Manohar Naik Sugali
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Manohar Naik Sugali has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Ashfaq Ahmad Najar
School of Computing Science and Engineering (SCOPE) VIT Bhopal University Sehore Madhya Pradesh India
5 shared publications - Hashida Haidros Rahima Manzil
Department of Computer Science, Central University of Kerala, Kasaragod, India
3 shared publications - Pamena Akanksha
Central University of Kerala
3 shared publications - Sruthi Krishnan
Department of Computer Science Central University of Kerala Periye, Kasaragod Kerala India
3 shared publications - Vanlalruata Hnamte
Mizoram University,Department of Mathematics and Computer Science,Aizawl,India
2 shared publications - Akanksha Pamena1 shared publication
Latest Publications
- 23 October 2026
- Published by Wiley in Security and Privacy
- Vol. 9 (6) , e70259
A Novel Consensus Ranking‐Based Ensemble Feature Selection for Efficient Phishing Attack Detection
Phishing attacks continue to represent a serious and evolving threat to individuals, businesses, and the wider field of cybersecurity. Despite advancements in machine learning‐based detection methods, the need to rely on large numbers of features and external data sources leads to various problems, such as greater computational complexity, reduced scalability, and higher latency. Because of these limitations, such systems are not feasible to implement in real world situations. In this study, a novel framework is presented which makes use of lightweight, third party independent features that have been optimized by means of an ensemble feature selection and consensus ranking procedure. The framework is able to efficiently find the most discriminative features by taking advantage of the individual advantages of filter, wrapper, and embedded methods. By using consensus ranking aggregation, this method reduces the biases inherent in each technique and at the same time improves the robustness and generalizability of the feature selection process. The framework was tested using the comprehensive and up to date PhiUSIIL Phishing URL Dataset, which includes more than 235 000 URLs, thus allowing for a thorough evaluation of its performance. We chose the following machine learning models, Random Forest, Decision Tree, Naive Bayes, and Logistic Regression, in order to assess the effectiveness of the best feature set. The reason for selecting these models rather than more advanced deep learning models is that they offer better interpretability, are more computationally efficient, have faster inference speeds, and require fewer resources. Our experiments demonstrated strong performance in detecting phishing URLs, indicating the effectiveness of the ensemble feature selection method and its promise for use in scalable real time detection of malicious websites.
- 23 October 2026
- Published by Wiley in Security and Privacy
- Vol. 9 (6) , e70259
A Novel Consensus Ranking‐Based Ensemble Feature Selection for Efficient Phishing Attack Detection
Phishing attacks continue to represent a serious and evolving threat to individuals, businesses, and the wider field of cybersecurity. Despite advancements in machine learning‐based detection methods, the need to rely on large numbers of features and external data sources leads to various problems, such as greater computational complexity, reduced scalability, and higher latency. Because of these limitations, such systems are not feasible to implement in real world situations. In this study, a novel framework is presented which makes use of lightweight, third party independent features that have been optimized by means of an ensemble feature selection and consensus ranking procedure. The framework is able to efficiently find the most discriminative features by taking advantage of the individual advantages of filter, wrapper, and embedded methods. By using consensus ranking aggregation, this method reduces the biases inherent in each technique and at the same time improves the robustness and generalizability of the feature selection process. The framework was tested using the comprehensive and up to date PhiUSIIL Phishing URL Dataset, which includes more than 235 000 URLs, thus allowing for a thorough evaluation of its performance. We chose the following machine learning models, Random Forest, Decision Tree, Naive Bayes, and Logistic Regression, in order to assess the effectiveness of the best feature set. The reason for selecting these models rather than more advanced deep learning models is that they offer better interpretability, are more computationally efficient, have faster inference speeds, and require fewer resources. Our experiments demonstrated strong performance in detecting phishing URLs, indicating the effectiveness of the ensemble feature selection method and its promise for use in scalable real time detection of malicious websites.
