Dr. Nayan Kumar Sarkar is currently serving as an Assistant Professor in the Department of Artificial Intelligence and Machine Learning, under the School of Computer Science and Engineering at IILM University, Greater Noida. He earned his M. Tech (IT) from Tezpur University (Central) and Ph.D. in Computer Science and Engineering (AIML) from the North Eastern Regional Institute of Science and Technology (NERIST), under the Ministry of Education, Government of India. With extensive teaching and research experience across reputed institutions, Dr. Sarkar has previously worked with some reputed institutes. His primary research interests include machine learning, deep learning and image classification. He has published several research articles in reputed SCI and Scopus-indexed journals and conferences. Beyond his academic achievements, he has also qualified UGC-NET (CS). He is also the recipient of the prestigious Anundoram Borooah Award (2006) and the MHRD GATE Scholarship (2014–2016).
EDUCATION
PhD (AIML)
RESEARCH, TEACHING, or OTHER INTERESTS
Computer Engineering, Management Information Systems
9
Scopus Publications
43
Scholar Citations
4
Scholar h-index
1
Scholar i10-index
Scopus Publications
PruferNET: An Augmented Graph Neural Network for Influential Node Detection in Social Networks Atashi Saha, Nayan Kumar Sarkar, Sanjoy Pratihar IEEE Access, 2026 Finding the prime nodes, or people in a social network, is very important because it helps in many important uses in different fields. This paper proposes a new way to classify social media nodes using Prufer sequence-based structural features with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Graph Neural Networks (GNNs)</i>. Utilizing the Facebook dataset, a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Minimum Spanning Tree (MST)</i> is built to generate the Prufer sequence and obtain graph-based features. To build a detailed feature set, properties such as label frequency, sequence entropy, subtree depth, unique label ratio, and transition influence are combined with traditional structural measures like degree centrality, closeness centrality, and PageRank. The method uses a GraphSAGE model for node classification, where nodes are tagged as Very Shallow, Shallow, or Deep according to their relative depth in the reconstructed Prufer tree. The results show that the integration of Prufer-based features significantly improves performance. The method attained a test accuracy of 99.63% after adding Prufer features with the traditional feature vectors. This is far better compared to 96.54% accuracy of the result of only traditional features. The precision, recall, and F1 scores for both categories show the strength and reliability of the proposed method. The results highlight the effectiveness of this combined graph-based representation. This is the proof of better performance, especially for those derived from spanning tree encodings such as Prufer sequences, with deep learning models. This system mainly captures hidden structural patterns and identifies influential nodes in large social graphs.
A Transfer Learning Technique to Identify Crop Leaf Diseases Nayan Kumar Sarkar, Sanjeet Kumar Borah, Rahul Anjana, Utpal Nandi, Moirangthem Marjit Singh 2026 2nd International Conference on Computing Sciences and Communications Iccsc 2026, 2026 The development of rapid and accurate detection methods is needed as tomato leaf diseases is an important constraint to agricultural productivity worldwide. In this research work we present a new approach using a deep transfer learning-based technique. The technique considers a pretrained Convolutional Neural Network (CNN) namely DenseNet201 for efficient and accurate identification of healthy and infected tomato leaf images. The pre-trained DenseNet201 was frozen up to 200^\textth layer to preserve its learned features. The features were passed through global average pooling and AReLU activation function. 2 0% dropout rate was also applied for reducing overfitting. The approach was trained on Tomato dataset for 100 epochs considering batch size of 100 and delivered classification accuracy (Acc) of 9 8. 9 3%. The model also delivered 98.67%, 98.63%, and 98.65% precision (Pr), recall (R e), and F 1-score (F 1), respectively. The relative performance investigation also reflects that the developed technique has delivered better results than several recent approaches. The results demonstrate that the pre-trained DenseNet201, signifies an effective tool for automated disease identification as well as can contribute to early response and precision agricultural strategies. However, further enrichments are desirable to progress the generalization ability of the models in diverse environmental circumstances and disease variants. As future directions, research can be addressed considering diverse datasets as well as implementing more robust techniques with enhanced scalability to enhance overall performance and real-world usability in precision agriculture.
