Dr Kamal Kumar is working as Associate
Professor in the Department of Information Technology,
IGDTW since December 2023. Previously, He was
associated with NIT Uttarakhand, Srinagar
(Uttarakhand). He served at NIT Uttarakhand from July
2018 to Dec 2023. He was also associated with University of Petroleum as
Associate Professor (July 2015 to June 2018) and
Maharishi Markandeshwar University, Mullana as
Associate Professor (Sep 2000 to June 2015). His total
experience is more than 24 years. He has served
several administrative responsibilities including HoD
CSE at NIT Uttarakhand, PTP Coordinator (NIT
Uttarakhand), Head Academics (UPES) among many
other. He has guided 04 PhD and supervising 04 PhD
scholars. He has guided 33 PG (M. Tech) students as
supervisor. He has published more than 60 Research
papers (18 SCI, 08 SCOPUS, 10 Referred, and 35
Conferences). He has published 01 Design Patent and
submitted 01 Patent with IPR for patent grant. He has
served as Chair NGCT 2018, TPC Chair NG
EDUCATION
PD, M. Tech. B. Tech.
RESEARCH, TEACHING, or OTHER INTERESTS
Computer Networks and Communications, Artificial Intelligence, Computer Vision and Pattern Recognition, Computer Vision and Pattern Recognition
MH-ARO: Meta-heuristic based adaptive routing for large scale opportunistic networks Soni Chaurasia and Kamal Kumar Wiley SummaryIn wireless sensor networks, sensor nodes (SNs) are placed throughout a wide area and gather information from the surroundings. SN used to detect and send data consumes a lot of energy and dies instantly, which causes network overhead issues. Due to network overhead, network faults occur and do not cover a significant region for data transmission. A meta‐heuristic‐based adaptive routing for large‐scale opportunistic networks (MH‐ARO) is proposed in this work to overcome these problems. The network exploits the dragonfly approach (DA) with opportunistic routing in this protocol. The DA is based on local and global search (GS) optimization. In local search (LS), each node assigns/uses a rank for data transmission and selects the forwarder node after applying the decision level matrix in a single group. In GS, an opportunistic network has multiple groups. Each group sends optimal data to the next group. Each group assigns a relay ranking based on the forwarder node's highest rank. The BS receives critical data from the relay and increases the survival of the node. Based on LS and GS, MH‐ARO is categorized into two parts: (1) optimal forwarder opt (OFO) and (2) optimal route selection (ORS). In OFO, forwarder node selection is based on the following factors: relay ranking, node density, group ranking, residual energy, forwarder distance, and relay distance. ORS follows the adaptive routing and sends the data to BS using the best optimal route. The BS receives critical data without network failure and increases the survival of the node. The maximum active node participates in the network and consumes less energy so that a node covers a large region for communication. Comparing MH‐ARO's performance to competing routing protocols, key performance indicators such as PDR, MSR, alive node, survivability, AEC, AD, and throughput are examined. The results demonstrate that the MH‐ARO performs noticeably better than its rivals regarding energy efficiency.
Detecting adversarial examples using constraint-based learning Vivek Kumar, Kamal Kumar, and Maheep Singh IEEE Deep learning models have been applied in various fields and continue producing exceptional results. However, these models are vulnerable to adversarial attacks, which are modified data samples maliciously crafted to cause misclassification by a target model. Fields like Network intrusion detection systems (NIDS) are security-critical; therefore, adversarial attacks pose a high risk in such applications. The attacker not only aims for evasion but also tries to launch an attack once the adversarial example has evaded detection by a NIDS. This paper presents a novel technique for detecting adversarial examples targeted towards a NIDS by learning the dependencies between features known as domain constraints. Our proposed method detects adversarial examples that violate these constraints learned during training. Compared with prevalent detection techniques, our method does not require generating adversarial examples for training, which reduces its computational requirement. We assume a grey-box threat model where the attacker can access the training data. We perform experiments on the CICIDS2017 dataset against five state-of-the-art adversarial attacks: FGSM, BIM, PGD, Deep fool and BIM. The results show that our method can detect up to 100% adversarial attacks, outperforming the contemporary schemes.
