Tandra Pal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Tandra Pal has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Chiranjib Koley
Electrical Engineering Department, National Institute of Technology Durgapur, Durgapur, West Bengal, India
2 shared publications - Prasenjit Dey
Computer Science and Engineering, National Institute of Technology Rourkela, Rourkela, Odisha, India
2 shared publications - Rahul Mondal
Department of Computer Science and Engineering, National Institute of Technology Durgapur, Durgapur, India
2 shared publications - Aayush Agarwal
Amity University,Department of Computer Science and Engineering, Amity School of Engineering and Technology,Noida,UP,India
1 shared publication - Amiya Biswas
Department of Mathematics, Durgapur Government College, Durgapur, India
1 shared publication Aniruddha Sinha
TCS Research
1 shared publication
Latest Publications
- 7 October 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Information Technology & Decision Making
- Vol. 25 (06) , 1-51
Intuitionistic Type-2 Fuzzy Soft Set-Based Decision Support Framework for Emergency Relief Supply Chain Planning
This study advances the theoretical concept of Intuitionistic Type 2 Fuzzy Set (IT2FS) to develop further Intuitionistic Type 2 Fuzzy Soft Set (IT2FSS). To this end, this work first develops the IT2FSS framework for decision-making, defining the fundamental set-theoretic operations (complement, union, and intersection) and elucidating their algebraic properties. Subsequently, the definitions of level sets, score functions, and accuracy functions for IT2FSS are provided, along with two decision-making algorithms. The practical applicability of the newly proposed theoretical framework is demonstrated in an innovative real-life case study on identifying key enabling factors for effective Humanitarian Supply Chain Management (HSCM) in emergency relief operations. This work proposes a novel framework for measuring preference scores in HSCM, encompassing three key aspects: Importance (I), Complexity (C), and Urgency (U). Based on experts’ opinions, the study reveals three dominant factors: supply chain financing, collaboration and coordination, and public governance, all of which are essential for an effective emergency relief operation. This work contributes to the literature by formalizing IT2FSS and its application, developing a granular soft computing decision analysis framework for HSCM to identify key enablers during crises, and providing a structured strategic decision-making approach for practitioners and policymakers in HSCM.
- 7 October 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Information Technology & Decision Making
- Vol. 25 (06) , 1-51
Intuitionistic Type-2 Fuzzy Soft Set-Based Decision Support Framework for Emergency Relief Supply Chain Planning
This study advances the theoretical concept of Intuitionistic Type 2 Fuzzy Set (IT2FS) to develop further Intuitionistic Type 2 Fuzzy Soft Set (IT2FSS). To this end, this work first develops the IT2FSS framework for decision-making, defining the fundamental set-theoretic operations (complement, union, and intersection) and elucidating their algebraic properties. Subsequently, the definitions of level sets, score functions, and accuracy functions for IT2FSS are provided, along with two decision-making algorithms. The practical applicability of the newly proposed theoretical framework is demonstrated in an innovative real-life case study on identifying key enabling factors for effective Humanitarian Supply Chain Management (HSCM) in emergency relief operations. This work proposes a novel framework for measuring preference scores in HSCM, encompassing three key aspects: Importance (I), Complexity (C), and Urgency (U). Based on experts’ opinions, the study reveals three dominant factors: supply chain financing, collaboration and coordination, and public governance, all of which are essential for an effective emergency relief operation. This work contributes to the literature by formalizing IT2FSS and its application, developing a granular soft computing decision analysis framework for HSCM to identify key enablers during crises, and providing a structured strategic decision-making approach for practitioners and policymakers in HSCM.
