US 11,688,017 B1
Computer system and method for detecting, extracting, weighing, benchmarking, scoring, reporting and capitalizing on complex risks found in buy/sell transactional agreements, financing agreements and research documents
James Callahan, Greenwich, CT (US)
Filed by James Callahan, Greenwich, CT (US)
Filed on Dec. 17, 2021, as Appl. No. 17/553,949.
Application 17/553,949 is a continuation of application No. 17/061,650, filed on Oct. 2, 2020, granted, now 11,205,233.
Claims priority of provisional application 62/909,494, filed on Oct. 2, 2019.
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 40/08 (2012.01); G06F 16/9538 (2019.01); G06Q 40/06 (2012.01); G06F 16/955 (2019.01); G06Q 10/10 (2023.01); G06F 16/958 (2019.01); G06Q 30/018 (2023.01); G06Q 40/12 (2023.01); G06Q 40/03 (2023.01); G06V 30/40 (2022.01); G06V 30/10 (2022.01)
CPC G06Q 40/08 (2013.01) [G06F 16/9538 (2019.01); G06F 16/9558 (2019.01); G06F 16/986 (2019.01); G06Q 10/103 (2013.01); G06Q 30/018 (2013.01); G06Q 40/03 (2023.01); G06Q 40/06 (2013.01); G06Q 40/12 (2013.12); G06V 30/10 (2022.01); G06V 30/40 (2022.01)] 12 Claims
OG exemplary drawing
 
1. A computer system for monitoring one or more transactional documents of a first transaction for changing risk profiles relative to prior transactions of a similar type, the computer system comprising:
an analyst computer device for an analyst that is tasked with analyzing the transactional documents of the first transaction for risk to parties to the first transaction, wherein the analyst computer device comprises is configured to display a graphical user interface (GUI) for the analyst to analyze the transaction documents; and
a back-end computer system that is in communication with the analyst computer device, wherein the back-end computer system comprises:
a transaction document database that stores the one or more transactional documents of the first transaction in word-searchable form, wherein the back-end computer system comprises an OCR component for converting transactional documents that are not word-searchable to word-searchable form, wherein the OCR component identifies individual characters in the transactional documents and identifies equivalent plain text characters for the identified individual characters using matrix matching and/or feature extraction;
a query database that stores pre-determined queries for the analyst to investigate in the one or more transactional documents for the first transaction, wherein for at least some of the pre-determined queries, the query database also stores corresponding suggestions in the one or more transactional documents for the analyst to review to respond to the query, wherein the suggestions are based on prior reviews of transactional documents for prior transactions that are similar to the first transaction; and
a back-end server for serving interactive query node tree displays to the analyst computer device that are displayed by the analyst computer device, wherein the interactive query node tree displays comprise an interactive query node tree display that displays an interactive query node tree, wherein:
each query node in the interactive query node tree corresponds to a separate query designed to assess risk for the first transaction and wherein each query node comprises a hyperlink;
upon the analyst activating the hyperlink for a first query node in the interactive query node tree display, a corresponding query for first query node is displayed for the analyst in a first query display, wherein the first query display further comprises:
means for the analyst to enter a response to the first query;
an evidence field for the analyst to cite a citation in the one or more transactional documents that supports the response to the first query;
a suggestion field suggesting one or more places in the one or more transactional documents for the analyst to review to determine the response to the first query; and
a next query selection button that, when activated by the analyst, causes a second query display to be displayed to the analyst, wherein the query for the second query display depends on the response by the analyst to the first query, and
the second query display comprises:
means for the analyst to enter a response to the second query;
the evidence field for the analyst to cite a citation in the one or more transactional documents that supports the response to the second query;
the suggestion field suggesting one or more places in the one or more transactional documents for the analyst to review to determine the response to the second query; and
the next query selection button that, when activated by the analyst, causes a third query display to be displayed to the analyst,
wherein the back-end computer system is configured to compute and display an overall risk score for the first transaction based on the analyst's responses to queries in the query node tree, and wherein the back-end computer system further comprises a document scoring module for identifying the one or more places in the one or more transactional documents for the first transaction to display in the suggestion fields for first and second query nodes, wherein the document scoring module comprises:
a document similarity comparison module for:
comparing a first passage responsive to the first query node of the one or more transactional documents for the first transaction to a first passage in a second transactional document stored in the transaction document database for a second transaction, wherein the first passage in the second transactional document is responsive to the first query node for the second transaction and the second transaction is a similar type of transaction to the first transaction, to identify the one or more places in the one or more transactional documents to display in the suggestion field for the first query node; and
comparing a second passage responsive to the second query node of the one or more transactional documents to a second passage in the second transactional document, wherein the second passage in the second transactional document is responsive to the second query node for the second transaction, to identify the one or more places in the one or more transactional documents for the first transaction to display in the suggestion field for the second query node,
wherein:
the document similarity comparison module uses cosine similarity scores to compare the passages of the first and second transactional documents;
the query database stores citations of the analyst in the evidence fields for the first and second query nodes to store as possible suggestions in the suggestion fields for the first and second query nodes for assessing transaction risk of a future transaction with the computer system;
the means for the analyst to enter the response to the first query provides a suggested response to the first query upon a determination by the document similarity comparison module that a similarity score for the first passage responsive to the first query node of the one or more transactional documents for the first transaction to the first passage in the second transactional document exceeds a threshold similarly score for the first query node; and
the means for the analyst to enter the response to the second query provides a suggested response to the second query upon a determination by the document similarity comparison module that a similarity score for the second passage responsive to the second query node of the one or more transactional documents for the first transaction to the second passage in the second transactional document exceeds a threshold similarly score for the second query node.