Review of biomedical knowledge and data representation with conceptual graphs

Area of Science:

• Knowledge representation and reasoning
• Computational linguistics
• Biomedical informatics

Background:

• Conceptual graphs (CG) theory provides a formal framework for knowledge representation.
• CGs are built upon ontologies of concept types and semantic relationships.
• This formalism separates knowledge into conceptual and domain-dependent levels for reusability.

Purpose of the Study:

• To present applications of conceptual graphs theory in the biomedical domain.
• To explore CGs for concept representation, classification, information retrieval, and natural language processing.
• To discuss the unifying role of CGs in knowledge-based systems.

Main Methods:

• Utilizing an ontology of concept types to define canonical conceptual graphs.
• Applying formation rules to derive new conceptual graphs from existing ones.
• Demonstrating CG applications in specific areas like biomedical data and NLP.

Main Results:

• Conceptual graphs theory offers a structured method for representing complex information.
• The formalism facilitates knowledge sharing and reuse across different domains.
• Applications in biomedical data and NLP demonstrate the versatility of CGs.

Conclusions:

• Conceptual graphs theory provides a robust and unifying framework for knowledge-based systems.
• Its application in biomedical informatics and natural language processing highlights its practical value.
• The separation of conceptual and domain-specific knowledge enhances representation reusability.