Thenmozhi E

Dr. Thenmozhi E


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 January 2026 in Future Internet
Verified
Journal Article
Open access
PDF file
SSSuresh K. S
Suresh K. S
TEThenmozhi E
Thenmozhi E
RRRadhakrishnan Rajamani
Radhakrishnan Rajamani
AKAnubhav Kumar
Anubhav Kumar
BBBalamurugan Balusamy
Balamurugan Balusamy
PKPrabu Kaliyaperumal
Prabu Kaliyaperumal
Cloud computing environments generate high-dimensional, large-scale, and highly dynamic network traffic, making intrusion diagnosis challenging due to evolving attack patterns, severe traffic imbalance, and limited availability of labeled data. To address these challenges, this study presents an unsupervised, cloud-centric intrusion diagnosis framework that integrates autoencoder-based representation learning with density-based attack categorization. A dual-stage autoencoder is trained exclusively on benign traffic to learn compact latent representations and to identify anomalous flows using reconstruction-error analysis, enabling effective anomaly detection without prior attack labels. The detected anomalies are subsequently grouped using density-based learning to uncover latent attack structures and support fine-grained multiclass intrusion diagnosis under varying attack densities. Experiments conducted on the large-scale CSE-CIC-IDS2018 dataset demonstrate that the proposed framework achieves an anomaly detection accuracy of 99.46%, with high recall and low false-negative rates in the optimal latent-space configuration. The density-based classification stage achieves an overall multiclass attack classification accuracy of 98.79%, effectively handling both majority and minority attack categories. Clustering quality evaluation reports a Silhouette Score of 0.9857 and a Davies–Bouldin Index of 0.0091, indicating strong cluster compactness and separability. Comparative analysis against representative supervised and unsupervised baselines confirms the framework’s scalability and robustness under highly imbalanced cloud traffic, highlighting its suitability for future Internet cloud security ecosystems.

26 September 2026 in Future Internet
Verified
Journal Article
Open access
PDF file
ASAshutosh Shankhdhar
Ashutosh Shankhdhar
VMVanitha Murugesan
Vanitha Murugesan
TEThenmozhi E
Thenmozhi E
SKSamia Kouki
Samia Kouki
SYSumendra Yogarayan
Sumendra Yogarayan
PKPrabu Kaliyaperumal
Prabu Kaliyaperumal
The increasing heterogeneity of Internet of Things (IoT) environments makes intrusion diagnosis challenging because device behaviours, traffic patterns, and attack distributions can vary across deployment conditions, while data-locality requirements limit centralized access to network data. This study proposes CA-AFiD (Context-Aware Adaptive Federated Intrusion Diagnosis), a framework designed to support adaptive and interpretable intrusion diagnosis across IoT–Fog–Cloud environments. CA-AFiD combines behaviour-aware representation learning using Transformer, BiLSTM, and attention mechanisms with a context-aware adaptive ensemble that adjusts learner contributions according to behavioural complexity, attack density, and device context. Federated learning is used to coordinate model updates across distributed Fog nodes without sharing raw traffic data, while attention-based interpretation, SHAP feature attribution, and ATT&CK-oriented contextualization provide explanatory information for diagnostic decisions. The framework was evaluated on the CIC-IoT-DIAD 2024 dataset across device-aware learning, imbalanced attack diagnosis, temporal sensitivity, federated learning, explainability, and operational-efficiency scenarios. Under the controlled homogeneous evaluation, the complete CA-AFiD framework achieved a 99.22% F1-score, while the heterogeneous evaluation achieved an overall 99.03% F1-score across the evaluated attack categories. Across the four evaluated minority attack categories, the average F1-score was 98.79%. Under federated learning, the global model achieved an average device-identification accuracy of 98.90% and an average anomaly-diagnosis F1-score of 98.90% across the participating Fog nodes, while ATT&CK mapping achieved 88–98% coverage. These results demonstrate the potential of CA-AFiD to provide adaptive, data-local, and interpretable intrusion diagnosis for heterogeneous IoT–Fog–Cloud environments.

1 November 2024 in Indonesian Journal of Electrical Engineering and Computer Science
Verified
Journal Article
Open access
PDF file
VSVijaya Kumar S
Vijaya Kumar S
MPMuthusamy Periyasamy
Muthusamy Periyasamy
RRRadhakrishnan Rajamani
Radhakrishnan Rajamani
TKTamilarasi Karuppiah
Tamilarasi Karuppiah
TEThenmozhi E
Thenmozhi E
Distributed computing has turned into an indispensable application administration because of the colossal development and fame of the internet. However, determining the allocation of various tasks to suitable service nodes is crucial. For the reasons expressed over, an effective booking strategy is expected to work on the framework’s exhibition. As a result, three-layer cloud dispatching (TLCD) design is introduced to further develop mission planning execution. The assignments should be arranged into various sorts in the primary layer in radiance of about their personalities clustering selection algorithm is composed of then recommended in second layer towards dispatch the undertakings to significant help bunches. Likewise, to further develop booking effectiveness, another planning technique for third stage proposes dispatching that job here to system thinking in a central server. As a rule, the proposed TLCD design yields the quickest work finishing time. Moreover, in cloud computing network architecture, load balancing and stability can be achieved.

