there is encrypted molecular structure data to be functionally predicted for functional prediction, Sufficiently, it is not necessary to decrypt the encrypted molecular structure data to generate the molecular structure data. Therefore, the function of the compound can be predicted by a simple process.
  • control unit 11 encrypts the molecular structure data A1 related to the structure of the compound according to a predetermined encryption algorithm to generate the encrypted molecular structure data A2 (encryption unit), and the encrypted molecular structure data A2 and the compound.
  • a learning model as a prediction model is generated based on the functional data A3 related to the function of (learning model generation unit). According to this, the learning model can be generated in the MI server 1.
  • the learning model is generated using the encrypted molecular structure data, it is sufficient to have the encrypted molecular structure data to be functionally predicted for the function prediction using the learning model, and the encrypted molecular structure data. There is no need to decode to generate molecular structure data. Therefore, the function of the compound can be predicted by a simple process.
  • control unit 11 acquires the molecular structure data A1 and the functional data A3 from the public DB server 2 (third data acquisition unit), and generates encrypted molecular structure data A2 based on the acquired molecular structure data A1.
  • Encryption unit the learning model data D2 is generated using at least the encrypted molecular structure data A2 and the functional data A3 acquired from the public DB server 2 (learning model generation unit). This makes it possible to generate a learning model using the information of a large number of compounds disclosed by the public DB server 2. Therefore, the accuracy of predicting the function of the compound by the learning model can be improved.
  • the encryption algorithm cannot be converted back to structural data before encryption.
  • the MI server 1 cannot specify the molecular structure data C1 by decoding the labeled encrypted molecular structure data C2L received from the client server 4. Therefore, the client can receive the information provision service by MI without disclosing the molecular structure data C1 which is confidential information to any outsider including the administrator of MI server 1.
  • the program 113c encrypts the control unit 11 as a computer provided in the MI server 1 for the client server 4 to perform encryption according to a predetermined encryption algorithm.
  • First data acquisition means for acquiring the labeled encrypted molecular structure data C2L of the function prediction target encrypted according to the encryption algorithm from the information providing means for providing the encryption algorithm information D1 and the client server 4.
  • the function is made to function as a prediction means for predicting the function of the compound corresponding to the labeled encrypted molecular structure data C2L to be predicted based on the learning model as a prediction model, and the prediction model is the structural data related to the structure of the compound.
  • the correlation between the encrypted structure data obtained by encrypting the data according to the encryption algorithm and the functional data related to the function of the compound is shown.
  • necessary processing can be performed without receiving the molecular structure data C1 which is confidential information from the client server 4. Therefore, since the confidential information of the client is not stored inside the MI server 1, the security of the confidential information can be enhanced.
  • functional prediction using the learning model it is sufficient to have the encrypted molecular structure data to be the functional prediction target, and it is not necessary to decode the encrypted molecular structure data to generate the molecular structure data, which is simple.
  • the function of the compound can be predicted by the treatment.
  • the first method as an information processing method includes an information providing step of providing encryption algorithm information D1 for executing encryption according to a predetermined encryption algorithm to the client server 4, and a client server. From 4, the first data acquisition step (data acquisition step) to acquire the labeled encrypted molecular structure data C2L of the function prediction target encrypted according to the above encryption algorithm, and the labeled encrypted molecular structure of the function prediction target.
  • the prediction model includes a prediction step of predicting the function of the compound corresponding to the data C2L based on a learning model as a prediction model, and the prediction model is obtained by encrypting the structural data related to the structure of the compound according to the above encryption algorithm. The correlation between the encrypted structure data and the functional data related to the function of the above compound is shown.
  • FIG. 7 is a diagram illustrating a second method for predicting the function of a compound.
  • FIG. 7 shows the flow of various data processing performed by the MI server 1, the public DB server 2, the reagent DB server 3, and the client server 4, and the flow of data transmission / reception between each server.
  • the encrypted molecular structure data to be learned "(a1) encrypted molecular structure data A2 included in the general data 113a” and "(a2) encrypted molecular structure data included in the client-derived data 113b".
  • C2 is used, and as the encrypted molecular structure data to be functionally predicted, "(b2) Molecular structure data acquired from the outside (for example, reagent DB server 3) among the encrypted molecular structure data B2 included in the general data 113a". "Encrypted molecular structure data B2 obtained by encrypting B1" is used.
  • the control unit 11 of the MI server 1 acquires the molecular structure data A1 to be learned and the corresponding functional data A3 from the public DB server 2 (step S201: third data acquisition step). ..
