Work place: Geomatics Department, Nanjing Forestry University, Nanjing, China
E-mail: yufeng788@163.com
Website:
Research Interests: Pattern Recognition, Information Theory, Computational Complexity Theory
Biography
Yufeng Shi, born in Yantai, Shandong Province, China, in March 1965. He received the Master’s degree in Surveying Engineering from Hefei University of Technology in 1997. In 2003, he received the Doctor of Engineering degree in Geodesy and Surveying Engineering from Shandong University of Science and Technology in Taian, China. From 2003 to 2005, he did post-doctoral research work in Wuhan University, China.
He worked in Shandong Institute of Building Material from 1985 to 1998 as a lecturer and Associate Professor, from 2003 to 2008, he worked in Shandong University of Technology as a Professor, and he worked in Hongkong Polytechnic University as a research fellow in 2004. Now, he works in Nanjing Forestry University. He has published over 30 first-author technical papers and taken part in about 20 projects. His personal research domain and interests are information pattern recognition theory and application, geomatics information processing theory and application, etc. His previous publications include: Numeric Information Pattern Recognition Theory and Application (Beijing, China: Science Press, 2007), A Total Entropy Model of Spatial Data Uncertainty (England, Journal of Information Science, Vol.32(3), 2006), etc.
Professor Shi is the member of the Computer Application Branch Committee of Chinese Academy of Forestry, and awarded the first prize for scientific and technological progress of Shandong University of Technology in 2005.
DOI: https://doi.org/10.5815/ijisa.2009.01.08, Pub. Date: 8 Oct. 2009
Based on fuzzy similarity degree, entropy, relative entropy and fuzzy entropy, the symmetric fuzzy relative entropy is presented, which not only has a full physical meaning, but also has succinct practicability. The symmetric fuzzy relative entropy can be used to measure the divergence between different fuzzy patterns. The example demonstrates that the symmetric fuzzy relative entropy is valid and reliable for fuzzy pattern recognition and classification, and its classification precision is very high.
[...] Read more.Subscribe to receive issue release notifications and newsletters from MECS Press journals