Automatic contract risk assessment based on sentence level risk criterion using machine learning

1. A computer-implemented method comprising:

training a machine learning (ML) model to perform natural language processing to classify one or more portions of contractual documents as standard or non-standard contractual clauses, using training data;

mapping, via a risk assessment engine using one or more computer processors, one or more sentences in a first contractual document to one or more risk categories associated with a type of risk posed by the first contractual document, the one or more risk categories relating to a risk to a party to the first contractual document;

identifying, via a classification engine using at least one of the one or more computer processors, risk-associated language of the one or more sentences based on the one or more risk categories, comprising:

identifying a non-standard contractual clause in the first contractual document using the trained ML model, based on determining that the non-standard contractual clause comprises language that is not standard for a contractual document; and

removing standard language of the one or more sentences based on the one or more risk categories;

mapping, via the risk assessment engine using the one or more computer processors, the risk-associated language of the one or more sentences to one or more risk criterion of a risk criterion document;

generating, via the risk assessment engine using the one or more computer processors, a first risk assessment based on the one or more risk criterion of the risk criterion document; and

delivering the first risk assessment for presentation to a user using a user interface.

2. The method of claim 1 , wherein the one or more risk categories are based on the risk criterion document.

3. The method of claim 1 , further comprising:

determining that the first risk assessment is indeterminate of risk;

combining data from at least one of i) language in the first contractual document excluding the one or more sentences, and ii) storage elements comprising information not included in the first contractual document; and

generating a second risk assessment based on the data.

4. The method of claim 3 , wherein the data is combined based on metadata associated with the risk criterion document.

5. The method of claim 4 , further comprising:

determining at least one of i) the first risk assessment indicates presence of a risk in the first contractual document, ii) the second risk assessment indicates presence of a risk in the first contractual document, and iii) the second risk assessment is indeterminate of risk, and

dispatching the first contractual document and the first or second risk assessment to a computer or device.

6. A system for assessing risk, comprising a risk assessment engine configured to use one or more computer processors to:

train a machine learning (ML) model to perform natural language processing to classify one or more portions of contractual documents as standard or non-standard contractual clauses, using training data;

map, by the one or more computer processors, one or more sentences in a first contractual document to one or more risk categories associated with a type of risk posed by the first contractual document, the one or more risk categories relating to a risk to a party to the first contractual document;

identify, by the one or more computer processors, risk-associated language of the one or more sentences based on the one or more risk categories, comprising:

identifying a non-standard contractual clause in the first contractual document using the trained ML model, based on determining that the non-standard contractual clause comprises language that is not standard for a contractual document;

map, by the one or more computer processors, the risk-associated language of the one or more sentences to one or more risk criterion of a risk criterion document;

generate, by the one or more computer processors, a first risk assessment based on the one or more risk criterion of the risk criterion document;

determine, by the one or more computer processors, that the first risk assessment is indeterminate of risk;

combine, by the one or more computer processors, data from at least one of i) language in the first contractual document excluding the one or more sentences, and ii) storage elements comprising information not included in the first contractual document;

generate, by the one or more computer processors, a second risk assessment based on the data; and

deliver, by the one or more computer processors, the second risk assessment for presentation to a user using a user interface.

7. The system of claim 6 , wherein the one or more risk categories are based on the risk criterion document.

8. The system of claim 6 , wherein the risk-associated language of the one or more sentences is identified by removing standard language of the one or more sentences based on the one or more risk categories.

9. The system of claim 6 , wherein the data is combined based on metadata associated with the risk criterion document.

10. The system of claim 9 , wherein the risk assessment engine is further configured to:

determine at least one of one of i) the first risk assessment indicates presence of a risk in the first contractual document, ii) the second risk assessment indicates presence of a risk in the first contractual document, and iii) the second risk assessment is indeterminate of risk, and

dispatch the first contractual document and the first or second risk assessment to a computer or device.

11. A computer-readable storage medium including computer program code that, when executed on one or more computer processors, performs an operation configured to:

train a machine learning (ML) model to perform natural language processing to classify one or more portions of contractual documents as standard or non-standard contractual clauses, using training data;

map one or more sentences in a first contractual document to one or more risk categories associated with a type of risk posed by the first contractual document, the one or more risk categories relating to a risk to a party to the first contractual document;

identify risk-associated language of the one or more sentences based on the one or more risk categories, comprising:

identifying a non-standard contractual clause in the first contractual document using the trained ML model, based on determining that the non-standard contractual clause comprises language that is not standard for a contractual document; and

removing standard language of the one or more sentences based on the one or more risk categories;

map the risk-associated language of the one or more sentences to one or more risk criterion of a risk criterion document;

generate a first risk assessment based on the one or more risk criterion of the risk criterion document; and

deliver the first risk assessment for presentation to a user using a user interface.

12. The computer-readable storage medium of claim 11 , wherein the one or more risk categories are based on the risk criterion document.

13. The computer-readable storage medium of claim 11 , the operation further configured to:

determine that the first risk assessment is indeterminate of risk;

combine data from at least one of i) language in the first contractual document excluding the one or more sentences, and ii) storage elements comprising information not included in the first contractual document; and

generate a second risk assessment based on the data.

14. The computer-readable storage medium of claim 13 , the operation further configured to:

determine at least one of one of i) the first risk assessment indicates presence of a risk in the first contractual document, ii) the second risk assessment indicates presence of a risk in the first contractual document, and iii) the second risk assessment is indeterminate of risk, and

dispatch the first contractual document and the first or second risk assessment to a computer or device.