Systems and methods for analyzing documents using machine learning techniques

1. A system for entity risk management, the system comprising:

at least one processor;

a display; and

a non-transitory computer-readable medium containing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising:

accessing document data associated with at least one of a transaction or an individual;

normalizing the document data;

classifying the normalized document data;

extracting model input data from the classified document data;

selecting, based on the classified document data, a machine learning model from among a plurality of machine learning models, the machine learning model having been trained with document data of a same document type as the classified document data;

applying the selected machine learning model to the extracted model input data to score the document data to generate a favorability output indicating a favorability of the transaction or individual, wherein the favorability output comprises an amount of risk associated with the transaction or individual;

generating analysis data based on the scored document data;

updating the selected machine learning model by modifying at least one model parameter based on the favorability output;

determine whether the favorability output satisfies an alert criterion; and

if the favorability output satisfies the alert criterion, generate an alert at the display.

2. The system of claim 1 , wherein:

the document data comprises unstructured data;

normalizing the document data comprises converting the unstructured data to structured data; and

classifying the normalized document data comprises identifying at least one marker in the document data.

3. The system of claim 2 , wherein the at least one marker includes a word, a phrase, a frequency of text, a position of text relative to a document, a position of text relative to other text in the document, a sentence, a number, or a pictographic identifier.

4. The system of claim 2 , wherein classifying the normalized document data comprises classifying the normalized document data as associated with a loan based on the at least one marker.

5. The system of claim 1 , wherein the non-transitory computer-readable medium contains further instructions that, when executed by the at least one processor, cause the at least one processor to apply a natural language processing (NLP) method to the classified document data to determine the model input data.

6. The system of claim 1 , wherein:

classifying the normalized document data comprises identifying at least one marker in the document data; and

an additional machine learning model associates the at least one marker with a document type.

7. The system of claim 1 , wherein the machine learning model was trained to generate the favorability output using historical data from at least one of:

a first financial institution associated with the document data; or

a second financial institution associated with additional document data.

8. The system of claim 1 , wherein:

the document data is associated with a financial institution; and

the machine learning model is trained to predict a change in model input data that will improve the favorability output.

9. A method for entity risk management, comprising:

accessing document data associated with at least one of a transaction or an individual;

normalizing the document data;

classifying the normalized document data;

extracting model input data from the classified document data;

selecting, based on the classified document data, a machine learning model from among a plurality of machine learning models, the machine learning model having been trained with document data of a same document type as the classified document data;

applying the selected machine learning model to the extracted model input data to score the document data to generate a favorability output indicating a favorability of the transaction or individual, wherein the favorability output comprises an amount of risk associated with the transaction or individual;

generating analysis data based on the scored document data;

updating the selected machine learning model by modifying at least one model parameter based on the favorability output;

determining whether the favorability output satisfies an alert criterion; and

if the favorability output satisfies the alert criterion, generating an alert.

10. The method of claim 9 , wherein:

the document data comprises unstructured data;

normalizing the document data comprises converting the unstructured data to structured data; and

classifying the normalized document data comprises identifying at least one marker in the first document data.

11. The method of claim 10 , wherein the at least one marker includes a word, a phrase, a frequency of text, a position of text relative to a document, a position of text relative to other text in the document, a sentence, a number, or a pictographic identifier.

12. The method of claim 10 , wherein classifying the normalized document data comprises classifying the normalized document data as associated with a loan based on the at least one marker.

13. The method of claim 9 , further comprising applying a natural language processing (NLP) method to the classified document data to determine the model input data.

14. The method of claim 9 , wherein:

classifying the normalized document data comprises identifying at least one marker in the document data; and

an additional machine learning model associates the at least one marker with a document type.

15. The method of claim 9 , wherein the machine learning model was trained to generate the favorability output using historical data from at least one of:

a first financial institution associated with the document data; or

a second financial institution associated with additional document data.

16. The method of claim 9 , wherein:

the document data is associated with a financial institution; and

the machine learning model is trained to predict a change in model input data that will improve the favorability output.