Durga Prasad Mohapatra
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Durga Prasad Mohapatra has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
- Sangharatna Godboley
National Institute of Technology Warangal
10 shared publications - Madhusmita Sahu
National Institute of Technology
7 shared publications - Swadhin Kumar Barisal
Computer Science and Engineering, ITER, Siksha ‘O’ Anusandhan Deemed to be University, Bhubaneswar, India
5 shared publications - Arpita Dutta
National University of Singapore
4 shared publications - Arup Abhinna Acharya
KIIT Deemed to be University
4 shared publications - Prafulla Kumar Behera
Indian Institute of Technology Madras
4 shared publications
Latest Publications
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Behavioral Heatmap-Driven Vision Transformer Framework for Malware Classification
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Behavioral Heatmap-Driven Vision Transformer Framework for Malware Classification
- 17 March 2026
- Published by Springer Nature in Software Quality Journal
- Vol. 34 (2) , 10
An efficient blending model for classifying security-related software vulnerabilities into software development life cycle phases
- 17 March 2026
- Published by Springer Nature in Software Quality Journal
- Vol. 34 (2) , 10
An efficient blending model for classifying security-related software vulnerabilities into software development life cycle phases
- Top Cited Papers
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. PP (5) , 1-16
ADAPT-MIND: A Cross-Domain Deep Learning Framework for Mental Health Signal Detection in Shakespearean and Social Media Texts
Cross-domain detection of mental distress signals in historical and contemporary texts is very challenging because of linguistic heterogeneity and the limited annotation resources. Adapter-Augmented Distillation And Prompt-Tuning – Memory-Augmented INference for Distress (ADAPT-MIND) proposes a unified framework that integrates adapter-enhanced LLM embeddings, graph-attentive temporal convolutions, memory-augmented transformers, adversarial–contrastive distillation and mutual information regularization, and mixed meta-learned soft prompts. Shared pre-processing and feature augmentation are applied to both Shakespearean utterances (10% expert-annotated, $\kappa = 0.82$ ) and 7489 student social-media posts (80/10/10 split), including pronoun variance, affective bigrams, syntactic entropy, and graph positional encodings. On five-fold cross-validation, ADAPT-MIND achieves accuracies of 0.912 for Shakespearean and 0.857 for students, with F1 scores of 0.911 and 0.856 and AUC of 0.948 and 0.903, respectively. Extensive ablation studies confirm an F1 drop of up to 4.2% by removing the graph-temporal module, proving its critical role in capturing multiscale narrative context. Error analysis shows misunderstanding of archaic idioms is responsible for 31%, while the rest is mainly due to mistaken academic stress, totaling 27%. Hyper-parameter optimization utilizes hierarchical Bayesian search over loss weights, learning rates, adapter dimensions, and convolution depth. The proposed approach delivers state-of-the-art performance across domains while ensuring interpretability and efficient parameter adaptation.
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. PP (5) , 1-16
Top Cited PapersADAPT-MIND: A Cross-Domain Deep Learning Framework for Mental Health Signal Detection in Shakespearean and Social Media Texts
Cross-domain detection of mental distress signals in historical and contemporary texts is very challenging because of linguistic heterogeneity and the limited annotation resources. Adapter-Augmented Distillation And Prompt-Tuning – Memory-Augmented INference for Distress (ADAPT-MIND) proposes a unified framework that integrates adapter-enhanced LLM embeddings, graph-attentive temporal convolutions, memory-augmented transformers, adversarial–contrastive distillation and mutual information regularization, and mixed meta-learned soft prompts. Shared pre-processing and feature augmentation are applied to both Shakespearean utterances (10% expert-annotated, $\kappa = 0.82$ ) and 7489 student social-media posts (80/10/10 split), including pronoun variance, affective bigrams, syntactic entropy, and graph positional encodings. On five-fold cross-validation, ADAPT-MIND achieves accuracies of 0.912 for Shakespearean and 0.857 for students, with F1 scores of 0.911 and 0.856 and AUC of 0.948 and 0.903, respectively. Extensive ablation studies confirm an F1 drop of up to 4.2% by removing the graph-temporal module, proving its critical role in capturing multiscale narrative context. Error analysis shows misunderstanding of archaic idioms is responsible for 31%, while the rest is mainly due to mistaken academic stress, totaling 27%. Hyper-parameter optimization utilizes hierarchical Bayesian search over loss weights, learning rates, adapter dimensions, and convolution depth. The proposed approach delivers state-of-the-art performance across domains while ensuring interpretability and efficient parameter adaptation.