Feature Selection in Intrusion Detection using Jaccard's Needham Similarity Algorithm Kamal Kumar, Kartik Gupta, Rohan Gupta, and Anshul Arora IEEE The Intrusion Detection System (IDS) plays a vital role in upholding network security, safeguarding network resources and infrastructures. This paper presents an in-depth methodology for enhancing intrusion detection using machine learning in the context of network security. The study integrates various techniques to measure classification performance, with focus on a unique feature selection algorithm called Jaccard-Needham similarity algorithm. This innovative approach quanti-fies dissimilarity between features, strategically deselecting those with the least similarity, thereby enhancing the model's discriminative power. The culmination of our research entails a thorough comparative analysis of machine learning methodologies, encompassing Decision Trees, Random Forest, XGBoost, Gaussian Naive Bayes (GNB), and Logistic Regression. The evaluation process is conducted using the Jaccards feature selection algorithm across different thresholds and is also compared with other existing feature selection techniques. The analysis reveals compelling results with our proposed approach giving highest accuracy (99.41%) at 0.0001 threshold for Jaccard index with top 15 features, outperforming the other evaluated feature selection techniques. This outcome underscores its effectiveness in identifying and selecting features conducive to superior model performance.
EEM-CRP: Energy-Efficient Meta-Heuristic Cluster-Based Routing Protocol for WSNs Soni Chaurasia, Kamal Kumar, and Neeraj Kumar Institute of Electrical and Electronics Engineers (IEEE) Wireless sensor network (WSN) is an infrastructure-less network that deploys many interconnected sensors to monitor physical environmental conditions and sends data to the cloud/base station. Sensors monitor, analyze, and route data in Internet of Things (IoT)/WSNs. Nodes consume maximum energy for processing and routing the data from the physical environment. This article proposes an energy-efficient meta-heuristic cluster-based routing protocol (EEM-CRP) for WSNs to address energy issues. EEM-CRP protocol exploits dragonfly meta-heuristic algorithm for selecting optimal cluster head (CH) and route. The dragonfly algorithm (DA) is a kind of decision-making approach (DMA) where it works for exploitation and exploration. DA-based EEM-CRP protocol is divided into two parts: 1) optimal cluster head selection (OCHS) and 2) optimal route selection (ORS). OCHS is based on the exploitation aspect of DA and it selects the CH based on parameters such as node density, residual energy, and distance. ORS is based on exploration and choosing an optimal path based on the random walk/levy distribution function. It covers a large region with reduced energy consumption and overcomes network overhead issues. EEM-CRP ensures the participation of the maximum number of active sensor nodes’ inflow of data without much or no delay. EEM-CRP performance is compared with LEACH, whale and gray wolves (WGWO), and hybridization of metaheuristic algorithm for dynamic cluster-based routing protocol (HMBCR) algorithms on important factors including packet delivery ratio (PDR), delay, number of active nodes, and average residual energy (ARE). According to the simulation results, the EEM-CRP significantly exceeds the methodologies used by its rivals in terms of energy efficiency.
QVD-Querying Video Databases for Event Related Frames Using Text Keywords Ravi Ranjan Kumar, Kamal Kumar, Anuj Kumar Jain, Vikrant Sharma, Neha Sharma, and Nitin Jain IEEE With increasing camera enabled devices huge video data is generated every second. With unlimited storage offered by cloud, most of this video data is moved to cloud storage directly or indirectly. But manually querying such a huge data is challenging for most of the people. We have proposed an indexing method where from tokens of query we can find related frames and videos in a short amount of time. Queries could be very specific e.g. “car accident in area * at time*” Thus video summarization is used to produce short summary of long videos. Manually checking these summaries is again challenging task. This work aims to generate algorithm to query any video database in fraction of time, with few keywords related to video. Aim of generating a comprehensive algorithm is achieved by creating summary of each video. And the aim is also to achieve to Generate caption for each frame and then generating cumulative summary from summarized video frames using LSTM. This is achieved by searching user keywords in these summaries and ranking related summaries using CNN.