  • the control unit 11 of the MI server 1 encrypts and encrypts the acquired molecular structure data A1 and the molecular structure data A1 stored in advance in the storage unit 113 according to the encryption algorithm indicated by the encryption algorithm information D1.
  • Generate molecular structure data A2 (step S202: encryption step).
  • the control unit 11 of the MI server 1 transmits the encryption algorithm information D1 to the client server 4 (step S203: information provision step).
  • the control unit 41 of the client server 4 that has received the encryption algorithm information D1 encrypts the molecular structure data C1 to be learned according to the encryption algorithm indicated by the encryption algorithm information D1 to generate the encrypted molecular structure data C2. (Step S204). Further, the control unit 41 transmits the encrypted molecular structure data C2 to be learned and the corresponding functional data C3 to be learned to the MI server 1. In response to this, the control unit 11 of the MI server 1 receives the encrypted molecular structure data C2 and the functional data C3 to be learned (step S205: first data acquisition step).
  • the control unit 11 of the MI server 1 has the encrypted molecular structure data A2 generated in step S202, the functional data A3 acquired from the public DB server 2 in step S201, the functional data A3 stored in advance in the storage unit 113, and the step.
  • the learning model data D2 is generated by machine learning based on the encrypted molecular structure data C2 and the functional data C3 acquired from the client server 4 in S205 (step S206: learning model generation step).
  • step S206 learning model generation step.
  • the encrypted molecular structure data A2 of the molecular structure data A1 acquired from the public DB server 2 the encrypted molecular structure data A2 of the molecular structure data A1 stored in advance in the storage unit 113, and the client. Only a part of the encrypted molecular structure data C2 acquired from the server 4 may be used.
  • control unit 11 of the MI server 1 acquires the molecular structure data B1 to be functionally predicted from the reagent DB server 3 (step S207: second data acquisition step), and the encryption algorithm indicated by the encryption algorithm information D1.
  • the encryption is performed according to the above to generate the encrypted molecular structure data B2 (step S208).
  • the control unit 11 of the MI server 1 applies the learning model represented by the learning model data D2 to the encrypted molecular structure data B2 generated in step S208, so that the compound corresponding to each encrypted molecular structure data is applied. Predict the function of (step S209: prediction step).
  • the control unit 11 compares the function prediction result with the function target value received from the client server 4, and identifies the encrypted molecular structure data B2 that matches the function prediction result image target value (step S210).
  • the control unit 11 transmits the molecular structure data B1 corresponding to the encrypted molecular structure data B2 specified in step S210 to the client server 4 (step S211).
  • the control unit 41 of the client server 4 identifies the compound according to the received molecular structure data B1 as a compound exhibiting a desired function.
  • the control unit 11 of the MI server 1 encrypts the structural data related to the structure of the compound according to a predetermined encryption algorithm to generate the encrypted structure data (encryption unit).
  • the learning model data D2 representing the correlation between the encrypted structure data and the functional data is generated based on the encrypted structure data and the functional data related to the function of the compound (learning model generation unit), and the client server 4 is provided with the learning model data D2.
  • the encryption algorithm information D1 for executing the encryption according to the encryption algorithm is provided (information providing unit), and the encryption molecular structure of the learning target encrypted according to the above encryption algorithm from the client server 4 is provided.
  • Data C2 and functional data C3 of the learning target related to the function of the compound corresponding to the encrypted molecular structure data C2 are acquired (first data acquisition unit), and the acquired encrypted molecular structure data C2 of the learning target and learning are acquired.
  • the training model data D2 is generated using at least the target functional data C3 (learning model generation unit).
  • the prediction accuracy of the learning model can be improved by using the information of a larger number of compounds.
  • a compound exhibiting a function desired by a client is often specified from a compound having a structure similar to that of an existing compound managed and owned by the client. Therefore, as in the second method, by generating a learning model using the encrypted molecular structure data C2 and the functional data C3 received from the client server 4, it is higher whether or not the client exhibits the desired function. A learning model that can be predicted with accuracy is obtained.
  • the learning model is generated using the encrypted molecular structure data, it is sufficient to have the encrypted molecular structure data to be functionally predicted for the function prediction using the learning model, and the encrypted molecular structure data. There is no need to decode to generate molecular structure data. Therefore, the function of the compound can be predicted by a simple process. This will result in
  • control unit 11 acquires the molecular structure data B1 related to the structure of the compound from the reagent DB server 3 that discloses the structure of the compound (second data acquisition unit), and encrypts the acquired molecular structure data B1.