- 2 January 2026
- Published by Springer Nature
- p. 335-360
Phishing Detection: Exploring Feature Representation and Defense Strategies
- 2 January 2026
- Published by Springer Nature
- p. 335-360
Phishing Detection: Exploring Feature Representation and Defense Strategies
- Open Access
- 24 December 2025
- Published by NASK National Research Institute in Applied Cybersecurity & Internet Governance
SAE-AM: Enhancing Network Intrusion Detection Using Sparse Autoencoders with Attention Modules
In response to the relentless evolution of cyber threats that continue to outpace traditional defence mechanisms, this study addresses key limitations in existing Intrusion Detection Systems (IDS), particularly those related to high dimensionality and computational inefficiency. We propose a novel framework, sparse autoencoders with attention modules (SAE-AM), which integrates SAE with both channel and positional attention mechanisms to enhance feature representation and optimise resource utilisation. While SAE effectively perform dimensionality reduction, the attention modules capture global dependencies across diverse input features. Leveraging a deep learning model, specifically a multi-layer perceptron classifier – our framework efficiently classifies normal and attack samples. Extensive evaluations on benchmark network intrusion datasets, CICIDS2017, NSL-KDD, and UNSW-NB15, demonstrate the robustness and superior performance of the proposed method, achieving 99% accuracy and minimal false alarm rates across all datasets. SAE-AM makes significant strides in overcoming the core challenges of NIDS by reducing dimensionality and improving computational efficiency. This novel approach contributes to enhanced network security, offering a lightweight and effective solution for real-world intrusion detection scenarios.
- 24 December 2025
- Published by NASK National Research Institute in Applied Cybersecurity & Internet Governance
Open AccessSAE-AM: Enhancing Network Intrusion Detection Using Sparse Autoencoders with Attention Modules
In response to the relentless evolution of cyber threats that continue to outpace traditional defence mechanisms, this study addresses key limitations in existing Intrusion Detection Systems (IDS), particularly those related to high dimensionality and computational inefficiency. We propose a novel framework, sparse autoencoders with attention modules (SAE-AM), which integrates SAE with both channel and positional attention mechanisms to enhance feature representation and optimise resource utilisation. While SAE effectively perform dimensionality reduction, the attention modules capture global dependencies across diverse input features. Leveraging a deep learning model, specifically a multi-layer perceptron classifier – our framework efficiently classifies normal and attack samples. Extensive evaluations on benchmark network intrusion datasets, CICIDS2017, NSL-KDD, and UNSW-NB15, demonstrate the robustness and superior performance of the proposed method, achieving 99% accuracy and minimal false alarm rates across all datasets. SAE-AM makes significant strides in overcoming the core challenges of NIDS by reducing dimensionality and improving computational efficiency. This novel approach contributes to enhanced network security, offering a lightweight and effective solution for real-world intrusion detection scenarios.
- 21 November 2025
- Published by Wiley in Security and Privacy
- Vol. 9 (1)
LIDS: A Novel Lightweight Intrusion Detection System for DDoS Attacks
In this study, we introduce, a novel Lightweight Intrusion Detection System (LIDS) specifically designed to address this critical trade‐off between detection performance and computational efficiency. LIDS is designed for ultra‐low‐latency DDoS detection and distinguishes itself through a unique three‐stage architecture. The pipeline begins with Principal Component Analysis (PCA)‐based feature reduction to mitigate the curse of dimensionality, extract the most informative features from network traffic data, and reduce the input complexity for the classifier. Each inference in an LIDS operates on a single network packet represented by its extracted flow‐level features, enabling real‐time, per‐packet decision‐making with ultra‐low latency. This is followed by a weighted sampling technique to effectively handle the severe class imbalance typically present in network security datasets, preventing the model from being biased towards benign traffic. Finally, a lightweight model was fitted with minimal parameters for attack detection. We conducted a comprehensive evaluation of the LIDS on three diverse benchmark datasets: CICDDoS2019, TON_IoT, and BoT‐IoT, and evaluated its performance on both binary (attack vs. benign) and multi‐class (specific attack type) classification tasks. The LIDS consistently achieved a detection accuracy of over 99% and F1‐scores exceeding 99% across all datasets for binary classification. Additionally, the multi‐class classification achieved a detection accuracy of over 95% and F1‐scores exceeding 95%. Crucially, the system demonstrates exceptional efficiency, with an end‐to‐end inference latency below 2 per packet when tested on commodity edge‐like hardware and a final model size of less than 300 KB. To further bolster the model's practicality, we provide an interpretability analysis using SHapley Additive exPlanations (SHAP) to identify the key features driving its predictions and conduct an adversarial robustness evaluation to confirm its resilience.
