Browsing by Autor "Vairavel Madeshwaren"

Now showing 1 - 9 of 9

A Review on Emoji Entry Prediction for Future Finance Market Analysis Using Convolutional Neural Network
(Turkish Online Journal of Qualitative Inquiry (TOJQI), 2025) Paul Narayanan; Kishore Kunal; Veeramani Ganesan; Singaravelu Ganesan; A. T. Jaganathan; Vairavel Madeshwaren
Textual and financial data in social media has come a long way in the present. Emojis, the primary focus of this study piece, allow emotions to be visually represented thanks to the advent of text-based digital communication. By adding visual currency attractiveness to text, emojis in digital communication enhance communication and open up new channels for innovation and exchange. The neural network model for text-based emoji entry prediction is highly optimised, however because of little knowledge in this field, it is more difficult to predict future emojis from images all the finance symbols. Emojis are a great alternative to linguistically independent, sentiment-aligned embeddings since they are consistent and convey a clear sentiment signal NSE and BSE market. Compared to models for text, models for symbolic description have received less attention. In this study,Main researchers employed CNN architecture for image classification together with an emoji2vec embedding into the word2vec model to predict emoji from photos apply in finance sector and finding. Additionally, we performed a sentiment analysis on the text to forecast upcoming emoji labels added. Our approach effectively communicates how the emojis relate to one another. The length of the search for incoming image-based emoji predictions has been optimised using this model.
Adoption of Circular Economy Principles: An Empirical Study of Green Strategies in Manufacturing Organizations
(2025) Kishore Kunal; C. Joe Arun; V. Selvakumar; Santhi Venkatakrishnan; Smriti Anand; Vairavel Madeshwaren
This study investigates the adoption of circular economy (CE) principles within the manufacturing sector, emphasizing the interconnected roles of green operational practices, stakeholder communication, corporate green image, and customer-driven sustainability demands. Circular economy frameworks advocate for closed-loop systems where materials are reused, waste is minimized, and product life cycles are extended to achieve environmental and economic sustainability. As global awareness of ecological challenges intensifies, consumers and regulatory bodies are increasingly expecting companies to demonstrate genuine commitments to sustainable operations. The research collected primary data from employees across three manufacturing firms actively engaging in environmental initiatives. Analytical tools such as SPSS and Smart PLS-SEM were employed to examine the relationships among the key variables. The results show a significant positive relationship between the adoption of CE principles more broadly and the successful communication of green practice implementation. CE adoption was further strengthened by upholding a visible green image. It's interesting to note that although they were significant, customer preferences had little influence on these strategic decisions. The significance of manufacturers institutionalizing sustainability through internal procedures and open stakeholder engagement is highlighted by these insights. To improve long-term environmental performance and shift towards a circular economy model, the study offers useful implications for industries.
AI-Driven Intelligent Control Strategies for Industrial Robotics: A Reinforcement Learning Approach
(2025) Kishore Kunal; M. Kathiravan; Vairavel Madeshwaren; T. Chandrakala; V. Ganesan; Sheifali Gupta
This study proposes an AI-driven adaptive control strategy to enhance the learning, adaptability, and autonomous performance of robotic ‎manipulators in dynamic and unstructured industrial environments. Moving beyond the limitations of conventional model-based controllers, ‎the research introduces a self-learning framework that integrates real-time sensor data from LiDAR and stereo vision cameras. This data ‎continuously informs and optimizes the robot’s motion trajectories in both simulated and real-world tasks. The system’s core innovation lies ‎in combining Reinforcement Learning (RL) with Deep Neural Networks (DNNs) for adaptive trajectory planning and error compensation. ‎Specifically, the Proximal Policy Optimization (PPO) algorithm is employed to fine-tune control strategies based on real-time sensory ‎feedback, allowing the robotic system to autonomously adapt to variations in object positions and unexpected disturbances. An Edge-AI ‎module is embedded into the architecture to enhance decision-making speed and reduce latency during task execution. Experimental ‎validation, including scenarios like arc welding and sealant dispensing, shows the proposed system outperforms traditional PID-based ‎adaptive controllers. The AI-driven solution demonstrated improved precision, faster convergence, and superior adaptability under complex ‎and fluctuating manufacturing conditions. The study also opens pathways for future integration of hybrid AI techniques—such as fuzzy ‎logic and genetic algorithms—for even more intelligent and responsive robotic systems‎.