- 22 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-13
A Novel Deep Learning Framework with Behavioral Heatmap, for Malware Classification
The classification of malware continues to be a significant cybersecurity problem because malicious software can rapidly evolve obfuscation and evasion mechanism to disguise its real behaviour. Common static analysis techniques can often fail to detect malware that is polymorphic or metamorphic, meaning that code structure changes but malicious intent remain the same. This paper proposes a dynamic behaviour-based malware classification framework as a solution to these issues. It generates visually understandable Custom RGB (Inferno-inspired) heatmap images by using API call sequences and argument features captured during execution. These perceptually consistent heatmaps, which transition from deep blue to magenta to bright yellow, enhance the visual separability of subtle behavioural differences across malware families. Three deep learning architectures are implemented and evaluated: Existing CNN Architecture, Proposed EfficientNet Model and Proposed CNN Model that employed transfer learning, for fine-grained feature extraction. The accuracy results for tests on 22,056 samples dataset from eight malware families were 98.76%, 98.88% and 99.12% respectively. The results demonstrate how behavioural visualisation, in conjunction with hybrid and multi-stage deep learning architectures, significantly enhances malware classification accuracy and resilience to obfuscation-based attacks. This approach provides a vision-driven, interpretable approach for cybersecurity systems of the future.
- 22 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-13
A Novel Deep Learning Framework with Behavioral Heatmap, for Malware Classification
The classification of malware continues to be a significant cybersecurity problem because malicious software can rapidly evolve obfuscation and evasion mechanism to disguise its real behaviour. Common static analysis techniques can often fail to detect malware that is polymorphic or metamorphic, meaning that code structure changes but malicious intent remain the same. This paper proposes a dynamic behaviour-based malware classification framework as a solution to these issues. It generates visually understandable Custom RGB (Inferno-inspired) heatmap images by using API call sequences and argument features captured during execution. These perceptually consistent heatmaps, which transition from deep blue to magenta to bright yellow, enhance the visual separability of subtle behavioural differences across malware families. Three deep learning architectures are implemented and evaluated: Existing CNN Architecture, Proposed EfficientNet Model and Proposed CNN Model that employed transfer learning, for fine-grained feature extraction. The accuracy results for tests on 22,056 samples dataset from eight malware families were 98.76%, 98.88% and 99.12% respectively. The results demonstrate how behavioural visualisation, in conjunction with hybrid and multi-stage deep learning architectures, significantly enhances malware classification accuracy and resilience to obfuscation-based attacks. This approach provides a vision-driven, interpretable approach for cybersecurity systems of the future.
- 1 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Malware Classification with n-gram Based NLP and Cosine Similarity
Malware classification is a fundamental aspect of cybersecurity, essential for detecting and mitigating threats. This paper introduces a malware classification technique that applies Natural Language Processing (NLP) methods alongside Cosine similarity. Our approach involves using $n$-grams of API call sequences, including the API names, their arguments, and categories, to characterize malware behavior. By computing Cosine similarity between these $n$-grams, we effectively capture both similarities and differences in malware behavior. Our experimental results show that different $n$-gram configurations exhibit varying classification capabilities, with some proving more effective for specific types of malware. Overall, our technique presents a promising solution for malware classification, leveraging NLP and Cosine similarity to improve the accuracy and efficiency of malware variant detection.
