Conceptual world representation natural language understanding system and method
A Natural Language Understanding system is provided for indexing of free text documents. The system according to the invention utilizes typographical and functional segmentation of text to identify those portions of free text that carry meaning. The system then uses words and multi-word terms and phrases identified in the free to text to identify concepts in the free text. The system uses a lexicon of terms linked to a formal ontology that is independent of a specific language to extract concepts from the free text based on the words and multi-word terms in the free text. The formal ontology contains both language independent domain knowledge concepts and language dependent linguistic concepts that govern the relationships between concepts and contain the rules about how language works. The system according to the current invention may preferably be used to index medical documents and assign codes from independent coding systems, such as, SNOMED, ICD-9 and ICD-10. The system according to the current invention may also preferably make use of syntactic parsing to improve the efficiency of the method.
1. A physical storage medium storing code representing instructions that when executed by a computer cause the computer to:
produce a hierarchical ontology including a plurality of concepts;
prohibit a link between a first concept from the plurality of concepts and a second concept from the plurality of concepts when a link between a parent concept linked hierarchically with the first concept and a parent concept linked hierarchically with the second concept is prohibited, the first concept being at any level within the hierarchy below a level of the parent concept associated with the first concept, the second concept being at any level within the hierarchy below a level of the parent concept associated with the second concept; and
provide a user with access to the hierarchical ontology.
2. The physical storage medium of claim 1 , wherein every concept from the plurality of concepts included within the hierarchical ontology, and ancestrally linked to and below the parent concept associated with the first concept, is defined based on all criteria from the set of criteria and based on at least one additional criterion different than each criterion from the set of criteria.
3. The physical storage medium of claim 1 , wherein a link between a first term from a document and a second term from the document is prohibited based on the prohibiting associated with the first concept and the second concept.
4. The physical storage medium of claim 1 , further comprising code representing instructions to cause the computer to:
permit a link between a third concept from the plurality of concepts and a fourth concept from the plurality of concepts only when a link between a parent concept linked hierarchically to the third concept and a parent concept linked hierarchically to the fourth concept is permitted.
5. The physical storage medium of claim 1 , wherein the parent concept is associated with a partial definition derived from a set of a criteria used to define the parent concept, the partial definition is used to define a relationship between terms from a document after the terms from the document have been associated with the parent concept.
6. The physical storage medium of claim 1 , wherein the hierarchical ontology is a formal ontology used within a natural language processing algorithm to extract a meaning from a text.
7. The physical storage medium of claim 1 , wherein a syntactically related set of concepts from the plurality of concepts is associated with a full definition.
8. The physical storage medium of claim 1 , wherein the plurality of concepts in the hierarchical ontology is associated with a plurality of terms in a lexicon.
9. A method, comprising:
receiving a file having a text; and
determining a meaning from the text based on a comparison of at least a portion of the text to at least one concept from a plurality of concepts arranged in hierarchical ontology, the plurality of concepts having a primary concept at a highest level of the hierarchical ontology, the primary concept being defined based on a set of criteria,
each concept from the plurality of concepts included within the hierarchical ontology, and ancestrally linked to and below the primary concept, being defined based on all criteria from the set of criteria and based on at least one additional criterion different than each criterion from the set of criteria;
associating a first term from the text with a first concept from the plurality of concepts;
associating a second term from the text with a second concept from the plurality of concepts; and
prohibiting a link between the first term and the second term when a link between the first concept and the second concept is prohibited based on a link between a parent concept of the first concept and a parent concept of the second concept, the link between the first concept and the second concept being a substantially similar type as the link between the parent concept of the first concept and the parent concept of the second concept.
10. The method of claim 9 , wherein the plurality of concepts including a parent concept, a first child concept ancestrally linked to and below the parent concept, and a second child concept ancestrally linked to and below the parent concept, the parent concept being associated with a partial definition derived from a set of a criteria used to define the parent concept,
the method further comprising:
associating a first term from the text with the parent concept; and
determining a relationship between a second term from the text and one of the first child concept or the second child concept based on the partial definition.
11. A method, comprising:
receiving a file having a text;
determining a meaning from the text based on a comparison of at least a portion of the text to at least one concept from a plurality of concepts arranged in hierarchical ontology, the plurality of concepts having a primary concept at a highest level of the hierarchical ontology, the primary concept being defined based on a set of criteria,
each concept from the plurality of concepts included within the hierarchical ontology, and ancestrally linked to and below the primary concept, being defined based on all criteria from the set of criteria and based on at least one additional criterion different than each criterion from the set of criteria;
associating a first term from the text with a first concept from the plurality of concepts;
associating a second term from the text with a second concept from the plurality of concepts; and
prohibiting a link between the first term and the second term when a link between the first concept and the second concept is prohibited based on a link between a parent concept of the first concept and a parent concept of the second concept, the link between the first concept and the second concept being a substantially similar type as the link between the parent concept of the first concept and the parent concept of the second concept, the link between the first concept and second concept, and the link between the parent concept of the first concept and the parent concept of the second concept being ancestrally linked.