International Journal of Research and Review in Applied Science, Humanities, and Technology
Not a member yet
92 research outputs found
Sort by
Deep Neural Network Approaches for Emotion Recognition in Human–Computer Interaction
Get PDFEmotion recognition has become a pivotal research domain in Human–Computer Interaction (HCI), as modern interactive systems increasingly aim to respond not only to explicit user commands but also to implicit emotional cues. Understanding human emotions allows intelligent systems to adapt their behaviour, enhance user experience, and support applications such as intelligent tutoring systems, healthcare monitoring, customer service automation, and social robotics. Traditional emotion recognition methods relied heavily on handcrafted features and shallow machine learning algorithms, which struggled with high-dimensional data, environmental variability, and real-time performance constraints. Recent advances in deep learning have significantly transformed emotion recognition research. Deep Neural Networks (DNNs), including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, and hybrid architectures, have demonstrated superior capability in learning hierarchical and discriminative representations directly from raw multimodal data. These models have enabled more accurate recognition of emotions from facial expressions, speech signals, textual inputs, and physiological signals. This paper presents an in-depth study of deep neural network approaches for emotion recognition in HCI. It systematically reviews existing literature, discusses methodological frameworks, explores tools and technologies used for implementation, and analyses experimental results obtained from deep learning-based emotion recognition systems. Special emphasis is placed on multimodal emotion recognition and hybrid deep architectures, which have shown substantial improvements over unimodal systems. The study also highlights key challenges such as dataset bias, cultural dependency of emotions, real-time deployment issues, and ethical considerations. Finally, the paper outlines future research directions focusing on explainable artificial intelligence, edge-based emotion recognition, and emotionally adaptive intelligent interfaces
AI-Based Analysis of Microbial Communities for Climate Impact Prediction
Get PDFClimate change is one of the most pressing global challenges of the 21st century, influencing ecosystems, biodiversity, and human societies. Microbial communities play a central yet often underappreciated role in regulating Earth’s climate through their involvement in biogeochemical cycles, including carbon sequestration, nitrogen fixation, and greenhouse gas emissions. Due to their rapid response to environmental changes, microbial ecosystems serve as early indicators of climatic perturbations. However, the intrinsic complexity, diversity, and high dimensionality of microbial datasets pose significant challenges for conventional analytical approaches. Recent advances in artificial intelligence (AI), particularly machine learning and deep learning techniques, have demonstrated exceptional potential in modelling non-linear, high-dimensional biological systems. This paper presents a comprehensive AI-based framework for analysing microbial community data to predict climate impacts. By integrating metagenomic sequencing data with environmental variables, the proposed approach leverages unsupervised learning for microbial pattern discovery, supervised deep learning models for climate-variable prediction, and explainable AI techniques to enhance interpretability. The study highlights how AI-driven microbial analysis can significantly improve prediction accuracy of climate-related parameters such as soil carbon flux, methane emissions, and ecosystem resilience under climate stress. Results indicate that AI models outperform traditional statistical techniques and provide meaningful ecological insights. This research establishes a robust interdisciplinary framework that bridges microbiology, climate science, and artificial intelligence, contributing to improved climate forecasting, environmental monitoring, and sustainable policy formulation
An AI-Driven Framework for Intelligent Decision Making in IoT-Based Smart Systems
Get PDFThe rapid proliferation of Internet of Things (IoT) devices has led to the generation of massive volumes of heterogeneous data across smart environments such as smart cities, healthcare, agriculture, transportation, and industrial automation. Traditional rule-based and static decision-making mechanisms are increasingly inadequate to handle the scale, complexity, and dynamic nature of IoT ecosystems. Artificial Intelligence (AI), particularly machine learning and deep learning techniques, has emerged as a transformative enabler for intelligent, autonomous, and adaptive decision-making in IoT-based smart systems. This paper proposes a comprehensive AI-driven framework for intelligent decision making in IoT-based smart systems, integrating data acquisition, preprocessing, intelligent analytics, and automated action layers. The framework leverages supervised, unsupervised, and reinforcement learning models to extract actionable insights from real-time and historical IoT data. A modular architecture is presented, supporting scalability, interoperability, and real-time responsiveness. Experimental evaluation across representative smart system use cases demonstrates improved decision accuracy, reduced latency, and enhanced system efficiency. The study further discusses challenges related to data privacy, model interpretability, and resource constraints, and outlines future research directions toward explainable AI and edge intelligence
