Recently researchers from Institute of Intelligent Machines,Hefei Institutes of Physical Science. developed a new wavelength selection algorithm based on combined moving window (CMW) and variable dimension particle swarm optimization (VDPSO) algorithm.

 
CMW retained the advantages of the moving window algorithm, and different windows could overlap each other to realize automatic optimization of spectral interval width and number. VDPSO algorithms improved the traditional particle swarm optimization (PSO) algorithm.

 
This new algorithm, which is called VDPSO-CMW, could search the data space in different dimensions, and reduce the risk of limited local extrema and over fitting.

 
Combined with the moving window strategy, the spectral data variables could be quickly selected.

 
Comparing with four high-performance variable selection algorithms BOSS, VCPA, iVISSA and IRF, the results showed that the algorithm could select the more important spectral information and improve the predictive ability of the model.

 
The algorithm was expected to be further applied to data analysis in the fields of genomics, proteomics, metabolomics and quantitative structure-activity relationship (QSAR).

This work was supported by Anhui Key Research and Development Program, the Strategic Priority Research Program of the Chinese Academy of Sciences, and the National Natural Science Foundation of China.

Link to the paper: A novel variable selection method based on combined moving window and intelligent optimization algorithm for variable selection in chemical modeling.

The frequency of variable selected by different methods  (image by ZHANG Pengfei)

Contact:

ZHOU Shu
Hefei Institutes of Physical Science (http://english.hf.cas.cn/)
Email: zhous@hfcas.ac.cn