Deep Chemometrics Masters Project
Deep Neural Networks and Spectroscopic Data
– a chemometric study
A Masters project is available in a collaboration between Wildcard Pharmaceutical Consulting and Department of Food, Spectroscopy and Chemometrics, KU. We are looking for a motivated student with interest in spectroscopy and computer programming to take on a project involving deep neural networks and spectroscopical data.
Recent advances in image analysis using deep neural networks will be leveraged to improve chemometric analysis of spectral data. For CNN’s this has been explored http://dx.doi.org/10.1016/j.aca.2016.12.010 with good results.
The project should explore the use of more advanced deep neural network image analysis techniques and compare the results with more traditional spectrum normalization and pre-treatment. Data augmentation need to be thought into a chemometric setting but could potentially remove the need for pretreatment and normalization as the network will learn to disregard these variations and “invent” its own pretreatment in the lower layers.
Spectrum picture By Deglr6328 at the English language Wikipedia, CC BY-SA 3.0,