- 4 April 2025
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 19 (1) , 1-16
Detection of cognitive load during computer-aided education using infrared sensors
Technology integration in modern education has transformed traditional teaching-learning methods, but maintaining student attentiveness during computer-aided activities remains challenging. Neuroimaging advancements provide valuable insights into cognitive processes. This study measures cognitive load during computer-aided education. We have collected functional near-infrared spectroscopy (fNIRS) brain signals while subjects perform mental tasks and rest. Three datasets have been considered to evaluate the performance of the proposed model. The first two datasets are open-access, and we prepare the third dataset by collecting fNIRS brain signals from 14 healthy subjects. Two feature extraction techniques are proposed: manual and automatic based on wavelet scattering transform (WST). A one dimensional convolutional neural network (1D CNN) is also proposed to automatically extract features through feature engineering and classification. For comparison, four machine learning classifiers, linear discriminant analysis (LDA), Naive Bayes (NB), k-nearest neighbors (KNN) and support vector machine (SVM), are also considered. Classification performance is evaluated using accuracy, precision, recall and F1-score across all datasets. Computational cost, i.e., the CPU time and memory utilization for extracting the features and testing the classifiers, is also evaluated. The results suggest that when considering four classifiers across three datasets and comparing among the manual and the WST-based feature extraction methods, the average performance of 1D CNN is superior in terms of classification accuracy (1.16 times higher), precision (1.10 times higher), recall (1.10 times higher) and F1-score (1.09 times higher). However, the CPU time and memory utilization for 1D CNN are significantly higher, 10.09 and 14.70 times, respectively. In comparison to four state-of-the-art deep learning models, the proposed 1D CNN also shows best classification accuracy (92.99%). The analysis of the results shows that identifying cognitive load, SVM with Gaussian kernel function on WST based methods, provides satisfactory classification performance with significantly less CPU time and memory utilization.
- 4 April 2025
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 19 (1) , 1-16
Detection of cognitive load during computer-aided education using infrared sensors
Technology integration in modern education has transformed traditional teaching-learning methods, but maintaining student attentiveness during computer-aided activities remains challenging. Neuroimaging advancements provide valuable insights into cognitive processes. This study measures cognitive load during computer-aided education. We have collected functional near-infrared spectroscopy (fNIRS) brain signals while subjects perform mental tasks and rest. Three datasets have been considered to evaluate the performance of the proposed model. The first two datasets are open-access, and we prepare the third dataset by collecting fNIRS brain signals from 14 healthy subjects. Two feature extraction techniques are proposed: manual and automatic based on wavelet scattering transform (WST). A one dimensional convolutional neural network (1D CNN) is also proposed to automatically extract features through feature engineering and classification. For comparison, four machine learning classifiers, linear discriminant analysis (LDA), Naive Bayes (NB), k-nearest neighbors (KNN) and support vector machine (SVM), are also considered. Classification performance is evaluated using accuracy, precision, recall and F1-score across all datasets. Computational cost, i.e., the CPU time and memory utilization for extracting the features and testing the classifiers, is also evaluated. The results suggest that when considering four classifiers across three datasets and comparing among the manual and the WST-based feature extraction methods, the average performance of 1D CNN is superior in terms of classification accuracy (1.16 times higher), precision (1.10 times higher), recall (1.10 times higher) and F1-score (1.09 times higher). However, the CPU time and memory utilization for 1D CNN are significantly higher, 10.09 and 14.70 times, respectively. In comparison to four state-of-the-art deep learning models, the proposed 1D CNN also shows best classification accuracy (92.99%). The analysis of the results shows that identifying cognitive load, SVM with Gaussian kernel function on WST based methods, provides satisfactory classification performance with significantly less CPU time and memory utilization.
- 12 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-11
A Deep Learning Technique for Real-Time Detection of Cognitive Load Using Optimal Number of EEG Electrodes
Cognitive load analysis has the potential to significantly enhance brain-computer interfaces (BCIs) by enabling adaptive assistance based on the cognitive state of individuals. This paper presents a real-time approach for detecting cognitive load through electroencephalogram (EEG) signals, with a focus on optimizing computational resources such as CPU time, memory and the number and positioning of EEG electrodes. The study investigates various brain regions, including the prefrontal, frontal, parietal, temporal, and occipital areas, which are critical for identifying cognitive shifts. By leveraging established knowledge of EEG frequency band changes, the research constructs a two-dimensional (2D) brain state image using the Lambert cylindrical equal-area projection and an appropriate interpolation method to represent active brain regions. These 2D images are then processed by a lightweight convolutional neural network (CNN) designed to distinguish between cognitive and resting states. To validate the proposed model, three EEG datasets were employed: one prepared by the authors through experiments involving 15 healthy subjects, and two publicly available datasets. The model achieved an overall accuracy of 95.81% for previously seen subjects and 92.73% for entirely new subjects, utilizing only five electrodes (one prefrontal and four frontal). Furthermore, the model is demonstrated to be suitable for implementation in digital systems with limited computational resources, while maintaining performance and meeting real-time system requirements.
