Multi-label Classification with Optimal Thresholding for Multi-composition Spectroscopic Analysis

TL;DR

This paper presents a multi-label neural network approach with optimal thresholding for identifying gas species in complex spectroscopic data, outperforming traditional methods under certain conditions.

Contribution

The paper introduces a novel multi-label neural network method with optimal thresholding tailored for spectroscopic analysis of gas mixtures, demonstrating improved performance over conventional techniques.

Findings

Outperforms binary relevance - partial least squares discriminant analysis with sufficient SNR and training data

Effective in cluttered environments with synthesized spectral datasets

Applicable to multi-gas spectroscopic identification tasks

Abstract

In this paper, we implement multi-label neural networks with optimal thresholding to identify gas species among a multi gas mixture in a cluttered environment. Using infrared absorption spectroscopy and tested on synthesized spectral datasets, our approach outperforms conventional binary relevance - partial least squares discriminant analysis when signal-to-noise ratio and training sample size are sufficient.