Early Detection of Anomalies in Photovoltaic Module Strings Using Decision Trees for MPPT Solar Charger Systems D.Ramesh Reddy, T. S. Saravanan, P . Subhashini, Saravana Selvan, Sonia Maria D'Souza, Thiyagesan M, M . DuraiRaj, Raja-sekhara Babu L International Journal of Basic and Applied Sciences, 2025 This study presents a decision tree (DT) based machine learning approach to detect early anomalies and faults in solar maximum power point tracking (MPPT) integrated photovoltaic (PV) module circuits. A four-panel array, built using a single diode model, is simulated to generate a synthetic, balanced dataset for training and evaluation. Among the different models tested, including neural networks (NN) and support vector classifiers (SVC), the DT model showed the best performance in precision and recall across all anomaly labels while maintaining simplicity and low computational cost. Currently, the model is limited to four module configurations, and the use of synthetic data may lead to overfitting when applied to real-world scenarios. Nevertheless, the methodology can be adapted to grid configurations with other known parameters. This work provides a practical basis for integrating early anomaly detection systems into PV installations, increasing operational efficiency and reducing maintenance costs.
A review of reinforcement learning approaches for autonomous systems in industry 4.0 BH Krishna Mohan, Pulicherla Padmaja, M Durairaj, P Nagamalleswrarao, K Srinivasarao, Sagaya Aurelia Industry 4 0 Key Technological Advances and Design Principles in Engineering Education Business and Social Applications, 2024 The integration of reinforcement learning (RL) into autonomous systems is investigated in this study, which also examines major developments, practical issues, practical applications, as well as optimization techniques. As significant trends, deep reinforcement learning (RL), transfer learning, and interaction with other machine learning paradigms increase flexibility. Critical issues, including sample inefficiency along with elevated computational demands call for creative solutions. Real-world uses for RL in robotics, autonomous driving, including health care, demonstrate its revolutionary potential. Domain-specific modification and ethical concerns are suggested to increase RL’s effectiveness. Interdisciplinary frameworks including the investigation of new fields are some future research directions.
Highlighting bugs in software development codes using SDPET for enhancing security N.A. Bhaskaran, M. Durairaj Measurement Sensors, 2023 The requirement for high-quality, inexpensive software that can be maintained is being driven by the rise in demand for automated online software systems. Early defect identification in SDLCs (Software Development Life Cycles) results in early adjustments and eventually on-time delivery of maintainable software, satisfying the client and fostering his trust in the development team. Many MLTs (Machine Learning Techniques) have been put out in the last ten years to increase SDP accuracy. Most of the suggested SDPs frameworks and models employ ANNs (Artificial Neural Networks), which are a popular MLTs. Software defect data are hampered by a number of problems, including duplication, correlation, feature relevance, and missing samples. However, because to the under/over fitting issues, most existing SDPs utilising ANNs have low accuracy. SDPET (Software Defect Predictions Ensemble Technique), an ensemble learning technique to produce accurate SDPs based on the AdaBoost algorithm, is proposed. The proposed schema's efficacy against RFs (Random Forests) and GBs(Gradient Boosts) for needed values through experiments. The experiment results verify that the suggested SDPET has good accuracy in training and better accuracy in test datasets when compared with other methods. The original obtained dataset was cleaned of unnecessary features, converted to csv, and is stored as dataset. csv.
