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

Dr. Kushal Kanwar

Dr. Kushal Kanwar
Assistant Professor (SG)
(91) 01792-239365
kushal.kanwar@juit.ac.in, kushal.kanwar@juitsolan.in
For More Information Click here


Biography :

Dr Kushal Kanwar qualified GATE (2012), UGC-NET (2015), and received his PhD (2021) in Computer Science and Engineering from Panjab University, Chandigarh, India. He received GATE fellowship for his ME (CSE) and is also a recipient of the prestigious Visvesvaraya PhD Scheme for Electronics & IT Scholarship. He holds seven years of teaching experience and his academic contributions include SCI and SCOPUS indexed scientific papers and book chapters published by Emerald, Elsevier, and Springer.

Area of Interest:
Complex Networks, Artificial Intelligence, Machine Learning, Modelling & Simulation, and Quantum Computation.

Open Project Titles:

1. Development of count (wc) clone utilities for corpus statistics in C/C++.
On the Unix/Linux operating system, word count (wc) is a very popular utility to extract statistics of text files supplied to it as arguments. This small project will be developed in C/C++. It will help the student to explore the concept of command line arguments, file handling, and string processing.

2. Leveraging Transfer Learning for Spice Classification.
Spice classification is not much explored in the literature of machine learning. Moreover, benchmark datasets are not available for Spice Classification. Under this project, a small dataset of spices will be created and transfer learning will be leveraged to classify them.

3. Vehicle License plate detection in Wild.
License Plate Detection is a problem that has not been explored much in the Indian context. Moreover, due to differences in plate style, climatic conditions, and recording quality of videos/images, models trained for other countries don't work well in the Indian context. Under this project, Deep Learning Models and Techniques will be explored to suit the Indian Context.

4. Local Information based Edge Ranking Centrality Measure for Complex Networks
Ranking edges of complex networks is a hot area of research in network science with many applications, such as the dwindling spread of harmful contagions like fake news, biological infections, etc. Under this project, new edge centrality measures will be designed by only using local information of the edges for faster ranking.