Convolutions of String Data Structures: A Case Study of Finding Number of Palindrome Substrings in a String Kamal Kumar, Anuj Kumar Jain, Raj Gaurang Tiwari, Nitin Jain, Deden Witarsyah, and Alok Misra IEEE In computer science, substring search or string Matching is a vulnerable problem when text resources are very large. Productivity of diverse scraping applications depend on the effectiveness of searching algorithms. Palindrome string finding is one of these problems which has its use in many applications. The problem is about finding the number of palindrome substrings within a string. This paper discusses a naive and dynamic programming-based algorithm for finding several palindrome strings. The time complexity of algorithm has also been analyzed and an effort has been made to improve it.
Machine Learning-Based Detection of Cardiovascular Disease using Classification and Feature Selection Anuj Kumar Jain, Kamal Kumar, Raj Gaurang Tiwari, Nitin Jain, Vinay Gautam, and Naresh Kumar Trivedi IEEE The application of ameliorative datasets to foretell chronic diseases has grasped considerable attention amongst data scientists all around the globe. The study is based on the Cardiovascular disease dataset. Cardiovascular disease is a broad term concerning disorders of the heart and blood vessels and is ranked to be the topmost cause of demise worldwide. Various feature selection techniques have been applied to each of the given classification algorithms and the features which give the best accuracy are then utilized for that particular algorithm.
Analysis of API Architecture: A Detailed Report Kamal Kumar, Anuj Kumar Jain, Raj Gaurang Tiwari, Nitin Jain, Vinay Gautam, and Naresh Kumar Trivedi IEEE Web services provided by various companies in the form of various APIs (Application Programming Interfaces) have increased to a very large extent in recent years. Service Oriented Architectures (SOA) and a message exchange protocol SOAP (Simple Object Access Protocol) have become prominent methods to serve the data exchange process for web services. But Recent reports state that a large number of companies are migrating from SOAP-based architecture to Representational State Transfer or REST. REST architecture is a resourceoriented architecture where resources are the entities of the system. The APIs are implemented around these resources. Also, REST architecture is stateless architecture as the client and server do not store. RESTful services return resources in some representation format. REST is a way by which a company provides a method to use its services that is provided by them exclusively. Various websites that were SOAP oriented went to REST implementation to simplify and modernize their operations. It was found that APIs were using simple HTTP verbs with no headers. REST web APIs are developed using HTTP protocol. Therefore, it makes use of all the HTTP properties like HTTP verbs, HTTP responses, header information, etc. The work presented here elaborates on the analysis of APIs using HTTP protocols and how the APIs are validated.
