Content analyzer and recommendation tool

A computer program product comprising a computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to perform an operation comprising:

• analyzing a plurality of documents, using one or more natural language processing (NLP) models, to identify a first plurality of entities in the plurality of documents;
• processing the plurality of documents with the one or more NLP models to identify a first plurality of relationships among the first plurality of entities;
• storing each of the first plurality of entities in a data store, wherein the data store further includes an indication of the identified first plurality of relationships among the first plurality of entities;
• receiving a draft correspondence;
• determining that the draft correspondence includes a first entity and a second entity, wherein the first entity and the second entity have a first relationship;
• upon determining that the first relationship conflicts with at least one of the first plurality of relationships stored in the data store, generating a suggested revision for the draft correspondence;
• receiving an indication that the suggested revision has been rejected;
• and
• updating the data store based on the rejection.

A system comprising:

• one or more computer processors;
• and
• a memory containing a program which when executed by the one or more computer processors performs an operation, the operation comprising:
• analyzing a plurality of documents, using one or more natural language processing (NLP) models, to identify a first plurality of entities and a first plurality of details in the plurality of documents;
• processing the plurality of documents with the one or more NLP models to identify, for each respective entity in the first plurality of entities, one or more corresponding details in the first plurality of details that describe the respective entity, wherein a first detail is identified as describing a first entity;
• storing each respective entity of the first plurality of entities and the one or more corresponding details describing the respective entity in a data store;
• receiving a draft correspondence;
• determining that the draft correspondence includes a second entity and a second detail describing the second entity;
• determining that the second entity corresponds to the first entity, based on comparing the second entity with each of the first plurality of entities stored in the data store;
• upon determining that the second detail conflicts with the first detail, generating a suggested revision for the draft correspondence, wherein the suggested revision includes the first detail;
• receiving an indication that the suggested revision has been rejected;
• and
• updating the data store based on the rejection.