Melwin D. Souza
Research Subjects
Melwin D. Souza has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Ananth Prabhu Gurpur
Sahyadri College of Engineering and Management, Visvesvaraya Technological University, Belagavi, Adyar, India
3 shared publications - Chaithra2 shared publications
- Nayana Yadav M
Research Scholar Department of Computer Science and Engineering Sahyadri College of Engineering & Management Karnataka India
2 shared publications - Ananth Prabhu Gurpur
Moodlakatte Institute of Technology
1 shared publication - Arun K. Kadu
Department of Chemistry, University of Mumbai, Mumbai, 400098, India
1 shared publication G. Ananth Prabhu
Department of Computer Science & Engineering, Sahyadri College of Engineering & Management, Mangalore, India
1 shared publication
Latest Publications
- 1 January 2025
- Published by American Scientific Publishing Group (ASPG) LLC in Fusion: Practice and Applications
- Vol. 20 (2) , 229-29-248
Early Detection of Diseases in Hydroponic Saffron Crops Using a Diffused Concurrent Convolution Neural Network for Smart Farming
American scientific publishing group
- 1 January 2025
- Published by American Scientific Publishing Group (ASPG) LLC in Fusion: Practice and Applications
- Vol. 20 (2) , 229-29-248
Early Detection of Diseases in Hydroponic Saffron Crops Using a Diffused Concurrent Convolution Neural Network for Smart Farming
American scientific publishing group
- 20 December 2024
- Published by Springer Nature in SN Computer Science
- Vol. 6 (1) , 1-12
Advanced Breast Cancer Detection Using Spatial Attention and Neural Architecture Search (SANAS-Net)
- 20 December 2024
- Published by Springer Nature in SN Computer Science
- Vol. 6 (1) , 1-12
Advanced Breast Cancer Detection Using Spatial Attention and Neural Architecture Search (SANAS-Net)
- Open Access
- 4 November 2024
- Published by Dewspublication
- Vol. 20 (S14) , 1756-1785
Integrating AI with Cybersecurity: A Review of Deep Learning for Anomaly Detection and Threat Mitigation
The rapidly evolving landscape of cyber threats poses significant challenges to traditional security measures, necessitating more advanced and adaptive approaches to anomaly detection and threat mitigation. This review paper explores innovative hybrid deep learning techniques that aim to address the limitations of existing cybersecurity solutions. Current approaches often struggle with the increasing sophistication of attacks, the expanding attack surface due to Internet of Things (IoT) and cloud adoption, and the overwhelming volume and velocity of network data. Moreover, traditional machine learning models frequently fall short in detecting novel threats, adapting to evolving attack patterns, and providing explainable results—critical factors in effective cybersecurity management. The review covers a spectrum of innovations, including: (1) ensemble methods that improve generalization and robustness against adversarial attacks; (2) hybrid deep learning models that excel in analyzing both spatial and temporal aspects of network behaviour; (3) autoencoder-based anomaly detection integrated with supervised classifiers for improved threat categorization; and (4) reinforcement learning-enhanced systems for dynamic, adaptive defence strategies. We also explore the application of explainable AI techniques to hybrid models, addressing the critical need for interpretability in security decisions.
- 4 November 2024
- Published by Dewspublication
- Vol. 20 (S14) , 1756-1785
Open AccessIntegrating AI with Cybersecurity: A Review of Deep Learning for Anomaly Detection and Threat Mitigation
The rapidly evolving landscape of cyber threats poses significant challenges to traditional security measures, necessitating more advanced and adaptive approaches to anomaly detection and threat mitigation. This review paper explores innovative hybrid deep learning techniques that aim to address the limitations of existing cybersecurity solutions. Current approaches often struggle with the increasing sophistication of attacks, the expanding attack surface due to Internet of Things (IoT) and cloud adoption, and the overwhelming volume and velocity of network data. Moreover, traditional machine learning models frequently fall short in detecting novel threats, adapting to evolving attack patterns, and providing explainable results—critical factors in effective cybersecurity management. The review covers a spectrum of innovations, including: (1) ensemble methods that improve generalization and robustness against adversarial attacks; (2) hybrid deep learning models that excel in analyzing both spatial and temporal aspects of network behaviour; (3) autoencoder-based anomaly detection integrated with supervised classifiers for improved threat categorization; and (4) reinforcement learning-enhanced systems for dynamic, adaptive defence strategies. We also explore the application of explainable AI techniques to hybrid models, addressing the critical need for interpretability in security decisions.