  • the encrypted molecular structure data B2 of the function prediction target is generated (encryption unit), and the function of the compound corresponding to the generated encrypted molecular structure data B2 of the function prediction target is predicted based on the learning model data D2 (prediction unit). ). This makes it possible to identify a compound exhibiting a function desired by the client from among a large number of compounds published by the reagent DB server 3.
  • the program 113c generates encrypted structure data by encrypting the structural data related to the structure of the compound according to a predetermined encryption algorithm by the control unit 11 as a computer provided in the MI server 1.
  • Learning model generation means client server 4 that generates learning model data D2 representing the correlation between the encryption structure data and the functional data based on the encryption means, the encryption structure data, and the functional data related to the function of the compound.
  • the data C2 and the functional data C3 of the learning target related to the function of the compound corresponding to the encrypted molecular structure data C2 are made to function as a first data acquisition means (data acquisition means), and the learning model generation means is used.
  • the learning model data D2 is generated by using at least the encrypted molecular structure data C2 of the learning target and the functional data C3 of the learning target acquired by the first data acquisition means.
  • the second method as the information processing method includes an encryption step of encrypting the structural data related to the structure of the compound according to a predetermined encryption algorithm to generate the encrypted structure data, the encrypted structure data, and the compound.
  • the learning model generation step that generates the learning model data D2 that represents the correlation between the encrypted structure data and the functional data, and the client server 4 are encrypted according to the encryption algorithm.
  • the information providing step for providing the encryption algorithm information D1 for executing the above, the encrypted molecular structure data C2 of the learning target encrypted according to the above encryption algorithm from the client server 4, and the encrypted molecular structure data C2.
  • the learning model data D2 is generated by using at least the encrypted molecular structure data C2 and the functional data C3 to be learned. According to such a method, necessary processing can be performed without receiving the molecular structure data C1 which is confidential information from the client server 4. Therefore, since the confidential information of the client is not stored inside the MI server 1, the security of the confidential information can be enhanced.
  • FIG. 8 is a diagram illustrating a third method for predicting the function of a compound.
  • the third method as the encrypted molecular structure data to be learned, "(a1) encrypted molecular structure data A2 included in the general data 113a” and "(a2) encrypted molecular structure data included in the client-derived data 113b". "C2" is used, and as the encrypted molecular structure data to be functionally predicted, "(b3) Of the encrypted molecular structure data B2 included in the general data 113a, the molecular structure data B1 generated inside the MI server 1 is used. Encrypted molecular structure data B2 obtained by encryption is used.
  • steps S301 to S306 in the third method are the same as steps S201 to S206 in the second method, the description thereof will be omitted.
  • the control unit 11 of the MI server 1 mechanically and randomly generates a plurality of molecular structure data B1 using a genetic algorithm or the like (step S307), and encrypts the data according to the encryption algorithm indicated by the encryption algorithm information D1.
  • Generate a plurality of encrypted molecular structure data B2 step S308).
  • step S309 prediction step
  • the encrypted molecular structure data B2 generated in step S308 is used as the encrypted molecular structure data to be functionally predicted, and the function is predicted. Since steps S310 and S311 are the same as steps S210 and S211 of the second method, the description thereof will be omitted.
  • the control unit 11 of the MI server 1 generates the molecular structure data B1 (structural data generation unit), encrypts the generated molecular structure data B1, and encrypts the function prediction target.
  • the chemical molecular structure data B2 is generated (encryption unit), and the function of the compound corresponding to the generated encrypted molecular structure data B2 to be predicted is predicted based on the learning model data D2 (prediction unit). This makes it possible to increase the possibility that the compound exhibiting the desired function can be identified by the client even when sufficient data of the compound whose function is to be predicted cannot be obtained from the outside of the MI server 1.
  • FIG. 9 is a diagram illustrating a fourth method for predicting the function of a compound.
  • the fourth method as the encrypted molecular structure data to be learned, "(a1) encrypted molecular structure data A2 included in the general data 113a” and “(a2) encrypted molecular structure data included in the client-derived data 113b". "C2" is used, and "(b1) Labeled encrypted molecular structure data C2L included in the client-derived data 113b" is used as the encrypted molecular structure data to be functionally predicted.
  • steps S401 to S403 and S406 of the fourth method are the same as steps S301 to S303 and S306 of the third method, the description thereof will be omitted.
  • the control unit 41 of the client server 4 that has received the encryption algorithm information D1 in step S403 encrypts the molecular structure data C1 according to the encryption algorithm indicated by the encryption algorithm information D1, and the encrypted molecular structure data C2 to be learned. And the encrypted molecular structure data C2L with a label to be functionally predicted (step S404). Further, the control unit 41 transmits the encrypted molecular structure data C2 to be learned and the corresponding functional data C3 to be learned to the MI server 1, and the control unit 11 of the MI server 1 transmits the encryption to be learned. Receives the cryptographic structure data C2 and the functional data C3 (step S405: first data acquisition step).
