Machine Learning Methods Applied to Vibrational Spectroscopy:
New Horizons
Abstract
Machine Learning Methods have revolutionized almost every sphere of research over the last decade. Vibrational Spectroscopy is also not an exception to this. This keynote presentation explores the emerging horizon where the Machine Learning and Vibrational Spectroscopy meet. Machine learning methods are being used by the spectroscopists for functional group identification, spectrum prediction, spectrum-based structure recognition and explanatory analysis of spectroscopic data. Algorithms are being written to generate the data set with an artificial neural network that may be used to predict spectroscopic properties of interest. Since its advent, vibrational spectroscopy has been widely used to characterize the properties of test samples in various biomedical and engineering fields. Many of these tasks require the analysis of recorded vibrational signals to extract information, which can be achieved using machine learning. Broadly speaking, any task that analyses a signal for fetching the information can be regarded as a potential application of artificial intelligence. The conventional artificial intelligence methods adopt a knowledge-based analytical approach which has various limitations. However, new learning-based approach when implemented using specific machines has provided us with different machine learning methods which when combined with the previous spectroscopic techniques will open new routes for novel research.
Short Biography
Dr. Deepa Sharma is a theoretical physicist with expertise in computational nano-physics spanning across Spectroscopy, Optics, Condensed Matter Physics and Electronics. Her research work is focused mainly upon simulation and modelling of carbon nanomaterials and calculation of their electronic, spectroscopic and optical properties based on Density Functional Theory and Tight Binding Model. Her recent theoretical prediction of the possibility of proximity induced superconductivity in singlewalled carbon nanotubes has proven to be path-breaking paving a novel research pathway for the experimentalists to explore. She is serving as an assistant professor of Department of Higher Education, Government of Haryana, India.