In AR4IE we build linguistic computational linguistic (CL) models, theories and supporting natural language processing (NLP) tools to automate the extraction of insightful information from text documents. The linguistic models view the text documents as sequences of vectors of linguistic features including lexical, morphological, syntactic and semantic features. The theories define (1) atomic entities that match predicates over linguistic features, (2) relations between the entities and (3) relational entities that are used inductively in (2). The entities and relations form Graph entities and consequently graph analysis techniques apply to infer insightful knowledge and wisdom.
The supporting NLP tools implement the theory definitions with computational models based on logic, expert knowledge and statistics. The main applications at AR4IE are medical and Arabic documents.
Computational models for bacterial-antimicrobial resistance
Molecular Architecture of Spinal Chord Injury with Protein Interaction Networks
A Rich-Club Organization in Brain Ischemia Protein Interaction Network
Domain Specific Annotated Lebanese Electronic Medical Records