  • control unit 41 transmits the labeled encrypted molecular structure data C2L of the function prediction target to the MI server 1, and the control unit 11 of the MI server 1 receives the labeled encrypted molecular structure data C2L.
  • Step S407 first data acquisition step
  • the control unit 11 of the MI server 1 applies the learning model represented by the training model data D2 to each encrypted molecular structure data included in the labeled encrypted molecular structure data C2L acquired in step S407. , Predict the function of the compound corresponding to each encrypted molecular structure data (step S408: prediction step). Subsequent steps S409 and S410 are the same as steps S108 and S109 of the first method, and thus description thereof will be omitted.
  • the control unit 11 of the MI server 1 acquires the encrypted molecular structure data C2 and the functional data C3 of the learning target encrypted from the client server 4 (first).
  • Data acquisition unit) the learning model data D2 is generated using at least the acquired encrypted molecular structure data C2 and functional data C3 (learning model generation unit), and the function encrypted according to the encryption algorithm from the client server 4 Acquired the labeled encrypted molecular structure data C2L of the prediction target (first data acquisition unit), and acquired the function of the compound corresponding to the labeled encrypted molecular structure data C2L of the prediction target based on the learning model data D2.
  • Predict Prediction unit
  • the client exhibits the desired function, and at the same time, the client exhibits the desired function from the encrypted molecular structure data C2 provided by the client.
  • the compound can be identified.
  • the present invention is not limited to the above embodiment and each modification, and various modifications can be made.
  • the encryption of the encrypted molecular structure data C2 (first method) acquired from the client server 4 and the encryption of the molecular structure data B1 acquired from the reagent DB server 3 The encrypted molecular structure data B2 (second method) and the encrypted molecular structure data B2 (third method) of the molecular structure data B1 generated inside the MI server 1 are exemplified, but the purpose is not limited thereto.
  • any encrypted molecular structure data obtained by encrypting the molecular structure data of the compound can be used, and the acquisition route thereof is not limited to that exemplified in this embodiment.
  • the encrypted molecular structure data C2 to be functionally predicted may be acquired from the client server 4.
  • the present invention is not limited to this, and a reversible encryption algorithm may be used.
  • a reversible encryption algorithm may be used.
  • the encryption algorithm is not limited to the one using a hash function.
  • the learning model is generated in the MI server 1, but the learning model is not limited to this, and an existing learning model (for example, a learning model generated in an external device) may be used as it is.
  • the MI server 1 does not have to have a learning model generation function (learning model generation unit).
  • the MI server 1 when the encrypted molecular structure data of the function prediction target is acquired from the outside of the MI server 1, the MI server 1 has a function (encryption unit) for encrypting the molecular structure data. It does not have to be.
  • the prediction model used for predicting the function of a compound is not limited to this learning model.
  • the prediction model for example, a deductive prediction model that predicts the function of the compound from a known principle or regularity of the compound may be used. Even when the deductive prediction model is used, the MI server 1 does not have to have a learning model generation function (learning model generation unit). Further, when the encrypted molecular structure data of the function prediction target is acquired from the outside of the MI server 1, the MI server 1 does not have to have a function (encryption unit) for encrypting the molecular structure data. ..
  • each of the MI server 1, the public DB server 2, the reagent DB server 3, and the client server 4 has been described by using an example consisting of a single server device, but the present invention is not limited to this. Any of these server devices may be replaced with a system consisting of a plurality of devices. For example, at least a part of the program and data stored in the storage unit 113 of the MI server 1 may be stored in a storage device external to the MI server 1.
  • the present invention can be used for information processing devices, programs and information processing methods.
  • MI server information processing device
  • Control unit Encryption unit, learning model generation unit, information provision unit, first to third data acquisition unit, prediction unit, structural data generation unit
  • 111 CPU 112 RAM 113 Storage unit 113a General data 113b Client-derived data 113c
  • Program 12 Operation unit 13
  • Display unit 14 Communication unit 15
  • Bus 2 Public DB server (database) 3 Reagent DB server (second external device) 4 Client server (first external device) 41
  • Control unit 411 CPU 412 RAM 413 Storage unit 413a Program 42 Operation unit 43
  • Display unit 44 Communication unit 45
  • Compound information processing system A1, B1, C1 Molecular structure data (structural data)
  • A2, B2, C2 Encrypted molecular structure data (encrypted structure data)
  • C2L labeled encrypted molecular structure data Encrypted structure data
  • C3 Function data
  • D1 Encryption algorithm information
  • D2 Learning model data N N Communication network