RECENT SCHOLAR PUBLICATIONS
Deep learning based automated vein recognition using swin transformer and super graph glue model K Bhushan, S Singh, K Kumar, P Kumar Knowledge-Based Systems 310, 112929 2025
NIDS-DA: Detecting functionally preserved adversarial examples for network intrusion detection system using deep autoencoders V Kumar, K Kumar, M Singh, N Kumar Expert Systems with Applications, 126513 2025
BMST-Net: bidirectional multi-scale spatiotemporal network for salient object detection in videos G Sharma, M Singh, SC Kumain, K Kumar Signal, Image and Video Processing 19 (1), 1-9 2025
Fuzzy decision-making: Introduction and features R Arora, G Kaur, K Kumar Strategic Fuzzy Extensions and Decision-making Techniques, 1-8 2025
NIDS-CBAD: detecting adversarial attacks in network intrusion detection systems using domain constraints V Kumar, K Kumar, M Singh International Journal of Machine Learning and Cybernetics, 1-22 2024
Four-body-fragmentation dynamics of acetylene by highly-charged-ion impact K Kumar, MAKA Siddiki, J Mukherjee, H Singh, D Misra Physical Review A 110 (6), 062811 2024
Development of petroleum-derived polymeric additive to enhance the bituminous properties with the use of a machine-learning model M Awasthi, V Joshi, R Upadhyay, A Kukrety, AK Verma, P Kumar, ... Sustainable Chemistry for the Environment 8, 100186 2024
Multi-attribute group decision making based on p,q-quasirung orthopair fuzzy Yager prioritized weighted geometric aggregation operator of p,q-quasirung A Redhu, K Kumar Granular Computing 9 (4), 75 2024
Multiattribute decision-making based on TOPSIS technique and novel correlation coefficient of q-rung orthopair fuzzy sets V Patel, H Kumar, A Redhu, K Kumar Granular Computing 9 (4), 74 2024
Strain-induced bandgap engineering in 2D ψ-graphene materials: a first-principles study K Kumar, NH de Leeuw, J Adam, AK Mishra Beilstein Journal of Nanotechnology 15 (1), 1440-1452 2024
PINK1 controls RTN3L-mediated ER autophagy by regulating peripheral tubule junctions R Chidambaram, K Kumar, S Parashar, G Ramachandran, S Chen, ... Journal of Cell Biology 223 (12), e202407193 2024
Pristine and Cu-decorated ψ-graphene-based materials for CO2 activation: A DFT (D)+ U Study K Kumar, N de Leeuw, J Adam, AK Mishra 2024
A Comprehensive Image Processing Framework for Early Diagnosis of Diabetic Retinopathy. K Yadav, Y Alharbi, EJ Alreshidi, A Alreshidi, AK Jain, A Jain, K Kumar, ... Computers, Materials & Continua 81 (2) 2024
Interoperability: A Conceptual Framework S Chaurasia, K Kumar Reshaping Intelligent Business and Industry: Convergence of AI and IoT at 2024
Harnessing bright upconversion luminescence from PEG-coated NaGdF4: Ho3+/Yb3+ nanoparticles for contrast enhancement in dual mode OCT imaging and optical temperature sensing K Shwetabh, A Banerjee, R Poddar, K Kumar Ceramics International 50 (19), 34956-34965 2024
Chromium adsorption efficiency by functional polymeric nanocomposite membrane: A case study for environmental sustainability S Prabhakar, JP Singh, K Kumar, SG Prasad, D Roy Polymer Engineering & Science 64 (10), 4774-4785 2024
Process for the preparation of higher-grade vg bitumens using sulfur-based polymeric additives (sbpa) T Senthilkumar, K Kumar, A Kumar, V Joshi, U Kumar, SK Ganguly, ... US Patent App. 18/612,469 2024
Synthesis of polyesters derived from glycerol and phthalic anhydride and its application for bitumen modification V Joshi, K Kumar, K Nandu Krishna, T Senthilkumar Journal of Applied Polymer Science 141 (36), e55924 2024
Hindrance of complete fusion in O + Ho reaction at above the barrier energies: Role of angular momentum M Gull, S Ali, K Kumar, IA Rizvi, A Agarwal, M Kumar, S Prajapati, ... The European Physical Journal Plus 139 (9), 825 2024
Rainfall analysis using FUCOM weighted logarithmic distance measure based on probabilistic dual hesitant preference values Rohit, K Kumar, R Bhardwaj, G Kaur Water Resources Management, 1-20 2024
MOST CITED SCHOLAR PUBLICATIONS
