• home
• current students
• graduate
• events

Presenter: Luyun Gan
Supervisor:

Date: Wed, August 21, 2019
Time: 13:00:00 - 14:00:00
Place: EOW 430

ABSTRACT

Spectroscopic analysis sees plural applications in  physics, chemistry, bioinformatics, geophysics, astronomy, etc. It has been widely used for detecting mineral samples, gas emission, and food volatiles. Machine learning algorithms for spectroscopic analysis focus on either regression or single-label classification problems.  Using multi-label classification to identify multiple chemical components from the spectrum, is under explored. In this thesis, we implement multi-label neural networks with optimal thresholding (FNN-OT) to identify gas species among a multi gas mixture in a cluttered environment. For mutually independent and randomly correlated gas data, FNN-OT increases its recall without much sacrifice of its precision. For positively correlated gas data, FNN-OT is capable of capturing information of positive label correlation from noisy datasets.