Financial time series prediction using deep computing approaches M. Durairaj, Ch. Suneetha, BH. Krishna Mohan Journal of Autonomous Intelligence, 2023 <p class="Abstracttitle">A financial time series is chaotic and non-stationary in nature, and predicting it outcomes is a very complex and challenging task. In this research, the theory of chaos, Long Short-Term Memory (LSTM), and Polynomial Regression (PR) are used in tandem to create a novel financial time series prediction hybrid, Chaos+LSTM+PR. The first step in this hybrid will determine whether or not a financial time series contains chaos. Following that, the chaos in the time series is modeled using Chaos Theory. The modeled time series is fed into the LSTM to obtain initial predictions. The error series obtained from LSTM predictions is fitted by PR to obtain error predictions. The error predictions and initial predictions from LSTM are combined to obtain final predictions. The effectiveness of this hybrid is examined by three types of financial time series (Chaos+LSTM+PR), including stock market indices (S&amp;P 500, Nifty 50, Shanghai Composite), commodity prices (gold, crude oil, soya beans), and foreign exchange rates (INR/USD, JPY/USD, SGD/USD). The results show that the proposed hybrid outperforms ARIMA (autoregressive integrated moving average), Prophet, CART (Classification and Regression Tree), RF (Random Forest), LSTM, Chaos+CART, Chaos+CART, and Chaos+LSTM. The results are also checked for statistical significance.</p>
Statistical Evaluation and Prediction of Financial Time Series Using Hybrid Regression Prediction Models M. Durairaj, B. H. Krishna Mohan International Journal of Intelligent Systems and Applications in Engineering, 2021 : Financial time series are chaotic by nature, which makes prediction difficult and complicated. This research employs the new hybrid model for the prediction of FTS which comprises Long Short-Term Memory (LSTM), Polynomial Regression (PR), and Chaos Theory. First of all, FTS is tested for the presence of chaos, in this hybrid model. Later, using Chaos Theory, the time series is modelled with the chaos existence. The model time series will be entered in LSTM for initial forecasts. The sequence of errors derived from LSTM forecasts is PR appropriate for error predictions. Error forecasts and original model forecasts are applied to produce the final hybrid model forecasts. Performance testing of the hybrid model (Chaos+LSTM+PR) is conducted using three categories namely foreign exchange, commodity price and stock-market indices. The hybrid model proposed in this study, in compliance with MSE, Dstat and Theil’s U, is proved superior to the individual models like ARIMA, Prophet, LSTM and Chaos+LSTM. The execution of these various hybrid proposed methods is done mainly using Python, additionally, the authors used Gretl® and R for some methods respectively. Ultimately, the final result of this hybrid model describes with a better result than the existing prediction models and it is proved using various types of FTS like Foreign exchange rates, commodity prices, and stock market indices respectively. Hence, the result shows that the proposed hybrid models of Chaos+LSTM+PR achieved with better prediction rate than the existing models on the nine datasets executed.
MBC-ODCA algorithm to select an optimal datacenter for resource allocation in mobile cloud computing Journal of Advanced Research in Dynamical and Control Systems, 2017
A comparison of the perceptive approaches for preprocessing the data set for predicting fertility success rate International Journal of Control Theory and Applications, 2016
A review on big data analytics tools for telecommunication industry International Journal of Control Theory and Applications, 2016
A review on affective computing International Journal of Control Theory and Applications, 2016
ThreV - An efficacious algorithm to thwart MAC spoof DoS attack in wireless local area infrastructure network Indian Journal of Science and Technology, 2014
