| CPC G06Q 40/04 (2013.01) [G06Q 20/4016 (2013.01); G06Q 30/0185 (2013.01); G06Q 30/0215 (2013.01); G06Q 40/06 (2013.01); G05B 2219/31396 (2013.01); G06N 20/00 (2019.01)] | 10 Claims |
|
1. A computer-implemented method comprising:
onboarding asset data defining an asset to be listed for trading at an Alternative Trading System (ATS), comprising:
utilizing appropriate models, from among a plurality of different models, based on document type to extract a portion of the asset data from each document in a plurality of documents; and
formulating a checklist reflecting a summary of the asset data collectively including data for listing the asset for trading;
utilizing a further model annotating the plurality of documents forming a plurality of annotated documents indicating correct examples and incorrect examples; and
automatically, as part of a continuous training cycle, and concurrently with onboarding the asset data, training the plurality of different models based on the plurality of annotated documents, wherein the plurality of different models includes at least a regulatory model and an anti-fraud security measure model, the training evolving and improving performance of the plurality of different models, including:
receiving a selection of a regulatory model from among the plurality of different models, wherein the regulatory model identifies modifications to the formulated checklist and asset data locations within the plurality of submitted documents based on changing laws and regulations associated with the asset;
iteratively training the regulatory model utilizing at least a subset of the plurality of annotated documents and at least one previously annotated document until a first user feedback score is above a first predetermined threshold indicative of model improvement for at least one of: recall or precision, including:
running the regulatory model producing first responses;
receiving the first user feedback score indicating correctness of the produced first responses; and
checking the first feedback score relative to the first predetermined threshold; and
publishing the regulatory model responsive to the user feedback score being above the first predetermined threshold;
receiving a selection of a first anti-fraud security measure model from among the plurality of different models, wherein the first anti-fraud security measure model identifies manipulative actions or irregularities within the plurality of submitted documents;
iteratively training the first anti-fraud security measure model in parallel with the regulatory model utilizing at least another subset of the plurality of annotated documents and at least one other previously annotated document until a second user feedback score is above a second predetermined threshold indicative of model improvement for at least one of: recall or precision, including:
running the first anti-fraud security measure model producing second responses;
receiving the second user feedback score indicating correctness of the second produced responses; and
checking the second user feedback score relative to the second predetermined threshold; and
publishing the first anti-fraud security measure model responsive to the user feedback score being above the second predetermined threshold;
receiving a selection of a second anti-fraud security measure model from among the plurality of different models, wherein the second anti-fraud security measure model identifies manipulative actions or irregularities within the plurality of submitted documents;
iteratively training the second anti-fraud security measure model in parallel with the regulatory model and first anti-fraud security measure model utilizing at least a further subset of the plurality of annotated documents and at least one further previously annotated document until a third user feedback score is above a third predetermined threshold indicative of model improvement for at least one of: recall or precision, including:
running the second anti-fraud security measure model producing third responses;
receiving the third user feedback score indicating correctness of the third produced responses; and
checking the third user feedback score relative to the third predetermined threshold; and
publishing the second anti-fraud security measure model side-by-side with the first anti-fraud security measure model responsive to the third user feedback score being above the third predetermined threshold.
|