Machine Learning–Based Signal Classification for Brain–Computer Interface Applications
Get PDFBrain–computer interfaces (BCIs) translate brain activity into control signals for external devices. Electroencephalography (EEG) is the most widely used noninvasively modality for BCIs because of its portability and temporal resolution, but EEG signals are low-SNR, nonstationary, and highly subject-specific, making classification challenging. This paper reviews state-of-the-art machine learning (ML) methods for EEG/BCI signal classification, proposes a comprehensive end-to-end methodology combining preprocessing, feature extraction (CSP, time-frequency features), and modern classifiers (LDA, SVM, ensemble methods, CNNs), and presents a sample experimental pipeline using public motor-imagery datasets. Results show that deep models (CNNs) typically outperform classical shallow classifiers when sufficient data or transfer learning is available, while filter-bank CSP and transfer learning remain effective for limited data and subject-specific calibration. The paper concludes with practical recommendations, limitations, and future research directions in transfer learning, domain adaptation, explainability, and privacy for BCI systems
Smart Agriculture: Leveraging IoT and Machine Learning for Sustainable Farming
Get PDFThe increasing global demand for food, along with the challenges posed by climate change and limited natural resources, calls for a shift from conventional farming to more intelligent, data-centric methods. This study investigates the use of Internet of Things (IoT) devices, cloud computing, and Machine Learning (ML) algorithms to support sustainable agricultural practices. A dataset containing 10,001 entries—including variables such as environmental conditions, soil nutrients, and crop data—was analysed to forecast crop yield. Multiple regression models were tested, with the Random Forest Regressor delivering the highest accuracy at 98.48%, significantly outperforming baseline models like Linear Regression, which scored 76.42%. The integration of cloud services facilitates scalable, real-time data handling and allows efficient processing of sensor data alongside predictive modelling. This research highlights the effectiveness of ensemble learning methods and connected infrastructure in delivering actionable insights for precision agriculture. In order to increase productivity and ensure sustainable resource use, the suggested framework encourages more intelligent choices in areas such as crop planning, soil management, and yield enhancement
Exploratory Data Analysis on Cardiovascular Health Dataset
Get PDFcardiovascular diseases encompass conditions that impact the heart and blood vessels, with symptoms such as fatigue, dizziness, chest pain, discomfort, palpitations, and edema. Three major life-threatening conditions include high blood pressure, high cholesterol, and diabetes, which can lead to a diminished quality of life. The vast amount of healthcare data, or big data, contains valuable insights that can be extracted through Exploratory Data Analysis (EDA) to identify inaccuracies, locate pertinent data, verify assumptions, and assess the degree of association between exploratory factors. This is a crucial tool across industries for uncovering hidden patterns and forecasting future trends. This study examines a refined cardiovascular disease (CVD) dataset to identify clinical and demographic patterns linked to heart disease. The dataset includes 308,854 patients and 23 features, covering demographics (such as sex and age category), clinical variables (e.g., BMI, height, weight), health behaviors (e.g., smoking, exercise), and chronic conditions (e.g., diabetes, heart disease). Descriptive analysis showed that individuals with heart disease had a higher average BMI (29.6 vs. 28.5) and weight (86.9 kg vs. 83.3 kg) compared to those without. About 34% of patients were classified as obese (BMI > 30), indicating a significant at-risk group. Correlation analysis revealed a strong link between weight and BMI, with age showing a modest positive correlation with both BMI and weight. Boxplots indicated that patients with heart disease consistently had higher BMI and more extreme values, suggesting obesity as a major risk factor. K-means clustering analysis identified three distinct subgroups, potentially representing different risk profiles based on age, weight, and BMI. These findings highlight key variables and transformations such as obesity indicators, age-BMI interactions, and cluster memberships for future predictive modeling of cardiovascular risk. Overall, this paper underscores the significance of data analysis in healthcare and its potential to transform the industr
Reinforcement Learning Approaches for Energy-Efficient IoT Resource Allocation
Get PDFThe IoT has become a paradigm shift and already has connected billions of devices in the healthcare, transportation, production, and smart cities sectors. Since this growth is exponential, a great challenge has been provision of resources (particularly its energy efficiency). IoT devices are described as having low power, computing power, and bandwidth. The non-uniform and extremely dynamic nature of the IoT environment cannot be practically addressed using the classical optimization models. It can be quite promising to use the reinforcement Learning (RL) to attain autonomous and adaptive decisions in the resources allocation based on the data reduction without energy consumption. The article shall include a literature review of reinforcement learning systems to effectively distribute the IoT resources in terms of energy consumption. It introduces the