US 11,940,955 B2
Method for data structure relationship detection
Hisham Abu-Abed, Toronto (CA); Xiuzhan Guo, Toronto (CA); and Joel Ian Tousignant-Barnes, Toronto (CA)
Assigned to ROYAL BANK OF CANADA, Toronto (CA)
Filed by ROYAL BANK OF CANADA, Toronto (CA)
Filed on Apr. 22, 2022, as Appl. No. 17/726,869.
Application 17/726,869 is a continuation of application No. 16/353,804, filed on Mar. 14, 2019, granted, now 11,314,693.
Claims priority of provisional application 62/642,916, filed on Mar. 14, 2018.
Prior Publication US 2022/0253406 A1, Aug. 11, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/174 (2019.01); G06F 16/90 (2019.01); G06F 16/901 (2019.01); G06F 16/9038 (2019.01); G06F 16/906 (2019.01); G06N 7/00 (2023.01)
CPC G06F 16/1744 (2019.01) [G06F 16/9024 (2019.01); G06F 16/9038 (2019.01); G06F 16/906 (2019.01); G06N 7/00 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A computer system for automated estimation of relationships among a plurality of data elements, the system comprising:
a data receiver configured to receive one or more input data sets including the plurality of data elements, the one or more input data sets including a set of data records C;
a classifier engine computer processor configured to:
establish one or more linkage relations among data records in the set of data records C, the one or more linkage relations maintained as an adjacency matrix data structure on a non-transitory computer memory;
extend the one or more linkage relations into equivalence relations, updating the adjacency matrix data structure; and
generate an output data set by transforming the set of data records C using at least the adjacency matrix data structure, the output data set compressed relative to the set of data records C;
wherein the equivalence relations include custom character, and D1 and D2 are defined as subsets of data records in the set of data records C, and the one or more equivalence relations are used to partition the set of data records C into one or more partitions for generating the output data set, the output data set generated by transforming the set of data records C in accordance to the one or more partitions;
wherein the equivalence relations include the relation:
given two sets of data sets custom character1 and custom character2, a transitive equivalence relation custom character on custom character1, and a set function ƒ: custom character1→custom character2 such that D1custom characterD2 implies ƒ(D1)=ƒ(D2) then induced functions are provided showing equivalence classes of D1 can be linked to D2;
wherein the equivalence relations include the relation:
Let custom character1 and custom character2 be equivalence relations on X and Y, respectively and let ƒ: X→Y be a function such that xcustom character1y⇒ƒ(x)custom character2ƒ(y);
where ƒ is defined by ƒ([x])=[ƒ(x)]:

OG Complex Work Unit Math