Back to results
US 10,951,695 B2Granted
System and methods for identification of peer entities
AssigneeAon Global Operations SE, Singapore Branch
InventorOisin Lyons
Life of this patent
Estimating the term…
Abstract
In an illustrative embodiment, systems and methods for automatically identifying peer organizations of a subject organization within a transactional platform identify peer organizations of the subject organization by accessing characteristic features and transactional features of the subject organization, providing a portion of the features to similarity analysis models, and obtaining, through executing the models, predicted peer organizations. The predicted peers may be presented to a user arranged in priority order with some peer features of the respective organization and a control for selecting the respective peer. The user may select a subset of the predicted peers for use in comparison metrics.
Claims20
A system for automatically identifying peer organizations of a subject organization within a transactional platform, the system comprising:
- processing circuitry;
- and
- a non-transitory computer readable medium having instructions stored thereon, wherein the instructions, when executed by the processing circuitry, cause the processing circuitry to:
- receive, via a network from a remote computing device of a user, a request for peer comparison with the subject organization,
- in real time responsive to receiving the request, identify a plurality of peer organizations of the subject organization, wherein identifying the plurality of peer organizations comprises
- determining a plurality of features of the subject organization, wherein the plurality of features comprises i) one or more characteristic features of the subject organization including at least one of industry or region and ii) one or more transactional features of the subject organization,
- generating, from the plurality of features of the subject organization, one or more organization data sets, each organization data set comprising at least a portion of the one or more characteristic features and/or at least a portion of the one or more transactional features of the subject organization,
- wherein each of the one or more organization data sets is customized to features used to train a respective model of one or more models, wherein each model of the one or more models is trained to detect similarities between a given organization and a plurality of other organizations according to the features in the respective organization data set,
- for each of the one or more organization data sets,
- obtaining, through executing the respective organization data set on the corresponding model of the one or more models trained using the features of the respective organization data set, a plurality of candidate organizations of the plurality of other organizations sharing one or more similarities with the subject organization according to the features of the respective organization data set, and
- selecting, from the plurality of candidate organizations to each organization data set of the one or more organization data sets, the plurality of peer organizations,
- prepare, for presentation at the remote computing device, a graphical user interface for presenting at least a portion of the plurality of peer organizations, wherein, for each peer of the portion of the plurality of peer organizations, the respective peer is represented along with i) at least a portion of the plurality of peer features of the respective peer within the graphical user interface and ii) a control for selecting the respective peer for use in comparison metrics,
- receive, via the graphical user interface, indication of a set of peer organizations selected from the portion of the portion of the plurality of peer organizations, and
- analyze the set of peer organizations in view of the subject organization to determine one or more features of the plurality of features correlating between the subject organization and the set of peer organizations for use in training an additional model of the one or more models, wherein
- the one or more organization data sets lack the one or more features.
12 more elements
A method for automatically identifying peer entities of a subject entity within a transactional platform, the method comprising:
- receiving, by processing circuitry via a network from a remote computing device of a user, a request for peer comparison with the subject entity;
- in real time responsive to receiving the request, identifying, by the processing circuitry, a plurality of peer entities of the subject entity, wherein identifying the plurality of peer entities comprises
- determining a plurality of features of the subject entity, wherein the plurality of features comprises i) one or more characteristic features of the subject entity including at least one of industry or region and ii) one or more transactional features of the subject entity,
- generating, from the plurality of features of the subject entity, one or more entity data sets, each entity data set comprising at least a portion of the one or more characteristic features and/or at least a portion of the one or more transactional features of the subject entity,
- wherein each of the one or more entity data sets comprises features used to train a respective model of one or more models, wherein
- each model of the one or more models is trained to detect similarities between a given entity and a plurality of other entities according to the features in the respective entity data set, and
- each data set of the one or more entity data sets represents feature correlations identified through analyzing the plurality of features of identified groups of peer entities for commonalities,
- for each of the one or more entity data sets
- obtaining, through executing the respective entity data set on the corresponding model of the one or more models trained using the features of the respective entity data set, a plurality of candidate entities sharing one or more similarities with the subject entity according to the features of the respective organization data set, and
- selecting, from the plurality of candidate entities to each entity data set of the one or more entity data sets, the plurality of peer entities;
- and
- preparing, by the processing circuitry for presentation at the remote computing device, a graphical user interface for presenting at least a portion of the plurality of peer entities, wherein, for each peer of the plurality of peer entities, the respective peer is represented along with i) at least a portion of the plurality of peer features of the respective entity within the graphical user interface and ii) a control for selecting the respective peer for use in comparison metrics.
9 more elements
Dependencies are inferred from the claim text.