theoretical models of RL, Markov Decision Process, Q-learning and Deep Reinforcement Learning (DRL) and applies them to maximize the power consumption, bandwidth allocation and offloading of computations. The paper discusses such popular RL-based architecture as Q-learning to dynamical spectrum accessing, Deep Q-Network to task allocation, and actor-critic architecture to power harvesting. It further talks about hybrid solutions using RL that could be used to solve the privacy and scalability problem by generalizing to non-metric type of edge computing and federated learning. It is revealed that the RL-based approaches is way better than the time-honoured heuristics since it accommodates the dynamical requirements of the network and consumes lesser powers but does not improve the performance of the Quality of Service (QoS). Scalability, speed of conversion, interpretability and practical application, however, remain an issue. As mentioned in the paper, reinforcement learning has been suggested as a strong paradigm to establish sustainable IoT ecosystems and that future research should also consider lightweight, explainable, and privacy-preserving instantiations of RL models, which can be implemented in the resource-constrained IoT setting
Ransomware Detection Using Machine Learning: Design, Analysis, and Review of Frameworks
Get PDFRansomware has become one of the most widespread and harmful types of cybercrime, disabling organizations and encrypting important data, which they then have to pay a ransom. As ransomware types are rapidly evolving, there is a growing degree to which signature-based techniques are ineffective. Machine learning (ML), and its capacity to learn based on patterns and to identify deviations, is a potentially effective solution to early detection and countermeasures of ransomware attacks. In this paper, a review of ransomware detection frameworks that use machine learning has been presented extensively. It studies both the analysis of the file (its features, sequences of opcodes), the analysis of the system (its behaviour, API calls, changes to registries), and a combination of both (hybrid methods). The accuracy, scalability and obfuscation resistance such as decision tree, random forest, support vector machine (SVM), and deep learning models consisting of CNNs and LSTMs are benchmarked. In this paper, the authors give the benefits of the ML-based detection, such as adaptive learning, reduced signature requirements, and zero-day ransomware, but also highlight limitations, such as data imbalance, adversarial example, and energy consumption. To beat these new solutions such as federated learning, explainable AI (XAI) or ensemble models, they are responded to. Recent studies have shown that ML models can be trained to have detection accuracy greater than 95% with balanced datasets, but adversarial manipulation remains a challenge. The paper also ends with a recommendation of future research directions such as privacy-preserving collaborative training, real-time lightweight ML based on endpoint protection, and blockchain integration to provide tamper-proof logging of ransomware activities
Innovative Strategies for Sustainable Environmental Management: AI and IoT-Based Approaches
Get PDFThrough their practice of environmental management people learn to protect resources they share with nature so future generations maintain sustainability. The recognition of sustainable environment preservation grew rapidly because of worsening climate change threats combined with growing pollution problems and dwindling resources along with declining biodiversity. Environmental degradation occurs as three modern issues unite population growth with industrial developments and urban construction activities. Worldwide governments along with organizations and communities strive to establish sustainable environmental management as their urgent mission to decrease environmental impacts. This paper introduces sustainable environmental management solutions by implementing IoT together with AI technology. To achieve maximum environmental impact experts in manufacturing should review combined technology applications for environmental challenges caused by climate change, pollution and resource utilization and waste management problems. AI alongside IoT enables organizations to develop innovative solutions which strengthen their operational excellence and maintain their sustainability initiatives
An Analytical and Systematic Review of Smart Farming's Challenges and Opportunities
Get PDFVarious industries have become more financially accessible due to technological advancements in various circumstances. Integrating Internet of Things technology in crop cultivation has shown benefits for multiple industries, such as agriculture and food production. The review paper below presents evidence of Internet of Things technology's impact on intelligent agriculture. This paper aims to review smart agriculture systems utilising Internet of Things-connected devices. The report has examined various essential aspects of smart agriculture and the advantages of Internet of Things technology. The review paper thoroughly discusses the different elements of the Internet of Things (IoT) technology. The application was found to have several areas for improvement, such as high cost, knowledge gap, and significant energy consumption. A rational discussion addresses the possible solutions to the raised issues. On the other hand, secondary qualitative methods, which use qualitative data, have facilitated discussions about the needs of smart agriculture. The paper shows significant knowledge about implementing Internet of Things systems in intelligent agriculture