Ke Liu

Ke Liu

Auto-generated profile, based on publicly available information

Ke Liu

Naval University of Engineering, Wuhan, 430033, P. R. China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2019 in TrAC Trends in Analytical Chemistry
Verified
Journal Article
Closed access
KLKe Liu
Ke Liu
DTDi Tian
Di Tian
CLChang Li
Chang Li
YLYingchao Li
Yingchao Li
GYGuang Yang
Guang Yang
YDYu Ding
Yu Ding

25 February 2026 in Journal of Analytical Atomic Spectrometry
Verified
Journal Article
Closed access
PDF file
JYJiujiang Yan
Jiujiang Yan
MZMingjun Zhou
Mingjun Zhou
KLKe Liu
Ke Liu
YLYang Li
Yang Li
JMJinxiu Ma
Jinxiu Ma
GZGeng Zhang
Geng Zhang
With the increasing demand for marine resources, the existing marine exploration technologies face multiple challenges.

28 July 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
JYJiujiang Yan
Jiujiang Yan
MZMingjun Zhou
Mingjun Zhou
YLYang Li
Yang Li
KLKe Liu
Ke Liu
ZZZhong Zheng
Zhong Zheng
GZGeng Zhang
Geng Zhang
The miniaturization and integration of laser-induced breakdown spectroscopy (LIBS) technology is one of its important development trends, and also the only way for it to be practical application from the laboratory. Aiming at the problems of limited computing power and relatively low model computing efficiency in the qualitative analysis of compact LIBS technology, a feature selection method based on the cumulative correlation coefficient (CCC) was proposed. It was verified by the classification based on linear discriminant analysis (LDA) algorithm for 12 kinds of igneous rock samples. Using this method, the classification accuracy (CA) and relative standard deviation (RSD) of the samples reached 94.8% and 12.4%, respectively, while only 91 analytical lines were required in its qualitative analysis model. Compared with traditional signal-to-noise ratio (SNR) (105 Lines) or signal-to-background ratio (SBR) (210 lines) methods, the feature dimension of the model was effectively reduced by 13.3% and 56.7%, respectively. The time complexity and space complexity have also been significantly reduced. It is of great significance for improving the analysis efficiency of the compact LIBS system. It provided a way that can be standardized and produced deterministic results for the feature selection process of LIBS, and ensured the reliability of the analysis model and result.