Regulatory compliance assessment and business risk prediction system
An electronic platform to measure a maturity or level of an entity in view of regulatory and business risks relating to regulatory compliance. The methods and systems can collect various data (e.g., regulatory agency reports, regulatory agency warning letters (e.g. FDA warning letters), internal and vendor company audit results, fines and settlement information, country business risks, regulatory agency product recalls, etc.) from various different data sources. The collected information is analyzed using machine learning techniques to determine a risk compliance level or score for one or more of an entity's companies, functions, control types, and locations arising from regulatory audit non-conformances. The risk compliance scores can be used to generate a risk prediction and identify one or more actions to be taken by the entity to improve or increase an associated compliance level.
1. A method comprising:
collecting, by a processing device, regulatory-related data associated with an entity, wherein the regulatory-related data comprises a plurality of data objects;
determining, by a neural network executed by the processing device, a first set of classification designations corresponding to the plurality of data objects, wherein the neural network generates each of the first set of classification designations based on a combination of function type data corresponding to each of the plurality of data objects, control type data corresponding to each of the plurality of data objects, and findings level data corresponding to each of the plurality of data objects, wherein the neural network is trained based on training data set comprising parsed text of a set of documents having an approved function type to enable the determining of the first set of classification designations;
determining a confidence level associated with a first classification designation of the first set of classification designations;
determining, by a heuristic pattern matching system executed by the processing device, a second set of classification designations corresponding to the plurality of data objects, wherein the heuristic pattern matching system generates each of the second set of classification designations based on a combination of function type data corresponding to each of the plurality of data objects, control type data corresponding to each of the plurality of data objects, and findings level data corresponding to each of the plurality of data objects;
assigning a resultant classification designation to a first data object of the plurality of data objects, wherein the resultant classification designation is determined based on a comparison of the first classification designation, a second classification designation of the second set of classification designations, and the confidence level;
calculating a risk compliance index score associated with the resultant classification designation, wherein the risk compliance index score comprises a compliance level of a set of compliance levels;
generating, based on the risk compliance index score, a recommended action corresponding to compliance by the entity with regulatory guidelines; and
generating a graphical user interface comprising a display of the risk compliance index score and the recommended action.
2. The method of claim 1 , further comprising:
iteratively collecting updated regulatory-related data associated with the entity;
determining, by the neural network and the heuristic pattern matching system, updated resultant classification designations; and
calculating an updated risk compliance index score associated with the updated resultant classification designations.
3. The method of claim 1 , wherein the regulatory-related data is collected from a plurality of different data sources.
4. A system comprising:
a processing device; and
a memory to store computer-executable instructions that, if executed, cause the processing device to perform operations comprising:
collecting, by a processing device, regulatory-related data associated with an entity, wherein the regulatory-related data comprises a plurality of data objects;
determining, by a neural network executed by the processing device, a first set of classification designations corresponding to the plurality of data objects, wherein the neural network generates each of the first set of classification designations based on a combination of: function type data corresponding to each of the plurality of data objects, control type data corresponding to each of the plurality of data objects, and findings level data corresponding to each of the plurality of data objects, wherein the neural network is trained based on training data comprising parsed text of a set of documents having an approved function type to enable the determining of the first set of classification designations;
determining a confidence level associated with a first classification designation of the first set of classification designations;
determining, by a heuristic pattern matching system executed by the processing device, a second set of classification designations corresponding to the plurality of data objects, wherein the heuristic pattern matching system generates each of the second set of classification designations based on a combination of function type data corresponding to each of the plurality of data objects, control type data corresponding to each of the plurality of data objects, and findings level data corresponding to each of the plurality of data objects;
assigning a resultant classification designation to a first data object of the plurality of data objects, wherein the resultant classification designation is determined based on a comparison of the first classification designation, a second classification designation of the second set of classification designations, and the confidence level;
calculating a risk compliance index score associated with the resultant classification designation, wherein the risk compliance index score comprises a compliance level of a set of compliance levels;
generating, based on the risk compliance index score, a recommended action corresponding to compliance by the entity with regulatory guidelines; and
generating a graphical user interface comprising a display of the risk compliance index score and the recommended action.
5. The system of claim 4 , wherein the operations further comprise:
reviewing an association between the first data object and the resultant classification designation to update the association.
6. The system of claim 4 , wherein the operations further comprise:
parsing the first data object of the plurality of data objects to determine a set of relevant text associated with first findings level data.
7. The system of claim 4 , wherein the operations further comprise:
determining a first classification designation associated with the first data object exceeds a threshold prediction reliability level.
8. The system of claim 7 , wherein the operations further comprise:
assigning the first data object to the first classification designation.
9. A non-transitory computer-readable storage device storing computer-executable instructions that, if executed by a processing device, cause the processing device to perform operations comprising:
collecting regulatory-related data associated with an entity, wherein the regulatory-related data comprises a plurality of data objects;
determining, by a neural network executed by the processing device, a first set of classification designations corresponding to the plurality of data objects, wherein the neural network generates each of the first set of classification designations based on a combination of: function type data corresponding to each of the plurality of data objects, control type data corresponding to each of the plurality of data objects, and findings level data corresponding to each of the plurality of data objects, wherein the neural network is trained based on training data comprising parsed text of a set of documents having an approved function type to enable the determining of the first set of classification designations;
determining a confidence level associated with a first classification designation of the first set of classification designations;
determining, by a heuristic pattern matching system executed by the processing device, a second set of classification designations corresponding to the plurality of data objects, wherein the heuristic pattern matching system generates each of the second set of classification designations based on a combination of function type data corresponding to each of the plurality of data objects, control type data corresponding to each of the plurality of data objects, and findings level data corresponding to each of the plurality of data objects;
assigning a resultant classification designation to a first data object of the plurality of data objects, wherein the resultant classification designation is determined based on a comparison of the first classification designation, a second classification designation of the second set of classification designations, and the confidence level;
calculating a risk compliance index score associated with the resultant classification designation, wherein the risk compliance index score comprises a compliance level of a set of compliance levels;
generating, based on the risk compliance index score, a recommended action corresponding to compliance by the entity with regulatory guidelines; and
generating a graphical user interface comprising a display of the risk compliance index score and the recommended action.
10. The non-transitory computer-readable storage device of claim 9 , the operations further comprising reviewing an association between the first data object and the resultant classification designation to update the association.
11. The non-transitory computer-readable storage device of claim 9 , the operations further comprising:
determining a first classification designation associated with the first data object exceeds a threshold prediction reliability level; and
assigning the first data object to the first classification designation.
12. The non-transitory computer-readable storage device of claim 9 , the operations further comprising:
calculating a plurality of risk compliance index scores each corresponding to a data object of the plurality of data objects, wherein the plurality of risk compliance index scores comprises a first risk compliance index score associated with the first data object based at least in part on a first resultant classification designation of the first data object; and
determining an overall risk compliance index score associated with the entity based on the plurality of risk compliance index scores corresponding to the plurality of data objects.