Intelligent Database Systems Lab Presenter: YU-TING LU Authors: Yong-Bin Kang, Pari Delir Haghighi, Frada Burstein ESA CFinder: An intelligent key concept finder from text for ontology development
Intelligent Database Systems Lab Outlines Motivation Objectives Methodology Experiments Conclusions Comments
Intelligent Database Systems Lab Motivation According to prior studies, concepts can be often described by noun phrases that are suitable for representing the key information within text documents. The main problem under this scheme is that not all the noun phrases can be considered as domain- specific concepts and useful for accurately conceptualizing domain knowledge.
Intelligent Database Systems Lab Objectives We propose a novel key concept finder named CFinder that combines NLP techniques, statistical knowledge and domain-specific knowledge from a corpus of documents in the target domain. The calculated weights are further enhanced by considering an inner structural pattern of the candidates.
Intelligent Database Systems Lab Methodology
Intelligent Database Systems Lab Methodology-Key concept candidate extraction Noun phrase extraction using POS tags Synonym finding and stopword removing Candidate enrichment EMS Emergency Medical Service
Intelligent Database Systems Lab Methodology - Weight calculation of key concept candidates Statistical knowledge Domain- specific knowledge
Intelligent Database Systems Lab Methodology-Key concept extraction c: candidate(key phrase) D: corpus d: document
Intelligent Database Systems Lab Experiments - The DO4MG ontology
Intelligent Database Systems Lab Experiments - Evaluation metrics
Intelligent Database Systems Lab Experiments
Intelligent Database Systems Lab Experiments
Intelligent Database Systems Lab Experiments
Intelligent Database Systems Lab Experiments
Intelligent Database Systems Lab Conclusions CFinder has a strong ability to improve the effectiveness of key concept extraction. CFinder can be built and combined to estimate their degrees of relevance in the target domain, and how the inner structural patterns of the candidates were further enhanced to identify key concepts within the knowledge sources.
Intelligent Database Systems Lab Comments Advantages - Improve the effectiveness of key concept extraction. - It’s not restricted to any domains or applications Applications - Text categorization - Text summarization - Information retrieval