US 11,983,200 B2
Processing system, processing device, processing method, program, and information storage medium
Yu Hirate, Tokyo (JP); and Hiroki Mihara, Tokyo (JP)
Assigned to Rakuten Group, Inc., Tokyo (JP)
Appl. No. 16/316,119
Filed by Rakuten Group, Inc., Tokyo (JP)
PCT Filed Jul. 31, 2017, PCT No. PCT/JP2017/027763
§ 371(c)(1), (2) Date Jan. 8, 2019,
PCT Pub. No. WO2019/026151, PCT Pub. Date Feb. 7, 2019.
Prior Publication US 2021/0357430 A1, Nov. 18, 2021
Int. Cl. G06F 16/28 (2019.01); G06F 17/18 (2006.01); G06Q 30/0203 (2023.01)
CPC G06F 16/285 (2019.01) [G06F 17/18 (2013.01); G06Q 30/0203 (2013.01)] 9 Claims
OG exemplary drawing
 
1. An information processing system comprising a first processing device and a second processing device, the information processing system comprising:
the first processing device comprising:
a memory storing one or more instructions; and
a processor configured to execute the one or more instructions to at least:
obtain first characteristics information indicating a characteristic of each of a plurality of users in a first environment relating to a first service,
divide the plurality of users into first clusters according to the obtained first characteristics information,
generate belonging information obtained by dividing the plurality of users into the first clusters, wherein the belonging information obfuscates the first characteristics information; and
transmit, to the second processing device in a second environment relating to a second service separate from the first service, the belonging information indicating which of the first clusters each of the plurality of users belongs to,
wherein the belonging information indicates which cluster a user belongs to using a vector in which an element associated with the cluster to which the user belongs is set to nonzero and elements associated with clusters to which the user does not belong are set to zero,
wherein the nonzero element is set to a degree of similarity between a centroid vector based on the first characteristics information of a cluster to which the user belongs and the first characteristics information obtained for the user, and
wherein the degree of similarity is a cosine of an angle formed between the first characteristics information of the user and the cluster to which the user belongs to; and
the second processing device comprising:
a memory storing one or more instructions; and
a processor configured to execute the one or more instructions to:
obtain second characteristics information indicating a characteristic of each of the plurality of users in the second environment,
divide the plurality of users into second clusters according to the transmitted belonging information and the obtained second characteristics information
divide the plurality of users into interim clusters corresponding to the first clusters according to the transmitted belonging information,
calculate an average of the second characteristics information obtained for users who belong to each of the interim clusters, and
divide the plurality of users into the second clusters according to degrees of similarity between the obtained second characteristics information and the calculated averages; and
wherein the processor of the second processing device is further configured to execute the one or more instructions to divide the plurality of users into the second clusters according to integration vectors obtained by performing a linear summation, for each user, of the obtained second characteristics information and vectors relating to the transmitted belonging information.