A Next-Generation AI based Predictive Model for Indian Rainfall using Real-Time API Data G. Chamundeswari, V.Santha Kumar, Prasad Babu Bairisetty, Chiranjeevi Aggala, Nagendra Kumar Yakkala, Ch.Venkatesh Ch.Venkatesh Proceedings of the 2026 International Conference on AI Driven Smart Systems and Ubiquitous Computing Icauc 2026, 2026 Accurate rainfall prediction is more useful for government sectors to prevent huge losses from heavy rainfall. Rainfall prediction affects various fields, including agriculture, water utilization, disaster management, and environmental condition prediction in India. In this paper, a Next-Generation Artificial Intelligence (NG-AI)–based predictive model is presented that accurately detects and predicts rainfall by combining real-time meteorological metrics from Weather APIs to improve dynamic rainfall prediction. The proposed NG-AI processes live climate predictors, such as ClimateStreamNet (CS-Net), enabling regular, up-to-date predictions. The proposed NG-AI refines the final output by integrating ensemble feature maps with enhanced classification layers, which are applied across multiple datasets. The NG-AI uses the Python API to find the accurate climatic conditions. The results show that the proposed approach achieved MAE (mm) of 5.32, MSE (mm<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) of 111.56, RMSE (mm) of 11.67, and MAPE (%) of 5.23. These results represent superior performance.
An Autonomus Learning Framework for Intelligent Loan Eligibility Assessment G.Ratna Kumari, K. Gopal Reddy, G.Chamundeswari, Prasad Babu Bairisetty, Chiranjeevi Aggala, Ch.Venkatesh Proceedings of the 2026 International Conference on AI Driven Smart Systems and Ubiquitous Computing Icauc 2026, 2026 In India, many people are suffering from multiple loan rejections. Even people with genuine financial reports are facing loan rejections, even for very minor amounts. It is difficult to determine the loan rejection reason using manual methods or existing ML algorithms. As per RBI guidelines, many financial companies use automated systems to identify accurate reasons. But the loan rejection rate remains high even for minor reasons, even with a high CIBIL score. In this paper, an Autonomous Learning Framework (ALF) is introduced to verify loan eligibility and accurately predict it in real time. The proposed approach is an automated system that extracts high-dimensional features from applicant data and refines eligibility based on minor features. The proposed ALF also refines the model by using continuous data collected from financial agencies for easier processing and decision-making. On the other hand, an explainable AI is integrated into the proposed approach, improving predictive performance and fairness, helping the banking sector and financial companies. The experimental results are conducted across various benchmark datasets and demonstrate high accuracy and other performance metrics.
Adversarial Deep Learning Models for Visual Image Recovery and Quality Enhancement Priyanka Kaushik, Rajendra Kumar Vairagi, Prasad Babu Bairysetti, Anish Gupta, Rohit Agarwal, Bhavik Kuchipudi Proceedings of the 4th IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation Iatmsi 2026, 2026 The paper focuses on the importance of restoring and preserving past visual materials, especially old and damaged images, in order to protect the heritage. It examines different image restoration methods in the treatment of degradation due to aging, fading and physical damages. A comparative discussion is offered between the old-fashioned techniques of restoration, including manual retouching, and the new ones based on deep learning and Generative Adversarial Networks (GANs), which emphasizes the necessity to find the balance between visual and historic authenticity. The main issues debated in the study are the limited number of high-quality reference images and difficulty in degradation patterns. Ethical issues that revolve around altering historical records are also discussed, noting the need to be as transparent as possible as well as promote good responsible restoration that can be done. In general, the paper has shown that the method of image enhancement by using GANs is promising and effective in improving the visual quality of historical images without compromising the originality of their nature, which is expected to be available to the generations to come.
Advancements and Challenges in Additive Manufacturing: Future Directions and Implications for Sustainable Engineering Raffi Mohammed, Abdul Saddique Shaik, Subhani Mohammed, Kiran Kumar Bunga, Chiranjeevi Aggala, Bairysetti Prasad Babu, Irfan Anjum Badruddin Advance Sustainable Science Engineering and Technology, 2025 This study explores the recent advancements in additive manufacturing (AM) and its significant effects on various industries such as aerospace, automotive, medical, and casting. The research investigates how AM has the potential to enhance design flexibility, reduce weight, and optimize material performance through developments like adaptive algorithms, topology-based process planning, and multi-objective optimization techniques. These advancements have resulted in near-net-shape casting, improved surface finishes, and enhanced structural integrity. However, the widespread adoption of AM in the commercial sector faces challenges such as high costs, limited material compatibility, and inconsistent build quality. This paper assesses these limitations and suggests solutions such as enhanced design algorithms, AI-driven process monitoring, and the creation of sustainable materials to address them. By overcoming these barriers, AM can smoothly integrate into industrial environments and revolutionize manufacturing processes. The study emphasizes the importance of further exploration of AM's potential to drive innovation, sustainability, and productivity across different sectors.
