Computer Science & Engineering and Information Technology (CSE&IT)

Dr. Sahil Sharma

Dr. Sahil Sharma
Assistant Professor (SG)
(91)01792-239294
sahil.sharma@juit.ac.in, sahil.sharma@juitsolan.in
For More Information Click here


Biography :

Dr. Sahil Sharma is currently working as Assistant Professor (Senior Grade) in the Department of Computer Science & Information Technology at Jaypee University of Information Technology, Waknaghat, Solan, Himachal Pradesh. He has more than 5 years of teaching and research experience. He has completed his Ph.D. thesis with title "Metaheuristic Approaches for Occlusion Invariant 3D Face Recognition Technique" in Computer Science & Engineering Department from Thapar Institute of Engineering and Technology, Patiala, Punjab. He obtained his M.E. with thesis title “Predicting Employability from User Personality using Ensemble Modelling” in Computer Science Information Security from Thapar University, Patiala, Punjab. He has obtained his B.Tech. in Computer Science & Engineering from Punjab Technical University, Jalandhar, Punjab. He has also qualified NTA UGCNET June 2019 (98.901 Percentile) and GATE CS 2013 (96.7193 Percentile). He has around 9 SCI/SCIE research publications and few conference publications.

Qualifications:
Ph.D. (Computer Science and Engineering).
Thapar Institute of Engineering and Technology, Patiala, India in (2016 - 2021).
Thesis Topic: Metaheuristic Approaches for Occlusion Invariant 3D Face Recognition Technique.

M. E. (Computer Science – Information Security).
Thapar Institute of Engineering and Technology, (Formerly Thapar University), Patiala, India in 2015.
Thesis Topic: Predicting Employability from User Personality using Ensemble Modelling

B. Tech. (Computer Science and Engineering).
Rayat and Bahra Institute of Engineering and Bio-Technology (Formerly part of Punjab Technical University, Jalandhar), Sahauran, Mohali, India in 2012

Interest Areas:
Machine Learning, Deep Learning, Pattern Recognition, Computer Vision, and Natural Language Processing

Open Project Titles:

1. 2.5D Face Recognition using Deep Learning Face depth image can be used for occlusion presence and gender prediction by transfer learning. You will be working on the overfitting problem and how augmentation helps overcoming it. Pre-processing of the dataset would be needed for further processing. Transfer learning based state-of-the-art 2D deep learning models (eg, AlexNet, VGG16, DenseNet121, ResNet18, and SqueezeNet) can be discussed along with their architecture to solve the problem.

2. Trading Stocks Based on Financial News Using Attention Mechanism Sentiment analysis of news headlines is an important factor that investors consider when making investing decisions. The sentiment analysis of financial news headlines impacts stock market values. Hence, financial news headline data are collected along with the stock market investment data for a period of time. Using Valence Aware Dictionary and Sentiment Reasoning (VADER) for sentiment analysis, the correlation between the stock market values and sentiments in news headlines would be established.

3. Spoken Language Identification Using Deep Learning The process of detecting language from an audio clip by an unknown speaker, regardless of gender, manner of speaking, and distinct age of the speaker, is defined as spoken language identification (SLID). The considerable task is to recognize the features that can distinguish between languages clearly and efficiently.

4. Distracted Driver Detection using Learning Representations In order to lower the likelihood of accidents caused by people, many driver monitoring systems (DMSs) have been proposed. Using general models trained with the data gathered during aberrant driving, traditional DMSs concentrate on detecting specific predetermined abnormal driving behaviours, such as drowsy or distracted driving. However, it is challenging to compile enough representative training data to build universal detection models that work with all drivers. To address this issue, you will be using deep learning-based techniques.