Lexical Semantic Analysis to support Ontology Maintenance Modeling of FMEA

Abstract

Business produces procedures written in natural language that are meant to store technical and engineering information, management decision and operation experience during production system life cycle. A maintenance procedure composes of consecutive logical arguments to determine step-by-step cause-effect events resulting in mitigative tasks. Therefore, the context meaning representation to mimic the purpose of a maintenance procedure is highly dependent upon word sense, syntax-semantic interface, and sematic features of argument. This paper proposes an event-based ontology approach to support failure-Mode-Effect Analysis (FMEA) using lexical semantic analysis and context meaning. At first, it maps argument structure into causal event structure in which event is a group of highly frequent contextual features or words logically linked together to shape structured arguments. Then, Dowty and Van Valin's decomposition model are employed into the format of [Event-State-Activity- Accomplishment -Result] to determine syntax-sematic interface and linking rules in causal chain. Also, we employ Van Valin's model to differentiate between active and causative accomplishments for punctual/non-punctual change of states in causal chain. Finally, metadata or hypernym of causal event is represented to accommodate ontology modeling for semantic extraction and cause-effect interpretation.