Ensemble-Based Machine Learning Approach For Fake News Detection On Telegram With Enhanced Predictive Accuracy
(Turkish Online Journal of Qualitative Inquiry (TOJQI), 2025) Poody Rajan Y; Kishore Kunal; Amutha Govindan; Korra Balu; Veeramani Ganesan; Vairavel Madeshwaren
The rapid proliferation of fake news on social media platforms has raised significant concerns about misinformation, particularly on messaging applications like Telegram. This trend poses a severe threat to public trust and social harmony. Detecting fake news in such environments requires the development of efficient machine learning (ML) models that can accurately identify misleading content while minimizing false positives and negatives. This research aims to propose a robust machine learning-based framework for detecting fake news on Telegram by analyzing text content and user interaction patterns. Data collection involved scraping a dataset from publicly available Telegram channels, which include both genuine and fake news articles with relevant metadata such as user reactions and engagement levels. To address the problem of fake news detection, a set of machine learning algorithms, including XGBoost, K-Nearest Neighbors (KNN), Decision Trees, and Naive Bayes, were explored. A novel ensemble-based approach, termed Ensemble Feature Fusion (EFF), is introduced, combining the strengths of multiple classifiers to enhance predictive accuracy and robustness against diverse fake news characteristics. Performance metrics such as Accuracy, Engagement-Weighted Accuracy (EWA), False Positive Cost (FPC) , Contextual Precision (CP), and Temporal Consistency Index (TCI) were evaluated in this research. Results indicate that the proposed model outperforms conventional ML techniques, demonstrating improved classification accuracy and reduced error rates in detecting fake news. This approach provides a promising solution to the growing problem of misinformation on Telegram.
Federated Learning for Secure AI-Driven Predictive Maintenance in Smart Manufacturing
(2025) Kishore Kunal; Vairavel Madeshwaren; S. Leena Nesamani; A. Banushri; V. Ganesan; Sheifali Gupta
Background: In the Industry 4.0 landscape, integrating artificial intelligence (AI) with smart manufacturing is essential for enhancing automated monitoring, predictive maintenance, and system optimization. However, traditional centralized AI model training poses critical risks to data privacy, security, and scalability, especially when sensitive operational data from factory machines is shared across platforms. Methods: This study proposes a decentralized, intelligent framework designed for real-time machine monitoring that enhances fault detection accuracy while safeguarding data privacy. The approach begins with real-time sensor data acquisition—capturing vibration, temperature, and acoustic signals from distributed factory units via edge devices. These signals undergo preprocessing and advanced feature extraction using Wavelet Transform and Empirical Mode Decomposition (EMD) to reveal critical fault characteristics. Results: A hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks is used for classification. CNNs are responsible for extracting spatial features, whereas LSTMs identify temporal dependencies in time-series. With the federated learning (FL) framework, model training can be done collaboratively across edge devices without the need to transfer sensitive raw data. Conclusion: This ensures security and enhances model generalization. Results from experiments indicate that the suggested FL-based hybrid model exceeds centralized architectures regarding detection accuracy, computational efficiency, and adaptability. This research provides a scalable and secure solution that enhances intelligent monitoring for Industry 4.0 systems.
Intelligent 3D Printing Algorithms: Accelerating Additive Manufacturing with Precision and Defect-Free Fabrication
(2025) M. Kathiravan; Kishore Kunal; Vairavel Madeshwaren; P. Lavanya; V. Ganesan; Sheifali Gupta
Additive Manufacturing (AM) has transformed modern production by enabling the fabrication of complex geometries with enhanced material ‎efficiency. However, traditional 3D printing techniques often face challenges such as incomplete fusion, material inconsistencies, and thermal ‎warping, which affect overall quality and productivity. This study introduces an intelligent 3D printing framework that integrates Artificial ‎Intelligence (AI) to enable real-time monitoring, defect detection, and adaptive process control, thereby addressing these limitations. The ‎proposed system utilizes Convolutional Neural Networks (CNNs) for computer vision-based quality inspection, enabling the detection of ‎structural anomalies during the printing process. Reinforcement Learning (RL) is employed for dynamic adjustment of parameters like nozzle ‎temperature, deposition speed, and material feed rate in response to real-time feedback, significantly reducing defect occurrence. Adaptive ‎machine learning algorithms like Random Forests and Gradient Boosting also facilitate process optimization and predictive maintenance. ‎Stereolithography (SLA), Selective Laser Sintering (SLS) and Fused Deposition Modeling (FDM) are among the AM platforms that use this AI-‎AI-enhanced closed-loop control approach. Material use, energy efficiency, production time, print quality and defect mitigation have all significantly ‎improved, as confirmed by experimental validation. With its ability to guarantee accuracy and dependability in contemporary 3D printing ‎processes, the framework shows great promise for developing industrial and biomedical applications‎.
