Coordinate descent based ontology sparse vector computing strategy and its applications


In recent years, as a semantic analysis and computational tool, ontology has been widely applied in many engineering applications. Many cases suggests that it’s confronted with countless big data source with the complex data structures. In order to relieve the dilemma, the sparse learning algorithms are introduced into the ontology similarity measuring and ontology mapping. In this setting, it should be a high dimensional expression of each ontology vertex, and the ontology algorithm should extract key component information effectively. Under such background, we consider the ontology sparse vector learning algorithm and application in different engineering applications. In this article, by means of coordinate descent minimization tricks, we present the ontology sparse vector optimization strategy and discuss the different transformation in different settings. At last, the new ontology sparse vector learning proceeding is applied to four engineering applications respectively to get its efficiency verified.

DOI: 10.1007/s10586-017-1283-8

7 Figures and Tables

Cite this paper

@article{Gao2017CoordinateDB, title={Coordinate descent based ontology sparse vector computing strategy and its applications}, author={Wei Gao and Muhammad Shoaib Sardar and Sohail Zafar and Zohaib Zahid}, journal={Cluster Computing}, year={2017}, pages={1-15} }