Human-centric innovations and sustainable transformations: Bridging industry 5.0 and society 5.0 for smart manufacturing and resilient communities Raffi Mohammed, Prasad Babu Bayrisetti, Chiranjeevi Aggala, Sarma S. Subramanya, Kiran Kumar Bunga, Subhani Mohammed, Ranga Jarabala, Jagan Mohan Rao Saride AI and Digital Nomads Shaping Global Industrial Technology Transitions, 2025 The fusion of Industry 5.0 and Society 5.0 is driving sustainable transformations and human-centric innovations that enhance smart manufacturing and resilient communities. This chapter explores how AI, IoT, blockchain, and digital twins boost human-machine collaboration, leading to improved efficiency and industrial resilience. Innovations also introduce challenges, such as ethical AI and workforce adaptation. Society 5.0 aims to create smart ecosystems that foster economic progress, while Industry 5.0 emphasizes human-driven decision-making and circular economy practices to minimize environmental impact. The chapter analyzes challenges and opportunities, providing strategies to bridge gaps and accelerate sustainable innovations for an intelligent future.
FUTURE TRENDS AND EMERGING TECHNOLOGIES IN MECHANICAL ENGINEERING: AN ANALYTICAL PERSPECTIVE Raffi Mohammed, Bairysetti Prasad Babu, Subramanya Sarma S, C. Sailaja, Subhani Mohammed, Kiran Kumar Bunga, Chiranjeevi Aggala Journal of Mechanics of Continua and Mathematical Sciences, 2025 Engineering is a specially designed course that includes the application of knowledge explicitly in the field of science and natural phenomena. The fields of engineering, technology, and physical sciences have been growing towards a new era of development and innovation across the globe. They include many fields, and one such significant area is mechanical engineering, which deals with the construction, working principles, and applications of various types of machines. Technical data of the products based on their scientific principles, along with parameters, are involved in the development of mechanical engineering. With this background, this study is designed to look forward to the future directions and emerging technologies in mechanical engineering. This review study investigated the future direction and emerging technology in mechanical engineering. It also highlighted the purpose and significance of mechanical engineering and discussed some of the research questions in mechanical engineering. Future directions of learning and technology, mechanical invention and development, the transportation industry, electric vehicles, and the artificial intelligence industrial revolution are also mentioned in this study. Mechanical engineering is a growing field of technology across the world. This review study indicated that it is essential to have upgraded knowledge and skills in the field of engineering and technology in this modern era. Many theories can be applied in the mechanical field with the support of upgrades in technology. The direction of mechanical engineering study is to learn the mechanical aspects of different technologies and the knowledge about that technology to optimize its use.
Overcoming Dataset Imbalances and Computational Challenges in IoT Intrusion Detection: A SMOTE-Enhanced Transformer-Based Model S. Venkatasubramanian, S. K. B. Pradeep Kumar Ch, Bairysetti Prasad Babu 2025 IEEE 14th International Conference on Communication Systems and Network Technologies Csnt 2025, 2025 Due to security holes in IoT-enabled networks, the massive adoption of IoT has unleashed a new wave of threats into computer networks. Due to their limited resources, these devices are ill-equipped to handle security measures, making them the most vulnerable part of our computer networks and putting our data and systems at risk. Some have suggested Intrusion Detection Systems (IDS) as a means to lessen the impact of intrusions connected to the IoT in an effort to solve this problem. Although IDS are essential for detecting threats, they are not widely used because of their high computing costs and reliance on labelled data. Finding a happy medium between fast detection and excellent accuracy is the objective of this model development effort. In order to accomplish this, to present a feature selection method based on the black-winged kite algorithm (BKA), a groundbreaking swarm intelligence technique that combines the Leader tactic and the Cauchy variation procedure to find the expansive appropriate convergence solution. BKA is based on the migratory and predatory instincts of black-winged kites. The study used few-shot learning techniques by leveraging transfer learning models and specifically using pre-trained architectures for categorisation. It also fine-tuned four pre-trained transformers. To take into account the imbalance in these datasets as a significant component and use the Synthetic Minority Oversampling Technique (SMOTE) to address it. The research team used BoT-IoT, TON-IoT, and CIC-DDoS2019 as benchmark datasets to assess the impact of the chosen features on the suggested model's performance. Metrics for evaluation include F1-score, recall, precision, and accuracy of detection. Evidence from experiments shows that the suggested ensemble model outperforms state-of-the-art AIDS topologies for IoT networks in terms of detection accuracy and efficiency.
