Swapnanil Das

Confident and passionate about learning new skills. Ambitious and driven individual ready and willing to work hard and learn from professionals. Brings outstanding computer and communication skills.

• Python programming, Webel Informatics Ltd., Kolkata, West Bengal, 07/2022 - 12/2022
• Design Analysis and Algorithms, Education World, Kolkata, West Bengal, 07/2021 - 12/2021
• Workshop in Autonomous Robotics, IIT Kharagpur 04/14/2022 - 04/18/2022
• VLSI Design Engineer, Industrial Training at e-COE, Bhubaneshwar 12/2019 - 01/2020

Comparing the Effectiveness of various Machine Learning Models in the Fault Detection in Steel plates, IOSR Journal of Computer Engineering (IOSR-JCE), 2278-0661, 2278-8727, 24, 4, II, 51, 62, http://www.iosrjournals.org, https://drive.google.com/file/d/1nThnY8P3Y3UKNS219UkpwLtKukx06ZP7/view?usp=share_link

Outbreak Prediction of Covid-19 using Machine Learning (Minor Project).                 05/2020 - 06/2020 

Developed a deep learning model using TensorFlow and PyTorch to analyze the data, uncover patterns, and predict outbreaks. I explored CNNs and LSTM networks to understand Covid-19 spread complexities. During model refinement, I discovered a previously unreported correlation between population density, mobility patterns, and outbreak speed in various regions.

Biomedical Signal Processing (Detection of R-peak by studying various ECG samples using MATLAB) (Major Project)                                                                                                        05/2021 - 07/2021 

I aimed to detect R-peaks in ECG signals using MATLAB. My responsibilities included collecting ECG samples from diverse sources and preprocessing them. Using MATLAB, I applied filters, normalization, and segmentation techniques to prepare the data for R-peak detection. Defect Detection of Steel Plates Using Machine Learning (Journal), 07/2022, 08/2022, Trained machine learning for automatic pattern recognition using various ML algorithms, including Logistic Regression, Decision Tree, and Random Forest. Additionally, a variety of feature extraction methods are applied to the training data to determine the accuracy. The Random Forest approach was concluded to be the most efficient method, with accuracy ratings of 78.58% for PCA, 76.42% for LDA, and 71.68% for Simple Classification.

• Prof Chandrima Roy, Assistant Professor (ECE Dept.), Heritage Institute of Technology, chandrima.roy@heritageit.edu
• Prof Amrita Banerjee, Assistant Professor (ECE Dept.), Heritage Institute of Technology, +91 9830691405, amrita.banerjee@heritageit.edu
• Prof (Dr.) Sayantani Datta, Associate Professor (ECE Dept.), Heritage Institute of Technology, +91 9831283371, sayantani.datta@heritageit.edu