Rapid analysis of meat floss origin using a supervised machine learning-based electronic nose towards food authentication

This paper presents the development and assessment of a compact portable electronic nose (e-nose) for authenticating the origin of meat floss. The e-nose uses a gas sensor array and supervised machine learning with a window time slicing method to classify beef, chicken, and pork flosses. Linear discriminant analysis (LDA) with five-window-extracted features achieved >99% accuracy in discriminating the meat floss types. The e-nose results were validated using Fourier-transform infrared (FTIR) spectroscopy and gas chromatography-mass spectrometry (GC-MS). The study found distinct volatile compound profiles for each meat type, with aldehydes being dominant in pork. The e-nose system shows promise for food authenticity testing.

Linda Ardita Putri, Iman Rahman, Mayumi Puspita, Shidiq Nur Hidayat, Agus Budi Dharmawan, Aditya Rianjanu, Sunu Wibirama, Roto Roto, Kuwat Triyana, Hutomo Suryo Wasisto