Theoretical framework of ANM and hybridization of ANM with ThreV in detecting and preventing DoS attacks in wireless infrastructure network International Journal of Applied Engineering Research, 2014
Criminal behavior analysis by using data mining techniques IEEE International Conference on Advances in Engineering Science and Management Icaesm 2012, 2012
Real-Time ECG Image Classification Using GA–PSO Optimized CNN-LSTM Model Trained on PTB-XL Dataset S Selvakumari, M Durairaj Indian Journal of Science and Technology 19 (13), 868-875 , 2026 2026
A Hybrid CNN-LSTM Framework for ECG Classification with Genetic Algorithm-Based Feature Optimization M Durairaj, S Selvakumari Indian J. Sci. Technol. 18 (31), 2509-2519 , 2025 2025 Citations: 2
A comparative study of optimization techniques in deep learning using the MNIST dataset S Selvakumari, M Durairaj Indian J. Sci. Technol 18 (10), 803-810 , 2025 2025 Citations: 10
A Review of Reinforcement Learning Approaches for Autonomous Systems in Industry 4.0 BHK Mohan, P Padmaja, M Durairaj, P Nagamalleswrarao, ... Industry 4.0 Key Technological Advances and Design Principles in Engineering … , 2024 2024
Highlighting bugs in software development codes using SDPET for enhancing security NA Bhaskaran, M Durairaj Measurement: Sensors 30, 100930 , 2023 2023 Citations: 5
Epilson Swarm Optimized Cluster Gradient and deep belief classifier for multi-attack intrusion detection in MANET S Dilipkumar, M Durairaj Journal of Ambient Intelligence and Humanized Computing 14 (3), 1445-1460 , 2023 2023 Citations: 38
Financial time series prediction using deep computing approaches M Durairaj, C Suneetha, M Krishna Journal of Autonomous Intelligence 6 (1), 558 , 2023 2023 Citations: 6
A convolutional neural network based approach to financial time series prediction DM Durairaj, BHK Mohan Neural Computing and Applications 34 (16), 13319-13337 , 2022 2022 Citations: 175
Detection of attacks using multilayer perceptron algorithm S Dilipkumar, M Durairaj Inventive Communication and Computational Technologies: Proceedings of … , 2022 2022 Citations: 4
Statistical evaluation and prediction of financial time series using hybrid regression prediction models M Durairaj, KBH Mohan International Journal of Intelligent Systems and Applications in Engineering … , 2021 2021 Citations: 13
Protein Secondary Structure Prediction Using FFA Optimized ANN M Durairaj, S Sivakumar, B Sangeetha, K Saravannan, K Saravanakumar Annals of the Romanian Society for Cell Biology 25 (5), 5257-5266 , 2021 2021
Enhanced Gradient Boosting Tree Classifier Using Optimization Technique for Water Quality Prediction M Durairaj, T Suresh Annals of the Romanian Society for Cell Biology 25 (2), 3860-3873 , 2021 2021 Citations: 1
Fuzzy probability based person recognition in smart environments M Durairaj, J Hirudhaya Mary Asha Journal of Intelligent & Fuzzy Systems 40 (5), 9437-9452 , 2021 2021 Citations: 3
Enhanced Gradient Boosting Tree Classifier using Optimization Technique for Water Quality Prediction M Durairaj, T Suresh Annals of the Romanian Society for Cell Biology 25 (2), 3860-3876 , 2021 2021 Citations: 1
A CLASSIFICATION MODEL WITH OPTIMIZATION BASED FEATURE SELECTION METHOD FOR INTRUSION DETECTION SYSTEM M Durairaj, D Radhika PalArch's Journal of Archaelogy of Egypt / Egyptology 17 (6), 9318-9334 , 2020 2020
A Chaotic Algorithm for Image Encryption M Durairaj, JHM Asha Image Processing and Capsule Networks: ICIPCN 2020, 218 , 2020 2020
The Appraised Structure for Improving Quality in the Compressed Image Using EQI-AC Algorithm M Durairaj, J Hirudhaya Mary Asha International Conference on Image Processing and Capsule Networks, 201-217 , 2020 2020
ExclusiveOR-Discrete Cosine Transform-A Chaotic Algorithm for Image Encryption and Decryption M Durairaj, J Hirudhaya Mary Asha International Conference on Image Processing and Capsule Networks, 218-232 , 2020 2020 Citations: 1
Interoperability in smart living network—a survey M Durairaj, J Hirudhaya Mary Asha International Conference on Communication, Computing and Electronics Systems … , 2020 2020 Citations: 7
Volatility Nature of Financial Time Series Applications during Covid-19 Era M Durairaj, BHK Mohan, M Manjusha DYNAMIC BUSINESS TRENDS AND INNOVATIONS IN CONTEMPORARY TIMES, 128-137 , 2020 2020
