Hierarchical entity recognition and semantic modeling framework for information extraction

At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to perform the following operations:

• receiving a configuration file, wherein the configuration file includes one or more custom entity definitions, and wherein each custom entity definition includes an entity recognition definition;
• generating an entity graph based on the custom entity definitions, wherein the entity graph defines relationships between the custom entity definitions;
• identifying a document of interest;
• parsing the document of interest utilizing the entity graph, wherein parsing the document of interest includes:
• identifying a plurality of tokens in the document of interest,
• determining, for each of the tokens, a document entity, wherein the document entity matches one of the custom entity definitions, and
• generating, based on the document entities, the custom entity definitions, the relationships between the custom entity definitions, and the document of interest, a document tree, wherein the document tree includes the document entities and
• relationships between the document entities;
• and
• providing the document tree to a user.

A system comprising one or more processors and memory operably coupled with the one or more processors, wherein the memory stores instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to perform the following operations:

• receiving a configuration file, wherein the configuration file includes one or more custom entity definitions, and wherein each custom entity definition includes an entity recognition definition;
• generating an entity graph based on the custom entity definitions, wherein the entity graph defines relationships between the custom entity definitions;
• identifying a document of interest;
• parsing the document of interest utilizing the entity graph, wherein parsing the document of interest includes:
• identifying a plurality of tokens in the document of interest,
• determining, for each of the tokens, a document entity, wherein the document entity matches one of the custom entity definitions, and
• generating, based on the document entities, the custom entity definitions, the relationships between the custom entity definitions, and the document of interest, a document tree, wherein the document tree includes the document entities and relationships between the document entities;
• and providing the document tree to a user.