Journal of Intelligent Data Analysis and Computational Statistics (p-ISSN: 3049-3056 e-ISSN: 3048-7080)

A Review of Rainfall Prediction Using Machine Learning and Deep Learning

Naushin Sindhi — 2026-09-08

Exact rainfall forecasting is necessary in disaster management, long-term planning, agriculture, flood control, and water resource planning. In the past decade, there has been rapid development and enhancement in terms of data and computing technologies. The review presents a detailed description of new developments in satellite-based rainfall prediction, hydrological estimation, modelling, and especially changes from traditional methods. Current Artificial Intelligence and Machine Learning systems used in ground-based and satellite datasets consist of popular sources in India. GPM IMERG, Meteorological Department observation data, and CHIRPS datasets support large-scale modelling, but they are still challenged by issues of unequal spatial coverage, time, and variability in climatic regions. It is critical to deal with these problems to develop better prediction models. Machine Learning methods such as Random Forest, Gradient Boosting, Ensemble methods, and Support Vector Machines have been shown to perform well in short-term forecasting of rainfall, primarily because they work with sound input variables and detect hidden regularities. Deep learning models, including LSTM, CNN-based, and hybrid deep network models, also increase predictive ability by modelling detailed, non-linear, and spatiotemporal interdependencies that traditional models do not tend to reflect. This Artificial Intelligence-driven extreme forecasting is particularly beneficial for systems dealing with localised and monsoon rainfall variability.

Smart Burnout Prediction System for Students Using Machine Learning

Subha Dharshini. G — 2026-09-24

Student burnout has become one of the most significant challenges in modern education due to increasing academic pressure, examinations, assignments, project deadlines, competitive environments, and extracurricular activities. Continuous exposure to these factors can lead to emotional exhaustion, reduced motivation, decreased concentration, and poor academic performance. If burnout is not identified at an early stage, it may negatively affect students’ mental health, learning outcomes, and overall well-being. Traditional burnout assessment methods mainly rely on surveys, counseling sessions, and manual observation, which are often time-consuming and unable to provide continuous monitoring. This paper proposes a Smart Burnout Prediction System that utilizes machine learning and Predictive Analytics techniques to identify students who are at risk of academic burnout. The proposed system analyzes various academic, behavioral, and lifestyle factors such as attendance percentage, internal assessment scores, assignment completion rate, study hours, sleep duration, stress levels, and participation in academic activities. The collected data is preprocessed and analyzed using machine learning algorithms, including Decision Tree, Random Forest, and Logistic Regression, to classify students into Low, Medium, and High Burnout Risk categories. The system further provides personalized recommendations and preventive measures to help students effectively manage stress, improve time management, maintain a healthy study-life balance, and enhance academic performance. By enabling early detection and intervention, the proposed solution supports educational institutions in monitoring student well-being and implementing proactive support strategies. The integration of artificial intelligence, machine learning, and educational analytics makes the system a reliable and intelligent tool for improving student success, reducing burnout-related issues, and promoting a healthier learning environment. The proposed framework contributes to the development of smart educational systems by combining data-driven decision-making with student-centered support mechanisms.

A Review on Seismic Intelligence: Bridging AI and IoT for Predictive Earthquake Mitigation

Kazi Kutubuddin Sayyad Liyakat — 2026-09-11

The catastrophic impact of seismic events necessitates a paradigm shift from reactive emergency responses to proactive, real-time predictive modeling. This paper explores the integration of Artificial Intelligence (AI) with the Internet of Things (IoT) to revolutionize earthquake detection and decision-making. By deploying dense, low-cost sensor networks capable of capturing high-frequency seismic vibrations, it can create a distributed web of "seismic ears" that transmit data to edge-computing nodes. It proposes a hybrid architecture where Machine Learning (ML) algorithms, specifically Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs), process streaming IoT data to distinguish between ambient anthropogenic noise and genuine pre-seismic tremors. This framework facilitates automated, millisecond-latency decision-making, including the triggering of smart-grid shutdowns, automated transport halts, and instant wide-area early warnings. By reducing the reliance on sparse, high-cost seismic stations, this research demonstrates that a decentralized AI-driven IoT approach can significantly enhance the resolution of geological monitoring and drastically reduce the window of uncertainty in disaster mitigation.

A Survey of Artificial Intelligence Techniques for Early Leukemia Detection: Advances in Deep Learning, Blood Smear Analysis, and Intelligent Hematology Systems

Varun Doddagoudar — 2026-08-31

This survey presents a comprehensive review of recent Artificial Intelligence (AI)-based approaches for automated leukemia detection and hematological image analysis. The reviewed studies demonstrate the evolution of automated blood-cell detection and counting using Convolutional Neural Networks (CNNs), followed by hybrid Deep Learning–Machine Learning approaches for leukemia classification, optimized CNN architectures, and portable Raspberry Pi-based diagnostic systems. The survey compares the methodologies, performance, advantages, and limitations of these approaches and identifies important research gaps, including limited datasets, computational complexity, restricted leukemia-subtype coverage, limited clinical validation, and the lack of comprehensive patient-monitoring capabilities. Based on these identified gaps, the survey discusses future directions toward integrated hematology intelligence platforms. The proposed HemaSight AI concept extends the reviewed approaches by considering multiple stages of hematological assessment, including early risk prediction, synthetic visualization, relapse monitoring, and AI-assisted patient communication. Thus, the survey establishes a progression from conventional blood-cell analysis toward intelligent, scalable, and clinically applicable hematology systems.

EyeSpeak: Smart Eye Tracking System for Assistive Communication

Dyuthi Venkatesh — 2026-09-11

Recent developments in eye tracking technology and human–computer interaction have facilitated the creation of communication systems for people with severe physical and speech disabilities. This literature survey explores research works related to eye tracking, eye gaze estimation, blink detection, facial landmark detection, virtual keyboard, and multimodal interaction for hands-free communication. Studies have implemented various approaches and used different technologies to improve the accuracy of eye movement detection, typing speed, and overall communication efficiency. For example, computer vision with OpenCV, MediaPipe, Dlib, Eye Aspect Ratio (EAR), convolutional neural networks (CNNs), machine learning (ML), Large Language Models (LLMs), and text-to-speech APIs were employed to achieve high performance, reliability, and usability. Moreover, some researchers suggested using head pose estimation, voice commands, predictive text, and advanced interaction techniques to simplify the communication process and reduce physical interaction. The developed eye-tracking systems and communication tools were successfully applied to assistive communication, Augmentative and Alternative Communication (AAC), healthcare, rehabilitation, human–computer interaction, and accessibility domains, benefiting people with Amyotrophic Lateral Sclerosis (ALS), cerebral palsy, quadriplegia, and other conditions. However, most solutions have shortcomings, such as reduced accuracy due to lighting, calibration, and hardware constraints; added weight; eye strain; the need for ocular tracking; slow typing; and the complexity of the underlying algorithms. To summarize, the articles mentioned in this literature survey offer a useful insight into current trends and approaches used to design effective communication systems and help build a solid foundation for creating low-cost, easy-to-use, and high-performance communication tools.