View Proposal

---

Proposer

Sarat Dass

Title

Classification Techniques on GC-IMS spectroscopy for the detection of origin of food samples.

Goal

Classification techniques based on spectrometry data

Description

This proposal investigates the effectiveness of a certain method of spectrometry, namely, gas chromatography hyphenated with ion mobility spectrometry (GC-IMS), as a method for detecting volatile organic compounds. GC-IMS methodology is being successfully used in the field of food fraud detection. The proposal here is to develop feature selection and classification techniques using neural networks to detect the place of origin of food samples based on their spectral “fingerprint”. Specifically, the dataset used in this proposal contains GC-IMS measurements of extra virgin olive oil from three countries: Greece, Spain and Italy, and the goal is to come up with classification techniques on the GC-IMS spectra to determine the country of origin.

Resources

1. Joscha Christmann, Sascha Rohn, Philipp Weller, “GC-IMS data on the discrimination between geographic origins of olive oils,” Data in Brief, Volume 45, 2022, URL: https://www.sciencedirect.com/science/article/pii/S2352340922009349 2. Joscha Christmann, Sascha Rohn, Philipp Weller, “gc-ims-tools – A new Python package for chemometric analysis of GC–IMS data,” Food Chemistry, Volume 394, 2022, URL: https://www.sciencedirect.com/science/article/pii/S0308814622014388

Background

Url

External Link

Difficulty Level

Moderate

Ethical Approval

None

Number Of Students

1

Supervisor

Sarat Dass

Keywords

classification, data analysis, machine learning, spectrometry data

Degrees

Bachelor of Science in Statistical Data Science