- 12 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-11
A Deep Learning Technique for Real-Time Detection of Cognitive Load Using Optimal Number of EEG Electrodes
Cognitive load analysis has the potential to significantly enhance brain-computer interfaces (BCIs) by enabling adaptive assistance based on the cognitive state of individuals. This paper presents a real-time approach for detecting cognitive load through electroencephalogram (EEG) signals, with a focus on optimizing computational resources such as CPU time, memory and the number and positioning of EEG electrodes. The study investigates various brain regions, including the prefrontal, frontal, parietal, temporal, and occipital areas, which are critical for identifying cognitive shifts. By leveraging established knowledge of EEG frequency band changes, the research constructs a two-dimensional (2D) brain state image using the Lambert cylindrical equal-area projection and an appropriate interpolation method to represent active brain regions. These 2D images are then processed by a lightweight convolutional neural network (CNN) designed to distinguish between cognitive and resting states. To validate the proposed model, three EEG datasets were employed: one prepared by the authors through experiments involving 15 healthy subjects, and two publicly available datasets. The model achieved an overall accuracy of 95.81% for previously seen subjects and 92.73% for entirely new subjects, utilizing only five electrodes (one prefrontal and four frontal). Furthermore, the model is demonstrated to be suitable for implementation in digital systems with limited computational resources, while maintaining performance and meeting real-time system requirements.
- 11 March 2024
- Published by Elsevier in Pattern Recognition
- Vol. 151
Discriminative Regularized Input Manifold for multilayer perceptron
- 11 March 2024
- Published by Elsevier in Pattern Recognition
- Vol. 151
Discriminative Regularized Input Manifold for multilayer perceptron
- 1 November 2023
- Published by Springer Nature
- p. 473-486
NASA-TLX Based Workload Assessment of Learning Tasks for Primary School Children
- 1 November 2023
- Published by Springer Nature
- p. 473-486
NASA-TLX Based Workload Assessment of Learning Tasks for Primary School Children
- 6 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 5 (1) , 115-126
A Hybrid Regularized Multilayer Perceptron for Input Noise Immunity
The immunity of multilayer perceptron (MLP) is less effective towards input noise. In this article, we have focused on the robustness of MLP with respect to input noise where noise can be additive or multiplicative. Here, we have proposed a DropConnect based regularized MLP to reduce the coadaptation among the neurons of the hidden layer. At first, we have empirically and statistically shown that by reducing coadaptation among the hidden neurons, an MLP can achieve better noise immunity. We have also empirically shown that an MLP with input noise injection and $l_{2}$ regularizer is an effective approach to improve its noise immunity. However, the results indicate that it does not adjust the coadaptation among the hidden neurons. Therefore, for further improvement, we have proposed a hybrid regularized MLP (HRMLP), where DropConnect is combined with the noise injection and $l_{2}$ regularizer. In addition to input noise, we have also verified the robustness of HRMLP with respect to $\mathbf {20\%}$ outliers in the data set. To justify the effectiveness of the proposed HRMLP, we have compared it with MLP, MLP with noise injection, MLP with $l_{2}$ regularizer, MLP with noise injection and $l_{2}$ regularizer, and MLP with DropConnect along with two state-of-the-art works based on $\mathbf {20}$ standard data sets. The experimental results for both noisy inputs and outliers confirm that the performance of HRMLP is significant compared to other methods.