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un dispositif de traitement d'informations, un programme, et un procédé de traitement d'informations, avec lesquels la sécurité d'informations confidentielles se rapportant à la structure d'un composé peut être accrue. Ce dispositif de traitement d'informations comprend : une unité de fourniture d'informations pour fournir des informations d'algorithme de codage pour effectuer un codage selon un algorithme de codage prescrit par rapport à un premier dispositif externe ; une première unité d'acquisition de données pour acquérir, du dispositif externe, des premières données de structure codées qui ont été codées selon l'algorithme de codage et qui doivent être soumises à une prédiction de fonction ; et une unité de prédiction pour prédire, sur la base d'un modèle de prédiction prescrit, une fonction d'un composé correspondant aux données de structure codées à soumettre à une prédiction de fonction. Le modèle de prédiction représente une relation de correspondance entre des données de structure codées, qui ont été obtenues par codage de données de structure se rapportant à la structure d'un composé selon l'algorithme de codage, et des données de fonction, qui se rapportent à une fonction du composé.
PCT/JP2021/015183 2020-05-14 2021-04-12 Dispositif de traitement d'informations, programme, et procédé de traitement d'informations WO2021229973A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2022522566A JPWO2021229973A1 (fr) 2020-05-14 2021-04-12

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2020084839 2020-05-14
JP2020-084839 2020-05-14

