Identifying correlated roles using a system driven by a neural network

A device, comprising:

• one or more memories;
• and
• one or more processors, operatively coupled to the one or more memories, to:
• receive data that identifies a first standardized set of titles and a first hierarchy of role classifications;
• receive data that identifies a first set of historical titles associated with a group of organizations;
• generate data that identifies a second standardized set of titles by using a system of adversarial neural networks to process the data that identifies the first standardized set of titles and the first hierarchy of role classifications;
• generate data that identifies a second set of historical titles by using the system of adversarial neural networks to process the data that identifies the first set of historical titles;
• generate data that identifies particular title-class identifiers by using the system of adversarial neural networks to process the data that identifies the second standardized set of titles and the data that identifies the second set of historical titles;
• receive a data model that includes a set of generative adversarial networks, including the system of adversarial neural networks, that have been trained to map titles to:
• a standardized set of titles for standardized roles within organizations that are identified, by the set of generative adversarial networks, based on historical titles associated with the organizations, and
• a hierarchy of role classifications that identify classes used to group the standardized roles;
• receive a request associated with standardizing a set of organization-specific roles within an organization of the group of organizations,
• wherein the request includes information comprising data that identifies a storage location for a set of titles for the set of organization-specific roles;
• obtain data that identifies the set of titles using the data that identifies the storage location;
• convert the data that identifies the set of titles to a set of vectors that represent semantic meanings of the set of titles;
• determine, by using the data model to process the set of vectors, a set of scores that indicate likelihoods of the set of titles mapping to a set of title-class identifiers, including the particular title-class identifiers, associated with the historical titles,
• wherein the set of title-class identifiers are to be used to associate the standardized set of titles to the hierarchy of role classifications that identify types of the standardized roles within the organizations, and
• wherein the set of scores are to be determined using a cost function of the data model that assigns weighted values to the set of title-class identifiers based on the information included in the request;
• identify, based on the set of scores, a subset of title-class identifiers, of the set of title-class identifiers, that associate particular titles, of the standardized set of titles, and particular role classifications that are part of a standardized hierarchy of role classifications;
• obtain information relating to particular standardized roles that is stored in association with the subset of title-class identifiers;
• provide the information relating to the particular standardized roles to a client device by generating a recommendation to modify or eliminate one or more tasks associated with the set of organization-specific roles;
• and
• perform one or more actions to cause one or more tasks associated with the set of organization-specific roles to be modified or eliminated based on the information relating to the particular standardized roles.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors of a device, cause the one or more processors to:
• receive data that identifies a first standardized set of titles and a first hierarchy of role classifications;
• receive data that identifies a first set of historical titles associated with a group of organizations;
• generate data that identifies a second standardized set of titles by using a system of adversarial neural networks to process the data that identifies the first standardized set of titles and the first hierarchy of role classifications;
• generate data that identifies a second set of historical titles by using the system of adversarial neural networks to process the data that identifies the first set of historical titles;
• generate data that identifies particular title-class identifiers by using the system of adversarial neural networks to process the data that identifies the second standardized set of titles and the data that identifies the second set of historical titles;
• receive a request associated with standardizing a set of organization-specific roles within an organization of the group of organizations,
• wherein the request includes information comprising:
• data that identifies a set of titles for the set of organization-specific roles, and
• an industry identifier of an industry in which the organization operates;
• convert the data that identifies the set of titles to a set of vectors that represent semantic meanings of the set of titles;
• determine, by using a data model to process the set of vectors, a set of scores that indicate likelihoods of the set of titles mapping to a set of title-class identifiers including the particular title-class identifiers,
• wherein the set of title-class identifiers are to be used to associate a standardized set of titles to a hierarchy of role classifications that identify classes used to group standardized roles within organizations, and
• wherein the set of scores are to be determined using a cost function of the data model that assigns weighted values to the set of title-class identifiers based on the information included in the request;
• identify, based on the set of scores, a subset of title-class identifiers, of the set of title-class identifiers, that associate particular titles, of the standardized set of titles, and particular role classifications that are part of a standardized hierarchy of role classifications;
• obtain information relating to particular standardized roles that is stored in association with the subset of title-class identifiers;
• provide the information relating to the particular standardized roles to a client device by generating a recommendation to modify or eliminate one or more tasks associated with the set of organization-specific roles;
• and
• perform one or more actions to cause one or more tasks associated with the set of organization-specific roles to be modified or eliminated based on the information relating to the particular standardized roles.