- Open Access
- 4 November 2024
- Published by Dewspublication in Nanotechnology Perceptions
- p. 1756-1785
Integrating AI with Cybersecurity: A Review of Deep Learning for Anomaly Detection and Threat Mitigation
The rapidly evolving landscape of cyber threats poses significant challenges to traditional security measures, necessitating more advanced and adaptive approaches to anomaly detection and threat mitigation. This review paper explores innovative hybrid deep learning techniques that aim to address the limitations of existing cybersecurity solutions. Current approaches often struggle with the increasing sophistication of attacks, the expanding attack surface due to Internet of Things (IoT) and cloud adoption, and the overwhelming volume and velocity of network data. Moreover, traditional machine learning models frequently fall short in detecting novel threats, adapting to evolving attack patterns, and providing explainable results—critical factors in effective cybersecurity management. The review covers a spectrum of innovations, including: (1) ensemble methods that improve generalization and robustness against adversarial attacks; (2) hybrid deep learning models that excel in analyzing both spatial and temporal aspects of network behaviour; (3) autoencoder-based anomaly detection integrated with supervised classifiers for improved threat categorization; and (4) reinforcement learning-enhanced systems for dynamic, adaptive defence strategies. We also explore the application of explainable AI techniques to hybrid models, addressing the critical need for interpretability in security decisions.
- 4 November 2024
- Published by Dewspublication in Nanotechnology Perceptions
- p. 1756-1785
Open AccessIntegrating AI with Cybersecurity: A Review of Deep Learning for Anomaly Detection and Threat Mitigation
The rapidly evolving landscape of cyber threats poses significant challenges to traditional security measures, necessitating more advanced and adaptive approaches to anomaly detection and threat mitigation. This review paper explores innovative hybrid deep learning techniques that aim to address the limitations of existing cybersecurity solutions. Current approaches often struggle with the increasing sophistication of attacks, the expanding attack surface due to Internet of Things (IoT) and cloud adoption, and the overwhelming volume and velocity of network data. Moreover, traditional machine learning models frequently fall short in detecting novel threats, adapting to evolving attack patterns, and providing explainable results—critical factors in effective cybersecurity management. The review covers a spectrum of innovations, including: (1) ensemble methods that improve generalization and robustness against adversarial attacks; (2) hybrid deep learning models that excel in analyzing both spatial and temporal aspects of network behaviour; (3) autoencoder-based anomaly detection integrated with supervised classifiers for improved threat categorization; and (4) reinforcement learning-enhanced systems for dynamic, adaptive defence strategies. We also explore the application of explainable AI techniques to hybrid models, addressing the critical need for interpretability in security decisions.
- Open Access
- 1 September 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (3) , 3339-3349
SANAS-Net: spatial attention neural architecture search for breast cancer detection
The utilization of mammography images plays a vital role in the prompt detection and treatment of breast cancer. Breast imaging techniques aid medical professionals in assessing the dimensions, morphology, and spatial orientation of breast lesions, facilitating the differentiation between benign and malignant conditions. Breast tissue can vary widely in terms of density, composition, and structure, leading to complexities in distinguishing between benign and malignant conditions. The primary contribution of this paper is the proposal of a spatial attention-based neural architecture search network (SANAS-Net) technique that incorporates a spatial attention mechanism, enabling the model to learn and prioritize key regions within mammograms (MMs). Multi-head attention is employed within the transformer blocks to effectively capture a wide range of spatial relations and feature interactions. Global contextual information was integrated into the transformer blocks by means of introducing positional embeddings. Several practical studies have been undertaken to verify the effectiveness of our methodology in identifying fully attentive networks that exhibit good performance in distinguishing between malignant and benign breast cancer cases. The experimental study reached a test accuracy of 89.95%, which is way higher than previously proposed algorithms for mammography imagebased breast cancer detection.
