Comparative Analysis on Machine Learning and One-Dimensional Convolutional Neural Network to Predict Surface Enhanced Raman Spectroscopy

• Published in: 2023 3rd International Conference on Computing and Information Technology (ICCIT) pp. 216 - 221
• Main Authors: Jamil, Nasrin Nadher, Khairi Kareem, Aythem
• Format: Conference Proceeding
• Language: English
• Published: IEEE 13.09.2023
• Subjects: Convolutional Neural Network; Convolutional neural networks; Decision Tree; Feature extraction; Machine Learning; Naïve Bayes; Predictive models; Raman scattering; Safety; Support Vector Machine; Support vector machines; Training
• DOI: 10.1109/ICCIT58132.2023.10273940

Surface-enhanced Raman spectroscopy (SERS) established on machine learning (ML) techniques have been used in intelligence investigation, food safety, biological recognition, and material study. However, ML techniques typically require additional preprocessing or attribute engineering, and conducting large data employing these techniques is problematic. Deep Learning (DL) techniques involve minimal input data by analyzing every probable attribute established in the network. Therefore, this study employed ML techniques. These are K-Nearest Neighbor (KNN), Naïve Bayes (NB) Support Vector Machine (SVM), and Decision Trees (DT). Then, the result is compared with a one-dimensional convolutional neural network structure (1D-CNN). In this research, we utilize Rhodamine 6G (R6G) as an objective molecule. The experimental results demonstrate the effectiveness of the 1D-CNN offered.