Sujan Kumar Roy
![Sujan Kumar Roy](/cs/department/people/faculty/roy/images/sujan-kumar-roy-personnel170.jpg)
- Assistant Teaching Professor, Computer Science
- Ph. D. Machine Learning with Computer Engineering and Signal Processing, Griffith University
- M. A. Sc. Electrical and Computer Engineering, Concordia University
- M. Sc. Computer Science and Engineering, University of Rajshahi
- B. Sc. Computer Science and Engineering, University of Rajshahi
Areas of Expertise
- Artificial Intelligence
- Machine Learning and Deep Learning
- Data Science and Data Mining
-
Applications of AI, ML, and DL toCybersecurity, Medical Image Analysis,
Healthcare Systems, and Speech Enhancement
- K. Roy, A. Nicolson, and K. K. Paliwal, "On Supervised LPC Estimation Training Targets for Augmented Kalman Filter-based Speech Enhancement," in Speech Communication, vol. 142, pp. 49-60, doi: https://doi.org/10.1016/j.specom.2022.06.004.
- K. Roy, A. Nicolson, and K. K. Paliwal, "DeepLPC-MHANet: Multi-Head Self-Attention for Augmented Kalman Filter-based Speech Enhancement," in IEEE Access, vol. 9, pp. 70516-70530, 2021, doi: 10.1109/ACCESS.2021.3077281.
- K. Roy, A. Nicolson, and K. K. Paliwal, "DeepLPC: A Deep Learning Approach to Augmented Kalman Filter-Based Single-Channel Speech Enhancement," in IEEE Access, vol. 9, pp. 64524-64538, 2021, doi: 10.1109/ACCESS.2021.3075209.
- K. Roy and K. K. Paliwal, " Robustness and Sensitivity Metrics-Based Tuning of the Augmented Kalman Filter for Single-Channel Speech Enhancement," in Applied Acoustics, Volume 185, 2022, doi: 10.1016/j.apacoust.2021.108355.
- K. Roy and K. K. Paliwal, " Robustness and Sensitivity Tuning of the Kalman Filter for Speech Enhancement," in signals, vol. 2, pp. 434-455, 2021, doi: 10.3390/signals2030027.
- K. Roy and K. K. Paliwal, " A Noise PSD Estimation Algorithm using Derivative-based High-pass Filter in Non-stationary Noise Conditions," in EURASIP Journal on Audio, Speech, and Music Processing, 32 (2021), doi: 10.1186/s13636-021-00220-9.
- K. Roy, A. Nicolson, and K. K. Paliwal. (2020) "A Deep Learning-Based Kalman Filter for Speech Enhancement," Proc. Interspeech 2020, pp. 2692-2696.
- K. Roy, A. Nicolson, and K. K. Paliwal, "Deep Learning with Augmented Kalman Filter for Single-Channel Speech Enhancement," 2020 IEEE International Symposium on Circuits and Systems (ISCAS), 2020, pp. 1-5.
- K. Roy and K. K. Paliwal, "Deep Residual Network-Based Augmented Kalman Filter for Speech Enhancement," 2020 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), 2020, pp. 667-673.
- K. Roy and K. K. Paliwal, "Causal Convolutional Encoder Decoder-Based Augmented Kalman Filter for Speech Enhancement," 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS) 2020, pp. 1-7.
- K. Roy and K. K. Paliwal, "Causal Convolutional Neural Network-Based Kalman Filter for Speech Enhancement," 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), 2020, pp. 1-6.
- K. Roy and K. K. Paliwal, "Sensitivity Metric-Based Tuning of the Augmented Kalman Filter for Speech Enhancement," 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), 2020, pp. 1-6.
Current Research Activities
- Deep Learning and Transfer Learning Assisted Wearable Sensor Data Analysis for Decision-Support Systems.
- Vision Transformer and Multimodal Learning- assisted Histological Image Analysis for Cancer Detection.