Image classification with deconvolution operation and augmentation Nayan Kumar Sarkar, Moirangthem Marjit Singh, Utpal Nandi Computer Journal, 2025 Several image classification approaches have been evolved over the years utilizing convolutional neural network (CNN). In convolution operation of CNN, the shifting of kernels to overlapping regions of the image learns redundant data as the images are strongly correlated in reality. The redundant data make the neural network training a challenging task. Again, Deep Learning methods evaluated on small dataset yields degraded performance. To deal with these issues, a proposal is made in this paper that uses deconvolution operation to minimize correlations from images and data augmentation technique to increase the size of datasets. Plant Village, Tomato, and Covid-19 datasets were used for evaluating the performance of the proposed method. 70% of the datasets were used for training, 10% for validation, and 20% for testing purposes. The CIFAR10, MNIST, and Mini-ImageNet datasets were also considered for performance evaluation. The proposed method performed better than other existing methods in terms of classification accuracy.
Ensemble Transfer Learning for Image Classification Nayan Kumar Sarkar, Moirangthem Marjit Singh, Utpal Nandi, Jyotsna Kumar Mandal Ecti Transactions on Computer and Information Technology, 2025 The deep learning (DL) techniques used for image classification might not deliver the desired level of classification accuracy as some features belonging to some class of a dataset are missed during feature extraction. The ensemble learning (EL) based model improves classification accuracy by combining the strengths of individual classifiers. As a result, those features that were missed during feature extraction by a specific DL technique will be taken care of by another DL technique in an ensemble DL approach. In this paper, averaging EL (AENet), weighted averaging EL (WAENet), and stacking EL (StackedNet) approaches are proposed, considering the DenseNet201, EcientNetB0, and ResNetRS101 as base models. The predictions of the base models are averaged to generate the AENet. The WAENet is constructed by assigning weights to each base model based on their prediction and then taking their average. Similarly, the Stacked-Net is developed by considering the DenseNet201, EcientNetB0, and ResNetRS101 as base-learners and ResNetRS101 as meta-learner. Analysed performance of the considered pre-trained base models and the developed EL models on the standard and application-specific datasets such as MiniImageNet, CIFAR10, CIFAR100, Plant Village (PV), Tomato, Covid-19 and 9IndianFood. 80% of the datasets were used to train and 20% to test the base and proposed models. The models are trained for an epoch of 30, considering a learning rate of 0.001 and adam optimizer. The stackedNet delivered better results than others.
Deep Learning Based Hyperspectral Image Classification: A Review For Future Enhancement Anish Sarkar, Utpal Nandi, Nayan Kumar Sarkar, Chiranjit Changdar, Bachchu Paul International Journal of Computing and Digital Systems, 2024 The use of Hyperspectral Image(HSI) has become prevalent in many sectors due to its ability to identify detailed spectral information (i.e., relationships between the collected spectral data and the object in the HSI data) that cannot be obtained through ordinary imaging.Traditional RGB image classification approaches are insufficient for hyperspectral image classification(HSIC) because they struggle to capture the subtle spectral information that exists within hyperspectral data.In the past few years, the Deep Learning(DL) based model has become a very powerful and efficient non-linear feature extractor for a wide range of computer vision tasks.Furthermore, DL-based models are exempt from manual feature extraction.The use of this stimulus prompted the researchers to use a DL-based model for the classification of Hyperspectral Images, which yielded impressive results.This motivation inspired the researchers to develop a DL-based model for the classification of hyperspectral images, which performed well.Deeper networks might encounter vanishing gradient problems, making optimization more difficult.To address this issue, regularisation and architectural improvements are being implemented.One of the key issues is that the DL-based HSIC model requires a large number of training samples for training, which is an important concern with hyperspectral data due to the scarcity of public HSI datasets.This article provides an overview of deep learning for hyperspectral image classification and assesses the most recent methods.Among all studied methods SpectralNET offers significantly better performance, due to the utilization of wavelet transformation.