- 6 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 5 (1) , 115-126
A Hybrid Regularized Multilayer Perceptron for Input Noise Immunity
The immunity of multilayer perceptron (MLP) is less effective towards input noise. In this article, we have focused on the robustness of MLP with respect to input noise where noise can be additive or multiplicative. Here, we have proposed a DropConnect based regularized MLP to reduce the coadaptation among the neurons of the hidden layer. At first, we have empirically and statistically shown that by reducing coadaptation among the hidden neurons, an MLP can achieve better noise immunity. We have also empirically shown that an MLP with input noise injection and $l_{2}$ regularizer is an effective approach to improve its noise immunity. However, the results indicate that it does not adjust the coadaptation among the hidden neurons. Therefore, for further improvement, we have proposed a hybrid regularized MLP (HRMLP), where DropConnect is combined with the noise injection and $l_{2}$ regularizer. In addition to input noise, we have also verified the robustness of HRMLP with respect to $\mathbf {20\%}$ outliers in the data set. To justify the effectiveness of the proposed HRMLP, we have compared it with MLP, MLP with noise injection, MLP with $l_{2}$ regularizer, MLP with noise injection and $l_{2}$ regularizer, and MLP with DropConnect along with two state-of-the-art works based on $\mathbf {20}$ standard data sets. The experimental results for both noisy inputs and outliers confirm that the performance of HRMLP is significant compared to other methods.
- 1 May 2021
- Published by Elsevier in Expert Systems with Applications
- Vol. 170, 114491
A comparison between metaheuristics for solving a capacitated fixed charge transportation problem with multiple objectives
- 1 May 2021
- Published by Elsevier in Expert Systems with Applications
- Vol. 170, 114491
A comparison between metaheuristics for solving a capacitated fixed charge transportation problem with multiple objectives
- 8 November 2019
- Published by SAGE Publications in Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
- Vol. 37 (6) , 7831-7838
A genetic algorithm for total graph coloring
A genetic algorithm (GA) belongs to the class of evolutionary algorithms and it is one of the most studied heuristic algorithms to solve graph coloring problems. In this paper, we propose a new GA algorithm for the total graph coloring problem. To the best of our knowledge, no algorithm based on a GA exists in the literature for total graph coloring. In the proposed approach, a novel encoding scheme is introduced, where all the edges and vertices of the graph are represented in a chromosome without any repetition. For the initialization of the population, a greedy algorithm is used to determine the total number of colors required for a total coloring of the graph. The number of colors is used as the fitness value of a chromosome which depends on the sequence of vertices and edges representing the chromosome. We introduce a convergence criteria for GA based on the total coloring conjecture. A two-point crossover and mutation operations, suitable for total coloring, are suggested. The proposed algorithm is applied on some well-known and standard graphs. In our computational tests, graphs are used with a maximum number of 690 vertices and 6650 edges of the graph, respectively. The proposed algorithm determines an optimal solution for 21 graphs among the 27 graphs. The solution of remaining the 6 graphs is near optimal and differs by at most one unit from the optimal value. The results show the effectiveness of the proposed approach.
- 8 November 2019
- Published by SAGE Publications in Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
- Vol. 37 (6) , 7831-7838
A genetic algorithm for total graph coloring
A genetic algorithm (GA) belongs to the class of evolutionary algorithms and it is one of the most studied heuristic algorithms to solve graph coloring problems. In this paper, we propose a new GA algorithm for the total graph coloring problem. To the best of our knowledge, no algorithm based on a GA exists in the literature for total graph coloring. In the proposed approach, a novel encoding scheme is introduced, where all the edges and vertices of the graph are represented in a chromosome without any repetition. For the initialization of the population, a greedy algorithm is used to determine the total number of colors required for a total coloring of the graph. The number of colors is used as the fitness value of a chromosome which depends on the sequence of vertices and edges representing the chromosome. We introduce a convergence criteria for GA based on the total coloring conjecture. A two-point crossover and mutation operations, suitable for total coloring, are suggested. The proposed algorithm is applied on some well-known and standard graphs. In our computational tests, graphs are used with a maximum number of 690 vertices and 6650 edges of the graph, respectively. The proposed algorithm determines an optimal solution for 21 graphs among the 27 graphs. The solution of remaining the 6 graphs is near optimal and differs by at most one unit from the optimal value. The results show the effectiveness of the proposed approach.