An advanced study on the similarity measures of intuitionistic fuzzy sets based on the set pair analysis theory and their application in decision making H Garg, K Kumar Soft Computing 22 (15), 4959-4970 2018 Citations: 295
TOPSIS method based on the connection number of set pair analysis under interval-valued intuitionistic fuzzy set environment K Kumar, H Garg Computational and Applied Mathematics 37, 1319-1329 2018 Citations: 295
Linguistic interval-valued atanassov intuitionistic fuzzy sets and their applications to group decision making problems H Garg, K Kumar IEEE Transactions on Fuzzy Systems 27 (12), 2302-2311 2019 Citations: 241
A novel exponential distance and its based TOPSIS method for interval-valued intuitionistic fuzzy sets using connection number of SPA theory H Garg, K Kumar Artificial Intelligence Review 53, 595-624 2020 Citations: 209
Connection number of set pair analysis based TOPSIS method on intuitionistic fuzzy sets and their application to decision making K Kumar, H Garg Applied Intelligence 48, 2112-2119 2018 Citations: 189
DOVIS: an implementation for high-throughput virtual screening using AutoDock S Zhang, K Kumar, X Jiang, A Wallqvist, J Reifman BMC bioinformatics 9, 1-4 2008 Citations: 143
Distance measures for connection number sets based on set pair analysis and its applications to decision-making process H Garg, K Kumar Applied Intelligence 48, 3346-3359 2018 Citations: 133
DOVIS 2.0: an efficient and easy to use parallel virtual screening tool based on AutoDock 4.0 X Jiang, K Kumar, X Hu, A Wallqvist, J Reifman Chemistry Central Journal 2, 1-7 2008 Citations: 126
Some aggregation operators for linguistic intuitionistic fuzzy set and its application to group decision-making process using the set pair analysis H Garg, K Kumar Arabian Journal for Science and Engineering 43, 3213-3227 2018 Citations: 117
Elevated serum cancer antigen 125 levels in advanced abdominal tuberculosis V Thakur, U Mukherjee, K Kumar Medical Oncology 18, 289-291 2001 Citations: 100
Properties of high modulus PEEK yarns for aerospace applications RI Shekar, TM Kotresh, PMD Rao, K Kumar Journal of applied polymer science 112 (4), 2497-2510 2009 Citations: 98
Improved possibility degree method for ranking intuitionistic fuzzy numbers and their application in multiattribute decision-making H Garg, K Kumar Granular Computing 4, 237-247 2019 Citations: 93
Multiple attribute group decision making based on advanced linguistic intuitionistic fuzzy weighted averaging aggregation operator of linguistic intuitionistic fuzzy numbers K Kumar, SM Chen Information Sciences 587, 813-824 2022 Citations: 91
Mapping of groundwater potential zones using the fuzzy analytic hierarchy process and geospatial technique AK Chaudhry, K Kumar, MA Alam Geocarto International 36 (20), 2323-2344 2021 Citations: 84
AREOR–Adaptive ranking based energy efficient opportunistic routing scheme in Wireless Sensor Network P Chithaluru, R Tiwari, K Kumar Computer Networks 162, 106863 2019 Citations: 83
A novel correlation coefficient of intuitionistic fuzzysets based on the connection number of set pair analysis and its application H Garg, K Kumar Scientia Iranica 25 (4), 2373-2388 2018 Citations: 82
Oroxylum indicum–a medicinal plant of North East India: An overview of its nutritional, remedial, and prophylactic properties DC Deka, V Kumar, C Prasad, K Kumar, BJ Gogoi, L Singh, RB Srivastava Journal of Applied Pharmaceutical Science 3 (4,), S104-S112 2013 Citations: 82
Systems level analysis and identification of pathways and networks associated with liver fibrosis MDM AbdulHameed, GJ Tawa, K Kumar, DL Ippolito, JA Lewis, ... PloS one 9 (11), e112193 2014 Citations: 80
A novel possibility measure to interval-valued intuitionistic fuzzy set using connection number of set pair analysis and its applications H Garg, K Kumar Neural Computing and Applications 32, 3337-3348 2020 Citations: 74
Manufacturing/remanufacturing based supply chain management under advertisements and carbon emissions process S Kumar, M Sigroha, K Kumar, B Sarkar RAIRO-Operations Research 56 (2), 831-851 2022 Citations: 64