Publications (1)

Publication Number Publication Date
WO2021229973A1 true WO2021229973A1 (fr) 2021-11-18

Family

ID=78525782

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/015183 WO2021229973A1 (fr) 2020-05-14 2021-04-12 Dispositif de traitement d'informations, programme, et procédé de traitement d'informations

Country Status (2)

Country Link
JP (1) JPWO2021229973A1 (fr)
WO (1) WO2021229973A1 (fr)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013038698A1 (fr) * 2011-09-14 2013-03-21 独立行政法人産業技術総合研究所 Système de recherche, procédé de recherche et programme
JP2018054765A (ja) * 2016-09-27 2018-04-05 日本電気株式会社 データ処理装置、データ処理方法、およびプログラム
WO2019004437A1 (fr) * 2017-06-30 2019-01-03 学校法人 明治薬科大学 Dispositif de prédiction, procédé de prédiction, programme de prédiction, dispositif de production de données d'entrée de modèle d'apprentissage et programme de production de données d'entrée de modèle d'apprentissage

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013038698A1 (fr) * 2011-09-14 2013-03-21 独立行政法人産業技術総合研究所 Système de recherche, procédé de recherche et programme
JP2018054765A (ja) * 2016-09-27 2018-04-05 日本電気株式会社 データ処理装置、データ処理方法、およびプログラム
WO2019004437A1 (fr) * 2017-06-30 2019-01-03 学校法人 明治薬科大学 Dispositif de prédiction, procédé de prédiction, programme de prédiction, dispositif de production de données d'entrée de modèle d'apprentissage et programme de production de données d'entrée de modèle d'apprentissage

Also Published As

Publication number Publication date
JPWO2021229973A1 (fr) 2021-11-18

Similar Documents

Publication Publication Date Title
US11520912B2 (en) Methods, media, apparatuses and computing devices of user data authorization based on blockchain
CN102156835B (zh) 内容管理软件的安全局部更新
JP6926429B2 (ja) データ処理装置、データ処理方法、およびプログラム
CN102318257B (zh) 用于信息网络的密钥分发方案
CN110768787B (zh) 一种数据加密、解密方法及装置
CN108520183A (zh) 一种数据存储方法及装置
CN111783075A (zh) 基于密钥的权限管理方法、装置、介质及电子设备
Zhang et al. Cryptanalysis and improvement of an efficient and secure medical image protection scheme
CN105027096A (zh) 保护特权计算操作的结果
CN101187965A (zh) 过滤对数据对象的访问
US7996892B2 (en) Method and apparatus for using a proxy to manage confidential information
CN104620226A (zh) 确保公共的、私人的和移动的装置中的私人信息的安全
US10826693B2 (en) Scalable hardware encryption
Mahalakshmi et al. IoT sensor-based smart agricultural system
KR20200057900A (ko) 개인정보 처리를 위한 암호화 시스템 및 방법
Chyrun et al. Information System for Connection to the Access Point with Encryption WPA2 Enterprise
CN110717809A (zh) 一种基于大数据的车辆交易方法、装置、电子设备及计算机可读介质
CN109412791B (zh) 密钥信息处理方法、装置、电子设备及计算机可读介质
CN112134882A (zh) 用于在网络中匿名化地传输数据的系统和方法
WO2021229973A1 (fr) Dispositif de traitement d'informations, programme, et procédé de traitement d'informations
US11139969B2 (en) Centralized system for a hardware security module for access to encryption keys
CN116361849A (zh) 一种加密数据库的备份数据加密、解密方法及装置
US20220343010A1 (en) System and Method to enable Shared SaaS Multi-Tenancy using Customer Data Storage, Customer-controlled Data Encryption Keys
Alhakami et al. A Usability Management Framework for Securing Healthcare Information System.
Rao et al. A novel approach for identification of hadoop cloud temporal patterns using map reduce

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21804244

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022522566

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21804244

Country of ref document: EP

Kind code of ref document: A1