Komenti

Komenti enables semantic text mining and ontology-driven information extraction from biomedical text by querying both explicit and inferred knowledge in biomedical ontologies.

Key Features:

• Reasoner-Enabled Semantic Query: Integrates a reasoner to query inferred as well as explicit knowledge from biomedical ontologies over text.
• Information Extraction Framework: Performs information extraction from textual sources by leveraging the semantic richness of biomedical ontologies.
• Vocabulary Construction and Context Disambiguation: Provides components for constructing vocabularies and disambiguating context within texts to improve term interpretation and categorization.
• Ontology-Based Analysis Tasks: Supports ontology-based analyses that identify relationships between concepts via logical reasoning beyond explicit textual statements.

Scientific Applications:

• Biomedical research: Enables ontology-driven characterization and analysis of biomedical literature and clinical records.
• Clinical audits and medication extraction: Automates clinical audits to extract medication data from patient records and medical text, including identifying medications prescribed to patients with hypertrophic cardiomyopathy.
• Patient cohort identification and risk detection: Identifies sub-cohorts such as patients with atrial fibrillation who are not receiving anticoagulation therapy and may be at higher stroke risk.

Methodology:

Integrates semantic reasoning with text mining using biomedical ontologies and employs algorithms for vocabulary construction and context disambiguation.