MOST CITED SCHOLAR PUBLICATIONS
Data Mining Applications In Healthcare Sector: A Study M Durairaj, V Ranjani International Journal of Scientific & Technology Research 2 (10), 29-35 , 2013 2013 Citations: 248
A convolutional neural network based approach to financial time series prediction DM Durairaj, BHK Mohan Neural Computing and Applications 34 (16), 13319-13337 , 2022 2022 Citations: 175
Educational Data mining for Prediction of Student Performance Using Clustering Algorithms M Durairaj, C Vijitha International Journal of Computer Science and Information Technologies 5 (4 … , 2014 2014 Citations: 125
Prediction Of Heart Disease Using Back Propagation MLP Algorithm M Durairaj, V Revathi International Journal of Scientific & Technology Research 4 (08), 235-239 , 2015 2015 Citations: 124
A Comparison of the Perceptive Approaches for Preprocessing the Data Set for Predicting Fertility Success Rate M Durairaj, R Nandhakumar International Journal of Control Theory and Applications (IJCTA) 9 (27 … , 2016 2016 Citations: 77
Applications of Artificial Neural Network for IVF Data Analysis and Prediction M Durairaj, P Thamilselvan Journal of Engineering, Computer & Applied Sciences (JEC & AS) 2 (9), 11-15 , 2013 2013 Citations: 68
A study on security issues in cloud based e-learning M Durairaj, A Manimaran Indian Journal of Science and Technology 8 (8), 757-765 , 2015 2015 Citations: 64
A Study on Virtualization Techniques and Challenges in Cloud Computing M Durairaj, P Kannan International Journal of Scientific and Technology Research 1 (1), 2277-8616 , 2012 2012 Citations: 56
Prediction of Diabetes Using Soft Computing Techniques - A Survey M Durairaj, G Kalaiselvi International Journal of Scientific & Technology Research 4 (3), 190-192 , 2015 2015 Citations: 39
Epilson Swarm Optimized Cluster Gradient and deep belief classifier for multi-attack intrusion detection in MANET S Dilipkumar, M Durairaj Journal of Ambient Intelligence and Humanized Computing 14 (3), 1445-1460 , 2023 2023 Citations: 38
A Hybrid Prediction System Using Rough Sets and Artificial Neural Networks M Durairaj, K Meena Journal of Innovative Technology & Creative Engineering 1 (7) , 2011 2011 Citations: 33
An empirical study on applying data mining techniques for the analysis and prediction of heart disease S Sivagowry, M Durairaj, A Persia Information Communication and Embedded Systems (ICICES), 265-270 , 2013 2013 Citations: 32
Mobile Augmented Reality and Location Based Service A Sagaya Aurelia, M Durairaj, O Saleh Advances in Information Sciences and Applications 2, 551-558 , 2014 2014 Citations: 31
Criminal behavior analysis by using data mining techniques K Zakir Hussain, M Durairaj, GRJ Farzana Advances in Engineering, Science and Management (ICAESM), 2012 International … , 2012 2012 Citations: 31
High relevancy low redundancy vague set based feature selection method for telecom dataset TS Poornappriya, M Durairaj Journal of Intelligent & Fuzzy Systems, 1-18 , 2019 2019 Citations: 30
PREDICTION OF DIABETES USING BACK PROPAGATION ALGORITHM M Durairaj, G Kalaiselvi International Journal of Emerging Technology and Innovative Engineering 1 (8 … , 2015 2015 Citations: 30
A New Authentication Scheme with Elliptical Curve Cryptography for Internet of Things (IoT) Environments M Durairaj, K Muthuramalingam International Journal of Engineering & Technology 7 (2), 119-124 , 2018 2018 Citations: 28
Data Mining Application on IVF Data For The Selection of Influential Parameters on Fertility M Durairaj, R Nandha Kumar International Journal of Engineering and Advanced Technology (IJEAT) 2 (6 … , 2013 2013 Citations: 27
A review of two decades of deep learning hybrids for financial time series prediction M Durairaj, BK Mohan International Journal on Emerging Technologies 10 (3), 324-331 , 2019 2019 Citations: 26
Choosing a spectacular Feature Selection technique for telecommunication industry using fuzzy TOPSIS MCDM M Durairaj, TS Poornappriya International Journal of Engineering & Technology 7 (4), 5856-5861 , 2018 2018 Citations: 24
