- Timetable
- Download all abstracts
- Plenary Lectures
- MSSJ Special Program
- Award Lecture
- Symposium Sessions（Day1, Day2）
- Fundamental Sessions（Day1, Day3）
- Young Researchers' Sessions (Int'l)（Day1, Day3）
- Young Researchers' Sessions（Day1, Day3）
- Poster Presentations（Day1, Day2, Day3）
- Evening Workshop
- Corporate Program
Symposium Sessions
Day 1, June 10（Mon.） 16:23-16:43 Room B (Convention Hall 200)
- 1B-S-1623
Peak Detection using Non-negative Matrix Factorization for GC/MS Scan Mode Data of Environmental Sample
(1CESS, 2NIES)
oNobutoshi Ohtsuka1, Kotaro Minomo1, Shunji Hashimoto2
The GC-MS scan mode measurement results in a series of mass spectrum at each retention time. As the data for environmental sample contain information on many compounds, detecting individual compounds may be difficult. Non-negative matrix factorization (NMF) is one of multivariate analysis, that has attended in recent years. NMF is a method for decomposing a matrix consisting only of elements greater than or equal to zero into a product of two matrices under nonnegative constraints. Applying NMF to the scan data should obtain mass-spectra and chromatograms for the individual chemicals. The NMF algorithm based on Kullback-Leibler divergence was modified to reduce the influence of column bleed on peak detection. In increasing the number of factors of NMF calculation, the obtained results are used as part of the initial values in order to reduce calculation time. Bayesian information criterion is used to determine the optimal number of factors (number of compounds). The peaks can be clearly detected using NMF with these modifications.