IoT and Blockchain in Supply Chain Management for Advancing Sustainability and Operational Optimization
(Turkish Online Journal of Qualitative Inquiry (TOJQI), 2025) St Mary'; Kishore Kunal; Vairavel Madeshwaren
The rapid advancement of IoT technologies has emerged as a key driver of sustainable development, reshaping industries and societal structures. This study critically examines the intersection of IoT and sustainability by analyzing contemporary literature on the subject. A comprehensive review of IoT-driven innovations highlights their transformative impact across sectors such as agriculture, smart cities, and resource management. The research investigates how digitalization, particularly within supply chains, redefines operational strategies and enhances sustainability metrics. With the integration of technologies like RFID, blockchain, and IoT under Industry 4.0, organizations are revolutionizing process efficiency, transparency, and environmental responsibility. To assess these implications, the study conducts two comparative simulation experiments involving a three-party supply chain in cheese production—one utilizing traditional methods and the other leveraging IoT-based innovations. Results reveal significant improvements in order management efficiency and compliance handling, underscoring the critical role of emerging technologies in fostering sustainable practices. The proposed framework provides valuable insights into the broader management implications of IoT adoption, reinforcing its potential as a catalyst for global sustainability initiatives.
Machine Learning Framework for Detecting Fake News and Combating Misinformation Spread on Facebook Platforms
(Turkish Online Journal of Qualitative Inquiry (TOJQI), 2025) Poondy Rajan Y; Kishore Kunal; Arun Palanisamy; Senthil Kumar Rajendran; Rupesh Gupta; Vairavel Madeshwaren
The spread of fake news on social media platforms like Facebook threatens societal harmony and undermines the reliability of information. To address this issue, this research employs machine learning techniques to construct a robust and scalable framework for detecting fake news. Using a well-curated dataset of labeled Facebook posts containing both authentic and fake news, the study ensures a balanced representation for effective learning. Textual data was transformed into numerical features through Term Frequency-Inverse Document Frequency (TF-IDF) preprocessing, enabling seamless integration with machine learning algorithms. A variety of classification models, including Support Vector Machines (SVM), Logistic Regression, Gradient Boosting, and Random Forest, were trained and evaluated. Six performance evaluations precision, accuracy, F1 score, recall, Matthews Correlation Coefficient (MCC), and area under the Receiver Operating Characteristic (ROC) curve—were used to measure model effectiveness. The results highlighted Gradient Boosting as the most effective algorithm, achieving superior accuracy and overall performance. This framework demonstrates the capability of machine learning to automate the detection of misinformation, offering a scalable and efficient solution for preserving content credibility on Facebook. The study contributes significantly to the broader effort of combating misinformation, ensuring the dissemination of reliable information, and safeguarding public trust on social media platforms
Stock Price Prediction in India: Comparing Stochastic Differential Equations with MCMC, LSTM, and ARIMA Models and Exploring a Hybrid Approach
(Turkish Online Journal of Qualitative Inquiry (TOJQI), 2025) S. Bhuvaneshwari; S. Nirmala Sugirtha Rajini; Paul Narayanan; Kishore Kunal; Vairavel Madeshwaren; Saranya Anbarasu
The inherent volatility and nonlinearity of stock prices make them a crucial challenge in financial markets. This study investigates how well stochastic differential equations (SDEs) with parameter estimation employ the Markov Chain Monte Carlo (MCMC) algorithm to model changes in Indian stock prices. To evaluate this methods predictive accuracy we compare its performance to that of conventional Long Short-Term Memory (LSTM) networks and AutoRegressive Integrated Moving Average (ARIMA) models. A probabilistic estimation of important parameters in the SDE model is made possible by the Bayesian inference framework used in the MCMC algorithm which successfully captures market uncertainties. Our results show each models advantages and disadvantages in predicting stock prices highlighting how well-suited each is for various time horizons and market circumstances. In order to take advantage of both stochastic modeling and deep learning capabilities we also suggest a novel hybrid model that combines SDE-MCMC with LSTM and ARIMA. Results from experiments show that by fusing the advantages of machine learning and statistics the hybrid model increases forecasting accuracy. In addition to offering insights for analysts and investors in making data-driven decisions this research advances stock price prediction techniques.