- 21 November 2025
- Published by Wiley in Security and Privacy
- Vol. 9 (1)
LIDS: A Novel Lightweight Intrusion Detection System for DDoS Attacks
In this study, we introduce, a novel Lightweight Intrusion Detection System (LIDS) specifically designed to address this critical trade‐off between detection performance and computational efficiency. LIDS is designed for ultra‐low‐latency DDoS detection and distinguishes itself through a unique three‐stage architecture. The pipeline begins with Principal Component Analysis (PCA)‐based feature reduction to mitigate the curse of dimensionality, extract the most informative features from network traffic data, and reduce the input complexity for the classifier. Each inference in an LIDS operates on a single network packet represented by its extracted flow‐level features, enabling real‐time, per‐packet decision‐making with ultra‐low latency. This is followed by a weighted sampling technique to effectively handle the severe class imbalance typically present in network security datasets, preventing the model from being biased towards benign traffic. Finally, a lightweight model was fitted with minimal parameters for attack detection. We conducted a comprehensive evaluation of the LIDS on three diverse benchmark datasets: CICDDoS2019, TON_IoT, and BoT‐IoT, and evaluated its performance on both binary (attack vs. benign) and multi‐class (specific attack type) classification tasks. The LIDS consistently achieved a detection accuracy of over 99% and F1‐scores exceeding 99% across all datasets for binary classification. Additionally, the multi‐class classification achieved a detection accuracy of over 95% and F1‐scores exceeding 95%. Crucially, the system demonstrates exceptional efficiency, with an end‐to‐end inference latency below 2 per packet when tested on commodity edge‐like hardware and a final model size of less than 300 KB. To further bolster the model's practicality, we provide an interpretability analysis using SHapley Additive exPlanations (SHAP) to identify the key features driving its predictions and conduct an adversarial robustness evaluation to confirm its resilience.
- 4 September 2025
- Published by Springer Nature in Cluster Computing
- Vol. 28 (10) , 1-28
TCSO-KD: a novel technique for optimal feature selection and knowledge distillation for efficient IoT intrusion detection systems
- 4 September 2025
- Published by Springer Nature in Cluster Computing
- Vol. 28 (10) , 1-28
TCSO-KD: a novel technique for optimal feature selection and knowledge distillation for efficient IoT intrusion detection systems
- 21 August 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 329
A novel framework for effective phishing URL detection using an LSTM-based siamese network
- 21 August 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 329
A novel framework for effective phishing URL detection using an LSTM-based siamese network
- 1 August 2025
- Published by Springer Nature
- p. 327-336
Deep Learning and Explainable AI: A Dual Approach to Network Security
- 1 August 2025
- Published by Springer Nature
- p. 327-336
Deep Learning and Explainable AI: A Dual Approach to Network Security
- 26 November 2024
- Published by Springer Nature in Cluster Computing
- Vol. 28 (2) , 1-28
AE-CIAM: a hybrid AI-enabled framework for low-rate DDoS attack detection in cloud computing
- 26 November 2024
- Published by Springer Nature in Cluster Computing
- Vol. 28 (2) , 1-28
AE-CIAM: a hybrid AI-enabled framework for low-rate DDoS attack detection in cloud computing
- 22 November 2024
- Published by Springer Nature
- p. 57-67
Network Intrusion Detection with SMOTE-ENN and Deep Learning Techniques
- 22 November 2024
- Published by Springer Nature
- p. 57-67
Network Intrusion Detection with SMOTE-ENN and Deep Learning Techniques
- 20 February 2024
- Published by Springer Nature in SN Computer Science
- Vol. 5 (3) , 1-12
A DNA Cryptosystem Using Diffie–Hellman Key Exchange
- 20 February 2024
- Published by Springer Nature in SN Computer Science
- Vol. 5 (3) , 1-12
A DNA Cryptosystem Using Diffie–Hellman Key Exchange