US 11,797,503 B2
Systems and methods for enhanced mapping and classification of data
David von Rickenbach, Baar (CH); David Oliver, London (GB); Daniella Tsar, Orpington (GB); Kim Hau, London (GB); Dorcas Mbwiti, Essex (GB); Johannes Schleith, London (GB); Guillaume Mosching, Geneva (CH); and Sanzio Monti, Zurich (CH)
Assigned to Thomson Reuters Enterprise Centre GmbH, Zug (CH)
Filed by Thomson Reuters Enterprise Centre GmbH, Zug (CH)
Filed on Oct. 12, 2020, as Appl. No. 17/68,818.
Application 17/068,818 is a continuation of application No. 16/181,680, filed on Nov. 6, 2018, granted, now 10,803,033.
Claims priority of provisional application 62/581,802, filed on Nov. 6, 2017.
Prior Publication US 2021/0026823 A1, Jan. 28, 2021
Int. Cl. G06F 16/00 (2019.01); G06F 16/22 (2019.01); G06N 20/00 (2019.01); G06F 11/07 (2006.01); G06Q 40/12 (2023.01); G06F 16/28 (2019.01); G06N 7/01 (2023.01)
CPC G06F 16/22 (2019.01) [G06F 11/0727 (2013.01); G06F 11/0793 (2013.01); G06F 16/285 (2019.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01); G06Q 40/12 (2013.12)] 19 Claims
OG exemplary drawing
 
1. A method of enhanced mapping of data to a target document, the method comprising:
obtaining, by one or more processors, one or more source documents from one or more data sources, wherein the one or more source documents comprise source transaction data, wherein the source transaction data comprises data having different formats, and wherein the different formats correspond to reporting requirements in one or more jurisdictions;
mapping, by the one or more processors, the source transaction data from at least one of the one or more source documents to at least one target document structure to generate mapped transaction data, wherein the at least one target document structure has a format that corresponds to reporting requirements of a target jurisdiction that is different than at least one of the one or more jurisdictions, and wherein the mapping includes:
identifying at least one feature of at least one source column defined in at least one data structure of the source transaction data, each of the at least one source column associated with an aspect of the source transaction data; and
applying the identified at least one feature of the at least one source column to a classification algorithm, wherein the classification algorithm is configured to determine that the at least one feature indicates that an aspect associated with the at least one source column corresponds to an aspect associated with at least one target column of target columns defined in the at least one target document structure, the target columns associated with aspects of transaction data; and
generating, by the one or more processors, a structured report that includes at least the mapped transaction data structured in accordance with the at least one target document structure.