Information processing device, information processing system, and information processing method to determine correlation of data

An information processing system, comprising:

• a first information processing device having k number of a first plurality of vectors including secure data as an element;
• and
• a second information processing device having m number of a second plurality of vectors including secure data as an element, wherein
• the first information processing device is configured to receive vector information regarding a vector Y selected from the m number of the second plurality of vectors, as encrypted data, from the second information processing device, wherein
• each of the m number of the second plurality of vectors has n number of elements, and
• each of the n number of the elements has a sample-based element value including binary data of 1 or 0,
• a data processing unit of the first information processing device is configured to:
• calculate a number of element-based sample identifiers each having a specific common value, from a vector X and the vector Y, wherein
• the number of the element-based sample identifiers is calculated based on a secure computation,
• the vector X is selected from the k number of the first plurality of vectors of the first information processing device,
• each of the k number of the first plurality of vectors has the n number of the elements, and
• each of the n number of the elements has the sample-based element value including the binary data of 1 or 0;
• determine a correlation between the vector X and the vector Y corresponding to the calculated number;
• sequentially calculate the element-based sample identifiers each having the specific common value for each of the second plurality of vectors selected from the m number of the second plurality of vectors to the vector X;
• sequentially determine the correlation between the vector X and each of the second plurality of vectors;
• extract a number of common sample identifiers from the element-based sample identifiers each having 1 in value in the elements of the vector X and from the element-based sample identifiers each having 1 in value in the elements of the vector Y;
• calculate an inner product value between the elements of the vector X and the elements of the vector Y based on the extraction;
• and
• sequentially extract common element-based sample identifiers each having 1 in value for each of the second plurality of the vectors selected from the m number of the second plurality of vectors to the vector X.

An information processing method executed by an information processing device, the information processing method including:

• determining a correlation between a vector X a vector Y, wherein
• the vector X is a vector selected from k number of a first plurality of vectors,
• each of the k number of the first plurality of vectors has n number of elements,
• each of the n number of the elements has a sample-based element value including binary data of 1 or 0,
• the vector Y is a vector selected from m number of a second plurality of vectors,
• each of the m number of the second plurality of vectors has the n number of the elements, and
• each of the n number of the elements has the sample-based element value including the binary data of 1 or 0;
• calculating a number of element-based sample identifiers each having a specific common value, from the vector X and the vector Y, wherein the number of the element-based sample identifiers is calculated based on a secure computation;
• determining the correlation between the vector X and the vector Y corresponding to the calculated number;
• sequentially calculating the element-based sample identifiers each having the specific common value for each of the second plurality of vectors selected from the m number of the second plurality of vectors to the vector X;
• sequentially determining the correlation between the vector X and each of the second plurality of vectors;
• extracting a number of common sample identifiers from the element-based sample identifiers each having 1 in value in the elements of the vector X and from the element-based sample identifiers each having 1 in value in the elements of the vector Y;
• calculating an inner product value between the elements of the vector X and the elements of the vector Y based on the extraction;
• and
• sequentially extracting common element-based sample identifiers each having 1 in value for each of the second plurality of the vectors selected from the m number of the second plurality of vectors to the vector X.

An information processing method executed in an information processing system including a first information processing device and a second information processing device, the information processing method comprising:

• receiving, by the first information processing device, vector information regarding a vector Y selected from m number of a second plurality of vectors, as encrypted data, from the second information processing device, wherein
• the second information processing device has the m number of the second plurality of vectors including secure data as an element,
• each of the m number of the second plurality of vectors has n number of elements, and
• each of the n number of the elements has a sample-based element value including binary data of 1 or 0;
• calculating, by a data processing unit of the first information processing device, a number of element-based sample identifiers each having a specific common value, from a vector X and the vector Y, wherein
• the number of the element-based sample identifiers is calculated based on a secure computation,
• the first information processing device has k number of a first plurality of vectors including secure data as an element,
• the vector X is selected from the k number of the first plurality of vectors of the first information processing device, and
• each of the k number of the first plurality of vectors has the n number of the elements, and
• each of the n number of the elements has the sample-based element value including the binary data of 1 or 0;
• determining, by the data processing unit of the first information processing device, a correlation between the vector X and the vector Y, corresponding to the calculated number;
• sequentially calculating, by the data processing unit of the first information processing device, the element-based sample identifiers each having the specific common value for each of the second plurality of vectors selected from the m number of the second plurality of vectors to the vector X;
• sequentially determining, by the data processing unit of the first information processing device, the correlation between the vector X and each of the second plurality of vectors;
• extracting, by the data processing unit of the first information processing device, a number of common sample identifiers from the element-based sample identifiers each having 1 in value in the elements of the vector X and from the element-based sample identifiers each having 1 in value in the elements of the vector Y;
• calculating, by the data processing unit of the first information processing device, an inner product value between the elements of the vector X and the elements of the vector Y based on the extraction;
• and
• sequentially extracting, by the data processing unit of the first information processing device, common element-based sample identifiers each having 1 in value for each of the second plurality of the vectors selected from the m number of the second plurality of vectors to the vector X.

A non-transitory computer-readable medium having stored thereon, computer-executable instructions which when executed by an information processing apparatus cause the information processing apparatus to execute operations, the operations comprising:

• determining a correlation between a vector X and a vector Y, wherein
• the vector X is a vector selected from k number of a first plurality of vectors,
• each of the k number of the first plurality of vectors has n number of elements,
• each of the n number of the elements has a sample-based element value including binary data of 1 or 0,
• the vector Y is a vector selected from m number of a second plurality of vectors,
• each of the m number of the second plurality of vectors has the n number of the elements, and
• each of the n number of the elements has the sample-based element value including the binary data of 1 or 0;
• calculating a number of element-based sample identifiers each having a specific common value, from the vector X and the vector Y, wherein the number of the element-based sample identifiers is calculated based on a secure computation;
• determining the correlation between the vector X and the vector Y corresponding to the calculated number;
• sequentially calculating the element-based sample identifiers each having the specific common value for each of the second plurality of vectors selected from the m number of the second plurality of vectors to the vector X;
• sequentially determining the correlation between the vector X and each of the second plurality of vectors;
• extracting a number of common sample identifiers from the element-based sample identifiers each having 1 in value in the elements of the vector X and from the element-based sample identifiers each having 1 in value in the elements of the vector Y;
• calculating an inner product value between the elements of the vector X and the elements of the vector Y based on the extraction;
• and
• sequentially extracting common element-based sample identifiers each having 1 in value for each of the second plurality of the vectors selected from the m number of the second plurality of vectors to the vector X.