US 11,049,042 B2Granted

Systems and methods for extracting specific data from documents using machine learning

AssigneeConvr Inc.

InventorHarish Neelamana

Life of this patent

Estimating the term…

Loading patent…

Abstract

Computer implemented systems and methods are disclosed for extracting specific data using machine learning algorithms. In accordance with some embodiments, a memory device that stores at least a set of computer executable instructions for a machine learning algorithm and a pre-fill engine; and at least one processor that executes the instructions that cause the pre-fill engine to perform functions that include: receiving electronic documents, seed dataset documents, and pre-fill questions; determining output data that enable navigation through the electronic documents using the machine learning algorithm; determining output questions that enable navigation through the electronic documents using the machine learning algorithm; determining output documents to enable navigation through the electronic documents using the machine learning algorithm; and presenting one or more answers for one or more of the output questions using a graphical user interface.

Claims14

Dependent claims

A system comprising:

a memory device that stores at least a set of computer executable instructions for a machine learning algorithm and a pre-fill engine, wherein the pre-fill engine is trained via a performance monitor that provides user feedback to the machine learning algorithm, and wherein the user feedback includes user verification of the output data, the output documents, and/or the one or more answers provided by the pre-fill engine;
and
at least one processor that executes the instructions that cause the pre-fill engine to perform functions that include:
receiving electronic documents, seed dataset documents, and pre-fill questions, wherein the electronic documents comprise a set of documents to be processed by the pre-fill engine, wherein the seed dataset documents comprise a set of documents providing a reference for determining output documents by the pre-fill engine, and wherein the pre-fill questions comprise a set of pre-fill questions to be answered using the pre-fill engine;
determining output data from the electronic documents and data references that enable navigation through the electronic documents using the machine learning algorithm, wherein the output data comprise data relevant to a particular field of analysis, and wherein the data references rely upon terminology, categories, classifications and ontology of the particular field of analysis;
determining output questions from the set of pre-fill questions and the data references that enable navigation through the electronic documents using the machine learning algorithm, wherein the output questions are relevant to the particular field of analysis;
determining output documents from the seed dataset documents and the data references to enable navigation through the electronic documents using the machine learning algorithm, wherein the output documents are relevant to the particular field of analysis according to the data references;
and
presenting one or more answers for one or more of the output questions using a graphical user interface, wherein at least part of the output data and at least one of the electronic documents are displayed side by side on one screen in the graphical user interface for verification, wherein a score is displayed for an entity being evaluated, and wherein the score is determined based on the output data and answers to the output questions.

6 more elements

A method, comprising:

receiving electronic documents, seed dataset documents, and pre-fill questions, wherein the electronic documents comprise a set of documents to be processed by a pre-fill engine of a computing device, wherein the seed dataset documents comprise a set of documents providing reference for determining output documents by the pre-fill engine, and wherein the pre-fill questions comprise a set of pre-fill questions to be answered using the pre-fill engine;
determining output data from the electronic documents and data references to enable navigation through the electronic documents using a machine learning algorithm executed by the computing device, wherein the output data comprise data relevant to a particular field of analysis, wherein the data references rely upon terminology, categories, classifications and ontology of the particular field of analysis, wherein the pre-fill engine is trained via a performance monitor that provides user feedback to the machine learning algorithm, and wherein the user feedback includes user verification of the output data, the output documents, and/or the one or more answers provided by the pre-fill engine;
determining output questions from the set of pre-fill questions and the data references to enable navigation through the electronic documents using the machine learning algorithm, wherein the output questions are relevant to the particular field of analysis;
determining output documents from the seed dataset documents and the data references to enable navigation through the electronic documents using the machine learning algorithm, wherein the output documents are relevant to the particular field of analysis based on the data references;
and
presenting one or more answers for one or more of the output questions using a graphical user interface, wherein at least part of the output data and at least one of the electronic documents are displayed side by side on one screen in the graphical user interface for verification, wherein a score is displayed for an entity being evaluated, and wherein the score is determined based on the output data and answers to the output questions.

3 more elements

A non-transitory computer-readable medium storing a set of instructions that, when executed by one or more processors of a computing device, cause the computing device to perform functions comprising:

receiving electronic documents, seed dataset documents, and pre-fill questions, wherein the electronic documents are a set of documents to be processed by a pre-fill engine of the computing device, wherein the seed dataset documents comprise a set of documents providing reference for the pre-fill engine, and wherein the pre-fill questions comprise a set of questions to be answered using the pre-fill engine;
determining output data from the electronic documents and data references to enable navigation through the electronic documents using a machine learning algorithm executed by the computing device, wherein the output data comprise data relevant to a particular field of analysis, wherein the data references rely upon terminology, categories, classifications and ontology of the particular field of analysis, wherein the pre-fill engine is trained via a performance monitor that provides user feedback to the machine learning algorithm, and wherein the user feedback includes user verification of the output data, the output documents, and/or the one or more answers provided by the pre-fill engine;
determining output questions from the set of pre-fill questions and data references to enable navigation through the electronic documents using the machine learning algorithm, wherein the output questions are relevant to the particular field of analysis;
determining output documents from the seed dataset documents and data references to enable navigation through the electronic documents using the machine learning algorithm, wherein the output documents are relevant to the particular field of analysis;
and
presenting one or more answers for one or more of the output questions using a graphical user interface, wherein at least part of the output data and at least one of the electronic documents are displayed side by side on one screen in the graphical user interface for verification, wherein a score is displayed for an entity being evaluated, and wherein the score is determined based on the output data and answers to the output questions.

3 more elements

Dependencies are inferred from the claim text.