Y
Yijun Sun
Researcher at University at Buffalo
Publications - 141
Citations - 6362
Yijun Sun is an academic researcher from University at Buffalo. The author has contributed to research in topics: Feature selection & AdaBoost. The author has an hindex of 36, co-authored 130 publications receiving 5500 citations. Previous affiliations of Yijun Sun include University of Texas MD Anderson Cancer Center & State University of New York System.
Papers
More filters
Journal ArticleDOI
Fecal microbiota in premature infants prior to necrotizing enterocolitis.
Volker Mai,Christopher Young,Maria Ukhanova,Xiaoyu Wang,Yijun Sun,George Casella,Douglas W. Theriaque,Nan Li,Renu Sharma,Mark L. Hudak,Josef Neu +10 more
TL;DR: The authors' observations suggest that abnormal patterns of microbiota and potentially a novel pathogen contribute to the etiology of NEC, a common disease in preterm infants.
Journal ArticleDOI
Complete Genome Sequence of Citrus Huanglongbing Bacterium, ‘Candidatus Liberibacter asiaticus’ Obtained Through Metagenomics
Yongping Duan,Lijuan Zhou,David G. Hall,Wenbin Li,Harshavardhan Doddapaneni,Harshavardhan Doddapaneni,Hong Lin,Li Liu,Cheryl M. Vahling,Dean W. Gabriel,Kelly P. Williams,Allan W. Dickerman,Yijun Sun,Tim R. Gottwald +13 more
TL;DR: Multi-protein phylogenetic analysis confirmed 'Ca. L. asiaticus' as an early-branching and highly divergent member of the family Rhizobiaceae, the first genome sequence of an uncultured alpha-proteobacteria that is both an intracellular plant pathogen and insect symbiont.
Journal ArticleDOI
Iterative RELIEF for Feature Weighting: Algorithms, Theories, and Applications
TL;DR: This paper proposes an iterative RELIEF (I-RELIEF) algorithm to alleviate the deficiencies of RELIEf by exploring the framework of the expectation-maximization algorithm.
Journal ArticleDOI
Local-Learning-Based Feature Selection for High-Dimensional Data Analysis
TL;DR: This paper considers feature selection for data classification in the presence of a huge number of irrelevant features, and proposes a new feature-selection algorithm that addresses several major issues with prior work, including problems with algorithm implementation, computational complexity, and solution accuracy.
Journal ArticleDOI
Adaptive boosting for SAR automatic target recognition
TL;DR: The results of large-scale experiments demonstrate that the novel automatic target recognition (ATR) scheme outperforms the state-of-the-art systems reported in the literature.