- Investigating Explainable AI (XAI) Algorithms for Clinical Decision-Support Systems.
- Investigating Multimodal Learning and Deep Learning Algorithms for Fake Image, Video, Audio, and Text Detection in Social Media.
- Investigating Large Language Model for Information retrieval and cyberbullying detection in social media for English and Bengali Language.
- Investigating Multimodal Learning and Deep Learning Algorithms for Speech Intelligibility Enhancement and Hearing-aids.
- Department of Computer Science & Engineering, University of Rajshahi, Bangladesh
- Associate Professor (Jan 2022 - Aug 2024)
- Assistant Professor (Mar 2013-May-2013, Mar 2016-Oct 2017, Aug 2021- Jan 2022)
- Lecturer (Mar 2010 - Mar 2013)
- School of Engineering and Built Environment, Griffith University, Australia
- Research Assistant (Oct 2017-July 2021): Design Machine Learning Algorithms for Computer Science, Signal/Image Processing, and Speech Enhancement.
- Department of Electrical and Computer Engineering, Concordia University, Canada
- Research Assistant (May 2013-April 2016): Working as a Researcher and Algorithm Developer in CRD Projects with Miranda Technology, and Microsemi Ltd, Canada.
- Teaching Assistant (Sep 2013-Dec 2015): Tutor and Capstone Project Supervisor.
Skills and Experiences
- Statistical Machine Learning Algorithms: I have expertise in Python and R, including libraries such as Scikit-learn and Pandas. I am proficient in machine learning optimization techniques such as grid search, random search, and Bayesian optimization. My experience covers a range of machine learning algorithms, including Decision Trees, Random Forests, SVM, XGBoost, AdaBoost, MLP, and various classification and clustering algorithms. Additionally, I am skilled in KNN, Bayesian methods, cluster analysis, pattern recognition, and multidimensional data analysis and visualization.
- Deep Learning Algorithms: I specialize in Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Residual Networks, Temporal Convolutional Networks, Deep Dense Networks, Convolutional Encoder-Decoders, and Multimodal Learning. My expertise also includes Generative Adversarial Networks (GANs), Transfer Learning, Unsupervised Learning, Transformer Networks, Language Models, Self-Attention Networks, Vision Transformers, and Explainable AI.
- I have extensive experience using Python, OpenCV, TensorFlow, PyTorch, R, and Keras on the Linux platform for implementing various machine learning and deep learning algorithms.
- I have expertise in Python-assisted GPU programming for big data processing and data pipelining, including data analytics, engineering pipelines for data aggregation and processing, database management, analysis, and visualization.
- I possess strong expertise on Java, C/C++, MATLAB, SQL, No SQL, C#, .NET, and PHP.
My extensive tenure in academia has provided me with invaluable insights into both learning and teaching. Drawing from these experiences, I have formulated a teaching philosophy centered on fostering independent problem-solving skills among students, particularly in the fields of computer science and data science. My aim is to promote literacy, critical thinking, effective communication, and collaboration.
In computer science, I emphasize algorithmic thinking and debugging strategies alongside teaching programming languages like Python and Java. This approach enables students to tackle complex coding challenges independently, enhancing their problem-solving abilities.
In data science courses, I integrate real-world data projects involving data collection, cleaning, analysis, and visualization using tools such as R and SQL. This hands-on experience helps students understand the entire data pipeline, from theoretical concepts to practical applications, while also considering ethical implications and data privacy.
Collaboration and communication are vital components of my teaching philosophy. I incorporate group projects and peer reviews to develop students' teamwork and communication skills, preparing them for professional environments where interdisciplinary collaboration is essential. Additionally, I require students to present their projects and research findings, helping them refine their public speaking and technical writing abilities.
Ultimately, my goal is to empower students to become independent, critical thinkers and lifelong learners. Through a supportive and engaging learning environment, I strive to inspire students to take ownership of their education and succeed in the rapidly evolving fields of computer science and data science.