- 1 September 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (3) , 3339-3349
Open AccessSANAS-Net: spatial attention neural architecture search for breast cancer detection
The utilization of mammography images plays a vital role in the prompt detection and treatment of breast cancer. Breast imaging techniques aid medical professionals in assessing the dimensions, morphology, and spatial orientation of breast lesions, facilitating the differentiation between benign and malignant conditions. Breast tissue can vary widely in terms of density, composition, and structure, leading to complexities in distinguishing between benign and malignant conditions. The primary contribution of this paper is the proposal of a spatial attention-based neural architecture search network (SANAS-Net) technique that incorporates a spatial attention mechanism, enabling the model to learn and prioritize key regions within mammograms (MMs). Multi-head attention is employed within the transformer blocks to effectively capture a wide range of spatial relations and feature interactions. Global contextual information was integrated into the transformer blocks by means of introducing positional embeddings. Several practical studies have been undertaken to verify the effectiveness of our methodology in identifying fully attentive networks that exhibit good performance in distinguishing between malignant and benign breast cancer cases. The experimental study reached a test accuracy of 89.95%, which is way higher than previously proposed algorithms for mammography imagebased breast cancer detection.
- Open Access
- 31 July 2023
- Published by Asian Publication Corporation in Asian Journal of Chemistry
- Vol. 35 (8) , 1963-1968
Synthesis, Spectral and Theoretical Characterization and Antimicrobial and Cytotoxic Studies of Some Transition Metal Complexes of 4-[(2E)-2-Benzylidenehydrazinyl]-7H-pyrrolo[2,3-d]pyrimidine
Present work describes the synthesis and characterization of complexes of 4-[(2E)-2-benzylidenehydrazinyl]-7H-pyrrolo[2,3-d]pyrimidine with Fe(II), Mn(II), Co(II), Ni(II), Pd(II), Zn(II), Cu(II), Cd(II) and Hg(II). The results of the elemental and spectroscopic analyses suggest a 1:2 metal-ligand ratio, [ML2]. The electronic spectra and magnetic moments of the metal(II) complexes suggest that Fe(II), Mn(II), Co(II) and Ni(II) complexes are in an octahedral geometry, while Pd(II) in a square planar ligand geometry. In contrast, the Cu(II), Zn(II), Cd(II) and Hg(II) are in tetrahedral geometry. The conductance measurements in nitrobenzene suggest the non-electrolytic nature of all the metal-coordination compounds. The tested biological species viz. Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Bacillus cereus, Proteus mirabilis, Klebsiella oxytoca, Aspergillus niger, A. flavus and Rhizopus stolonifer exhibited the moderate to strong antimicrobial activity. At 10 mg/mL concentration in nitrobenzene, cobalt(II) complex is observed to be a more promising antimicrobial than any other metal(II) complexes synthesized. The antioxidant potential of the synthesized metal(II) compounds was also evaluated using 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical scavenging assay and shows the high DPPH radical scavenging activity. The nickel complex had the highest DPPH radical scavenging activity compared to the other metal(II) complexes.
- 31 July 2023
- Published by Asian Publication Corporation in Asian Journal of Chemistry
- Vol. 35 (8) , 1963-1968
Open AccessSynthesis, Spectral and Theoretical Characterization and Antimicrobial and Cytotoxic Studies of Some Transition Metal Complexes of 4-[(2E)-2-Benzylidenehydrazinyl]-7H-pyrrolo[2,3-d]pyrimidine
Present work describes the synthesis and characterization of complexes of 4-[(2E)-2-benzylidenehydrazinyl]-7H-pyrrolo[2,3-d]pyrimidine with Fe(II), Mn(II), Co(II), Ni(II), Pd(II), Zn(II), Cu(II), Cd(II) and Hg(II). The results of the elemental and spectroscopic analyses suggest a 1:2 metal-ligand ratio, [ML2]. The electronic spectra and magnetic moments of the metal(II) complexes suggest that Fe(II), Mn(II), Co(II) and Ni(II) complexes are in an octahedral geometry, while Pd(II) in a square planar ligand geometry. In contrast, the Cu(II), Zn(II), Cd(II) and Hg(II) are in tetrahedral geometry. The conductance measurements in nitrobenzene suggest the non-electrolytic nature of all the metal-coordination compounds. The tested biological species viz. Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Bacillus cereus, Proteus mirabilis, Klebsiella oxytoca, Aspergillus niger, A. flavus and Rhizopus stolonifer exhibited the moderate to strong antimicrobial activity. At 10 mg/mL concentration in nitrobenzene, cobalt(II) complex is observed to be a more promising antimicrobial than any other metal(II) complexes synthesized. The antioxidant potential of the synthesized metal(II) compounds was also evaluated using 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical scavenging assay and shows the high DPPH radical scavenging activity. The nickel complex had the highest DPPH radical scavenging activity compared to the other metal(II) complexes.