1 August 2025 in International Journal of Informatics and Communication Technology (IJ-ICT)
Verified
Journal Article
Open access
PDF file
AAAbinaya Alagarsamy
Abinaya Alagarsamy
TEThenmozhi E
Thenmozhi E
SRS. P. Ramesh
S. P. Ramesh
TKTamilarasi Karuppiah
Tamilarasi Karuppiah
PKPrabu Kaliyaperumal
Prabu Kaliyaperumal
RPRajakumar Perumal
Rajakumar Perumal
In today’s world, the significance of network security and cloud environments has grown. The rising demand for data transmission, along with the versatility of cloud-based solutions and widespread availability of global resources, are key drivers of this growth. In response to rapidly evolving threats and malicious attacks, developing a robust intrusion detection system (IDS) is essential. This study addresses the imbalanced data and utilizes an unsupervised learning approach to protect network data. The suggested hybrid framework employs the CIC-IDS2017 dataset, integrating methods for handling imbalanced data with unsupervised learning to enhance security. Following preprocessing, principal component analysis (PCA) reduces the dimensionality from eighty features to twenty-three features. The extracted features are input into density-based spatial clustering of applications with noise (DBSCAN), a clustering algorithm. particle swarm optimization (PSO) optimizes DBSCAN, grouping similar traffic and enhancing classification. To address the imbalances in the learning process, the autoencoder (AE) algorithm demonstrates unsupervised learning. The data from the cluster is input into the AE, a deep learning algorithm, which classifies traffic as normal or an attack. The proposed approach (PCA+DBSCAN+AE) attains remarkable intrusion detection accuracy exceeding 98%, and outperforms five contemporary methodologies.

1 March 2026 in International Journal of Informatics and Communication Technology (IJ-ICT)
Verified
Journal Article
Open access
PDF file
SPSudhakar Periyasamy
Sudhakar Periyasamy
VAVijayalakshmi Alagarsamy
Vijayalakshmi Alagarsamy
PLPalani Latha
Palani Latha
KTKaruppiah Tamilarasi
Karuppiah Tamilarasi
TEThenmozhi E
Thenmozhi E
PKPrabu Kaliyaperumal
Prabu Kaliyaperumal
Distributed systems are integral to modern digital infrastructure, supporting communication and data exchange across various sectors. Ensuring security while maintaining quality of service (QoS) in such environments presents a significant challenge. This study introduces a dynamic network monitoring system (DNMS) that incorporates adaptive monitoring mechanisms and dynamic security metrics to safeguard distributed systems. The proposed architecture utilizes an event analyzer (EA) to evaluate and classify system events based on criticality, enabling secure transmission decisions and efficient threat detection. Experimental evaluations demonstrate the DNMS achieves a low processing overhead of 12%, supports a high data handling capacity of 5,000 requests per second, and maintains a latency of just 150 milliseconds. Additionally, it ensures strong compliance with regulatory standards-achieving 95% alignment with GDPR and 97% with ISO 27001- and high threat detection accuracy, with 98% for phishing, 94% for malware, and 96% for insider threats. These results confirm the framework’s effectiveness in enhancing adaptive security, offering scalable and regulation-compliant solutions for complex distributed environments.

1 March 2026 in International Journal of Informatics and Communication Technology (IJ-ICT)
Verified
Journal Article
Open access
PDF file
SPSathiyaseelan Periyasamy
Sathiyaseelan Periyasamy
AKAnubhav Kumar
Anubhav Kumar
MKMUTHULAKSHMI K
MUTHULAKSHMI K
TEThenmozhi E
Thenmozhi E
PKPrabu Kaliyaperumal
Prabu Kaliyaperumal
RPRajakumar Perumal
Rajakumar Perumal
The accelerating pace of digital transformation in the banking sector has highlighted the critical need for comprehensive cybersecurity strategies capable of countering evolving cyber threats. This study introduces an innovative intrusion detection framework tailored for banking environments, leveraging the CICIDS2017 and CSECICIDS2018 datasets for evaluation and validation. The proposed framework integrates data preprocessing, feature reduction, and advanced attack detection methods to enhance detection accuracy. A basic autoencoder is utilized for dimensionality reduction, streamlining input data while preserving essential attributes. The density-based spatial clustering of applications with noise (DBSCAN) algorithm is then applied for attack detection, enabling the detection of intricate attack patterns and their classification into specific attack groups. The proposed adaptive intrusion detection system (IDS) framework demonstrates outstanding performance, achieving precision, recall, F1-score, and accuracy rates exceeding 98%. Comparative evaluations against conventional techniques, such as support vector machines (SVM), long short-term memory (LSTM), and K-means, highlight its superiority in terms accuracy and computational efficiency. This research address key challenges, including high-dimensional datasets, class imbalance, and dynamic threat landscapes, offering a scalable and efficient solution to enhance the security of banking operations and enable proactive threat mitigation in the sector.