PruferNET: An Augmented Graph Neural Network for Influential Node Detection in Social Networks A Saha, NK Sarkar, S Pratihar IEEE Access , 2026 2026
A Transfer Learning Technique to Identify Crop Leaf Diseases NK Sarkar, SK Borah, R Anjana, U Nandi, MM Singh 2026 International Conference on Computing, Sciences and Communications … , 2026 2026
Department of Computer Science & Engineering ABES Engineering College, Ghaziabad, India S Kumar, U Pilania, S Garg, A Gupta, Y Singh, SS Shah, U Upadhyay, ... 2026
A Comparative analysis of Predictive Machine Learning Models on Student Mental Health Impact on Academic Outcomes SKRAAD Brajen Kumar Deka, Amit Dutta, Kshirod Sarmah*, Nayan Kumar Sarkar ... INTERNATIONAL JOURNAL OF ADVANCES IN SIGNAL AND IMAGE SCIENCES 12 (1), 924-932 , 2026 2026
Image Classification Using Deep CNN MM Singh, P Kumar, NK Sarkar International Conference on Sustainability and Emerging Technologies for … , 2025 2025
Image classification with deconvolution operation and augmentation N Kumar Sarkar, M Marjit Singh, U Nandi The Computer Journal 68 (2), 135-144 , 2025 2025 Citations: 2
Ensemble transfer learning for image classification NK Sarkar, MM Singh, U Nandi, JK Mandal ECTI Transactions on Computer and Information Technology (ECTI-CIT) 19 (1 … , 2025 2025 Citations: 2
A novel deep neural network model using network deconvolution with attention based activation for crop disease classification NK Sarkar, MM Singh, U Nandi Multimedia Tools and Applications 83 (6), 17025-17045 , 2024 2024 Citations: 7
Capsule Network Approach for Image Classification NK 1. Singh, M. M., & Sarkar In: Gunjan, V.K., Zurada, J.M. (eds) Proceedings of 4th International … , 2024 2024
Deep Learning Based Hyperspectral Image Classification: A Review For Future Enhancement B 5. Sarkar, A., Nandi, U., Sarkar, N. K., Changdar, C., & Paul . International Journal of Computing and Digital Systems 15 (1), 419-435 , 2024 2024 Citations: 9
Recent researches on image classification using deep learning approach NK Sarkar, MM Singh, U Nandi International Journal of Computing and Digital Systems 12 (1), 1357-1374 , 2022 2022 Citations: 16
Learning based image classification techniques NK Sarkar, MM Singh, U Nandi International Conference on Computational Intelligence in Communications and … , 2022 2022 Citations: 7
Image Classification Using Few Shot Learning MM Singh, NK Sarkar, U Nandi International Conference on Engineering, Applied Sciences and System … , 2017 2017
MOST CITED SCHOLAR PUBLICATIONS
Recent researches on image classification using deep learning approach NK Sarkar, MM Singh, U Nandi International Journal of Computing and Digital Systems 12 (1), 1357-1374 , 2022 2022 Citations: 16
Deep Learning Based Hyperspectral Image Classification: A Review For Future Enhancement B 5. Sarkar, A., Nandi, U., Sarkar, N. K., Changdar, C., & Paul . International Journal of Computing and Digital Systems 15 (1), 419-435 , 2024 2024 Citations: 9
A novel deep neural network model using network deconvolution with attention based activation for crop disease classification NK Sarkar, MM Singh, U Nandi Multimedia Tools and Applications 83 (6), 17025-17045 , 2024 2024 Citations: 7
Learning based image classification techniques NK Sarkar, MM Singh, U Nandi International Conference on Computational Intelligence in Communications and … , 2022 2022 Citations: 7
Image classification with deconvolution operation and augmentation N Kumar Sarkar, M Marjit Singh, U Nandi The Computer Journal 68 (2), 135-144 , 2025 2025 Citations: 2
Ensemble transfer learning for image classification NK Sarkar, MM Singh, U Nandi, JK Mandal ECTI Transactions on Computer and Information Technology (ECTI-CIT) 19 (1 … , 2025 2025 Citations: 2
PruferNET: An Augmented Graph Neural Network for Influential Node Detection in Social Networks A Saha, NK Sarkar, S Pratihar IEEE Access , 2026 2026
A Transfer Learning Technique to Identify Crop Leaf Diseases NK Sarkar, SK Borah, R Anjana, U Nandi, MM Singh 2026 International Conference on Computing, Sciences and Communications … , 2026 2026
Department of Computer Science & Engineering ABES Engineering College, Ghaziabad, India S Kumar, U Pilania, S Garg, A Gupta, Y Singh, SS Shah, U Upadhyay, ... 2026
A Comparative analysis of Predictive Machine Learning Models on Student Mental Health Impact on Academic Outcomes SKRAAD Brajen Kumar Deka, Amit Dutta, Kshirod Sarmah*, Nayan Kumar Sarkar ... INTERNATIONAL JOURNAL OF ADVANCES IN SIGNAL AND IMAGE SCIENCES 12 (1), 924-932 , 2026 2026
Image Classification Using Deep CNN MM Singh, P Kumar, NK Sarkar International Conference on Sustainability and Emerging Technologies for … , 2025 2025
Capsule Network Approach for Image Classification NK 1. Singh, M. M., & Sarkar In: Gunjan, V.K., Zurada, J.M. (eds) Proceedings of 4th International … , 2024 2024
Image Classification Using Few Shot Learning MM Singh, NK Sarkar, U Nandi International Conference on Engineering, Applied Sciences and System … , 2017 2017
