Journal
NEUROCOMPUTING
Volume 116, Issue -, Pages 382-391Publisher
ELSEVIER
DOI: 10.1016/j.neucom.2011.12.057
Keywords
Ontology; Knowledge retrieval; Latent semantic analysis; Document clustering
Categories
Ask authors/readers for more resources
The semantic retrieval of the engineering domain knowledge is critical in many engineering activities, e.g., product design and process planning. To address the problems with existing keyword-based and semantic-enable methods, we propose an ontology-based semantic retrieval scheme for knowledge search and retrieval from domain documents. In our scheme, domain ontology is first constructed using the graph-based approach to automating construction of domain ontology GRAONTO proposed by our group, and query semantic extension and retrieval are then adopted for semantic-based knowledge retrieval. For query semantic extension, latent semantic analysis is adopted to discover the latent semantic relationships between queries and ontology semantic features, and ontology semantic graph is used to represent the query. For semantic retrieval, a graph-based k-means method is proposed to partition the domain documents into several clusters, and a hierarchical searching strategy is employed for document retrieval. Finally, experimental results on the fixture design corpus verify the benefits of the proposed scheme. (C) 2012 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available