- 1 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Malware Classification with n-gram Based NLP and Cosine Similarity
Malware classification is a fundamental aspect of cybersecurity, essential for detecting and mitigating threats. This paper introduces a malware classification technique that applies Natural Language Processing (NLP) methods alongside Cosine similarity. Our approach involves using $n$-grams of API call sequences, including the API names, their arguments, and categories, to characterize malware behavior. By computing Cosine similarity between these $n$-grams, we effectively capture both similarities and differences in malware behavior. Our experimental results show that different $n$-gram configurations exhibit varying classification capabilities, with some proving more effective for specific types of malware. Overall, our technique presents a promising solution for malware classification, leveraging NLP and Cosine similarity to improve the accuracy and efficiency of malware variant detection.
- 1 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Enhancing Cryptographic Misuse Detection in Source Code Using AST and Machine Learning Techniques
Correctly implementing cryptographic security in software is challenging due to its complexity. This paper introduces a machine learning framework using Abstract Syntax Trees (ASTs) to identify cryptographic API misuse in Java code. The approach includes two models: a Per-Category Model, classifying vulnerabilities into nine specific types, and a Full Model, performing binary (secure/insecure) classification. The Per-Category Model achieved an average accuracy of $80 \%$, effectively identifying issues, especially in Public Key Cryptography (PKC) and Weak Cryptography (WC). The Full Model reached 78% accuracy with an AUC-ROC of 0.87, showing strong overall performance. Compared to traditional static analysis tools detecting only 35% of known issues, our method significantly improves accuracy and reduces false alarms. Leveraging AST-based features and Random Forest classifiers, our framework enhances cryptographic misuse detection, providing developers clearer and more actionable insights, thus promoting secure software development.- 1 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Enhancing Cryptographic Misuse Detection in Source Code Using AST and Machine Learning Techniques
Correctly implementing cryptographic security in software is challenging due to its complexity. This paper introduces a machine learning framework using Abstract Syntax Trees (ASTs) to identify cryptographic API misuse in Java code. The approach includes two models: a Per-Category Model, classifying vulnerabilities into nine specific types, and a Full Model, performing binary (secure/insecure) classification. The Per-Category Model achieved an average accuracy of $80 \%$, effectively identifying issues, especially in Public Key Cryptography (PKC) and Weak Cryptography (WC). The Full Model reached 78% accuracy with an AUC-ROC of 0.87, showing strong overall performance. Compared to traditional static analysis tools detecting only 35% of known issues, our method significantly improves accuracy and reduces false alarms. Leveraging AST-based features and Random Forest classifiers, our framework enhances cryptographic misuse detection, providing developers clearer and more actionable insights, thus promoting secure software development.- 25 June 2025
- Published by Springer Nature in Sādhanā
- Vol. 50 (3) , 1-17
Customized convolutional neural network with explainable AI for multimodal breast cancer detection
- 25 June 2025
- Published by Springer Nature in Sādhanā
- Vol. 50 (3) , 1-17
Customized convolutional neural network with explainable AI for multimodal breast cancer detection
- 24 May 2025
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (20) , 15469-15499
MMHBC-Net: a multi-modal hybrid approach for breast cancer classification
- 24 May 2025
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (20) , 15469-15499
MMHBC-Net: a multi-modal hybrid approach for breast cancer classification
- 24 May 2025
- Published by Springer Nature
- p. 217-251
Empowering Software Security: Leveraging Machine Learning for Anomaly Detection and Threat Prevention
- 24 May 2025
- Published by Springer Nature
- p. 217-251
Empowering Software Security: Leveraging Machine Learning for Anomaly Detection and Threat Prevention
- 26 September 2024
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 120
SMUP: A technique to improve MC/DC using specified patterns
- 26 September 2024
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 120
SMUP: A technique to improve MC/DC using specified patterns