A Hybrid Approach for Secure Data Retrieval and SQL Injection Attack Mitigation in Cloud Storage Using Merkle Array Hash Tree and Machine Learning Kusuma Sundara Kumar, L.L.S. Maneesha, Byrisetti Prasad Babu, M.Vani Pujitha, G.Diwakar, Suresh Betam 2nd Asian Conference on Intelligent Technologies Acoit 2025, 2025 Cloud computing has emerged as a crucial paradigm for data storage and processing owing to its scalability and cost-effectiveness. Outsourcing sensitive data to third-party cloud providers presents significant security problems, especially with secure data retrieval and query-based attacks. SQL injection continues to be one of the most common and significant risks. This study presents a hybrid methodology that combines a Merkle Array Hash Tree (MAHT) for safe data retrieval with a machine learning-based arrangement for the identification and inhibition of SQL injection attacks. The MAHT ensures data integrity, efficient verification, and trusted retrieval in geographically distributed cloud environments, while the machine learning model embedded in a daemon process validates and monitors user queries before execution. Experimental evaluation demonstrates the effectiveness of the proposed system in enhancing data security, reducing retrieval latency, and preventing malicious query attacks, thus offering a robust framework for secure cloud data management.
A Hybrid Approach for Heart Disease Prediction using Principal Feature Extraction and Deep Forest Modeling Prasad Babu Bairysetti, Obu Venkatesh Yadav, V T Ram Pavan Kumar M, V S N Murthy, Sreedhar Bhukya, Tulasi Ambica Veerla Proceedings of 7th International Conference on Inventive Material Science and Applications Icima 2025, 2025 This study proposes integrating advanced feature selection with the Deep Forest (DF) algorithm to enhance heart disease prediction. Heart diseases still contribute to the largest number of mortality in different parts of the world. This implies the necessity of efficient diagnostic models for timely, early detection, and intervention. To develop a robust predictive model characterized by accuracy and efficiency is the major aim of this research. For this purpose, the research uses Principal Feature Extraction (PFE), a feature selection technique that is developed to support the cascaded nature of decision trees in DF. A hybrid approach is adopted for feature selection, which removes noise and increases the accuracy of the prediction. The DF model has an accuracy of 96.36% with high precision and recall values using key features obtained from the analysis of the dataset. The integration of PFE with DF demonstrates the model's ability to capture both local and global patterns in data, which in turn enhances the classification performance. Future research should be conducted to apply this approach to larger and more diverse datasets to ensure broader generalizability. Moreover, the integration of real-time data could further enhance clinical applicability. The findings of this study contribute to the advancement of predictive models in healthcare, promoting earlier diagnosis and improved patient care.
Harnessing Deep Neural Networks for Accurate PCOS Diagnosis from Medical Images G.S.S.S.S.V. Krishna Mohan, S K B Pradeepkumar Ch, Bairysetti Prasad Babu, G. UsandraBabu, M.V.H. Bhaskara Murthy, Mahammad FiroseShaik 2024 International Conference on Computational Intelligence for Green and Sustainable Technologies Iccigst 2024 Proceedings, 2024 Polycystic Ovary Syndrome (PCOS), a rampant endocrine anomaly in ladies, necessitates accurate and efficient diagnostic methods. This study presents a groundbreaking approach using a Custom Convolutional Neural Network (CNN) in Python’s Jupyter Notebook using Keras library for PCOS detection through ultrasound image analysis. Leveraging the “PCOS Detection with Ultrasound Images” dataset from Kaggle, our CNN architecture not only outperforms conventional classifiers but also surpasses the reference Linear Discriminant and KNN classifiers, yielding the highest accuracy results. Optimized with the Adam optimizer and binary cross entropy lossfunction, our model marks a significant advancement in PCOS diagnostics. Beyond technical prowess, our research delves into the broader implications of PCOS, emphasizing the urgent need for improved diagnostic tools to address the widespread impact on women’s health across diverse demographics. This study showcases the superiority of our approach over existing models, particularly surpassing the accuracy and precision outcomes of clinical data - based projects. With a focus on pre-processed ultrasound images, our Pythonbased deep learning framework represents a transformative stride towards a more accurate, accessible, and effective PCOS diagnostic tool, offering promising prospects for advancing women’s healthcare.
