Systems and methods for data classification and governance

A system for data classification and governance comprising at least one computer including a processor, wherein the processor is configured to:

• retrieve data related to a first-level entity, wherein the first-level entity comprises at least one or more of an internal database, a stored dataset, and an internal platform, wherein the retrieved data includes internal user-activity data for the first-level entity including one or more of an employee query, an access report, an internal monitoring information, a download report, and a human resources (HR) employee hierarchy, and wherein the first-level entity is associated with a governance policy that is stored centrally within an organization and is programmatically accessed, wherein the governance policy is associated with one or more of:
• a data source, a storage location, a logical dataset, the HR employee hierarchy, and a line of business;
• process the data with one or more machine learning models, wherein the one or more machine learning models are configured to:
• add labels to the data;
• add classifications to the data based on the labels;
• generate nodes and edges in a graph based on the governance policy and known user interactions with the data across a usage lifecycle, wherein the nodes represent entities including databases, employees, and managers, and wherein the edges represent data sharing relationships, wherein the graph is dynamically updated based on real-time data inputs, monitored changes in governance policies, and includes a hierarchical structure that prioritizes sensitive data flows based on organizational roles and permissions, and further comprising a mechanism to enforce data retention policies based on predefined organizational rules stored in the governance policy, wherein the hierarchical structure is dynamically adjusted based on sensitivity scores assigned to data flows;
• resolve ambiguous nodes into specific nodes;
• predict edge probabilities in the graph;
• predict a flow of data through the graph based on the edge probabilities;
• process and parse a text of the governance policy related to identified data with natural language processing (“NLP”), wherein the governance policy is written in free text format, wherein the parsing identifies one or more specific word patterns that represent one or more of collected data, a declared usage, and a sharing policy;
• insert a data probe at an entity in the graph;
• identify, based on the data probes, the entities that receive the data;
• monitor injected data from the data probe into the graph as the injected data traverses the graph, wherein the monitoring includes tracking data flow anomalies and identifying unauthorized data sharing events;
• generate automated messages to one or more employees that are likely to receive the data based on the entity identification that request the one or more employees delete the data when an employee work task is completed, wherein the automated messages are sent preemptively before the data may be shared with the one or more employees;
• sending the requests preemptively, that is, before the data may be shared with the entity;
• identify any differences between what the governance policy specifies and what the one or more employee are requesting and what they have used in a past;
• and
• when the one of the one or more employees is requesting or sharing the data that is outside the permissions of the governance policy, denying the request and blocking access to the data based on the governance policy.

A non-transitory computer readable storage medium, including instructions stored thereon for data classification and governance, which instructions, when read and executed by one or more computer processors, cause the one or more computer processors to perform steps comprising:

• retrieving data related to a first-level entity, wherein the first-level entity comprises at least one or more of an internal database, a stored dataset, and an internal platform, wherein the retrieved data includes internal user-activity data for the first-level entity including one or more of an employee query, an access report, an internal monitoring information, a download report, and a human resources (HR) employee hierarchy, and wherein the first-level entity is associated with a governance policy that is stored centrally within an organization and is programmatically accessed, wherein the governance policy is associated with one or more of:
• a data source, a storage location, a logical dataset, the HR employee hierarchy, and a line of business;
• processing the data with one or more machine learning models, wherein the processing includes:
• adding labels to the data;
• adding classifications to the data based on the labels;
• generating nodes and edges in a graph based on the governance policy and known user interactions with the data across a usage lifecycle, wherein the nodes represent entities including databases, employees, and managers, and wherein the edges represent data sharing relationships, wherein the graph is dynamically updated based on real-time data inputs, monitored changes in governance policies, and includes a hierarchical structure that prioritizes sensitive data flows based on organizational roles and permissions, and further comprising a mechanism to enforce data retention policies based on predefined organizational rules stored in the governance policy, wherein the hierarchical structure is dynamically adjusted based on sensitivity scores assigned to data flows;
• resolving ambiguous nodes into specific nodes;
• predicting edge probabilities in the graph;
• predicting a flow of data through the graph based on the edge probabilities;
• processing and parsing a text of the governance policy related to identified data with natural language processing (“NLP”), wherein the governance policy is written in free text format, wherein the parsing identifies one or more specific word patterns that represent one or more of collected data, a declared usage, and a sharing policy;
• inserting a data probe at an entity in the graph;
• identifying, based on the data probes, the entities that receive the data;
• monitoring injected data from the data probe into the graph as the injected data traverses the graph, wherein the monitoring includes tracking data flow anomalies and identifying unauthorized data sharing events;
• generating automated messages to one or more employees that are likely to receive the data based on the entity identification that request the one or more employees delete the data when an employee work task is completed, wherein the automated messages are sent preemptively before the data may be shared with the one or more employees;
• sending the requests preemptively, that is, before the data may be shared with the entity;
• identifying any differences between what the governance policy specifies and what the one or more employee are requesting and what they have used in a past;
• and
• when the one of the one or more employees is requesting or sharing the data that is outside the permissions of the governance policy, denying the request and blocking access to the data based on the governance policy.