Search
73 articles for “contrastive learning”
-
Multi-Layer Perceptron Neural Network Based Person Identification Using Appearance Based Facial Feature
Abstract: This paper deals with secure person identification system using Back-Propagation learning neural network algorithm where appearance based facial feature and Principal Component Analysis based dimensionality reduction technique have been used. To extract the appearance based facial features, Viola-Jones method has used as a face detector, Stam’s method of Active Shape Model has applied for detecting the facial edges and image pre-processing method has used for eliminating the background noises. For …
Published in Current Trends in Signal Processing · Vol. 4, Issue 1, 2014 · pp. 26–34 Read article
-
AI-Based Preventive Healthcare Using Quantum Computing
Abstract: With its improved performance and capabilities, quantum machine learning (QML) is becoming a promising field, especially in the healthcare industry for tasks like early heart disease prediction. In this work, a Quantum Support Vector Classifier (QSVC) is proposed as the basic classifier for a bagging ensemble learning model. Shapley Additive explanations (SHAP) are used to evaluate the significance of each attribute in the predictions in order to improve explainability. Using …
Published in Journal of Nanoscience, NanoEngineering & Applications · Vol. 15, Issue 2, 2025 Read article
-
Machine Learning-Based Disease Prediction: A Comparative Analysis for Diabetes, Brain Tumor, and Parkinson's Disease
Abstract: This paper presents a web-based disease prediction system that integrates machine learning and deep learning techniques to assist in the early detection of Parkinson’s Disease, Diabetes, and Brain Tumors. By utilizing clinical data and MRI images, the platform provides rapid and interpretable predictions to support proactive health management. Logistic Regression models are applied to classify structured datasets for predicting Parkinson’s disease and Diabetes, making use of their effectiveness in binary …
Published in Current Trends in Signal Processing · Vol. 15, Issue 2, 2025 · pp. 44–54 Read article
-
Country Cluster Visualization Based on Agricultural Imports: Unsupervised Learning Approach
Abstract: The clustering algorithm used in this analysis makes it easier for policymakers to understand performance metrics. K-means clustering has been demonstrated to be a tool for analyzing countries' agricultural imports and a visualization tool for interpreting the results in this study. Authors have used agricultural imports from 190 nations from Knoema, a web-based open data platform. Cereals, meat, and coffee imports from 190 nations in 2016 are included in the …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 11, Issue 1, 2024 · pp. 1–7 Read article
-
Solar Panel Defect Detection Using Geospatially-Aware Deep Learning framework
Abstract: Large-scale photovoltaic (PV) systems demand reliable inspection techniques to maintain efficiency, as manual methods remain labor-intensive and inconsistent. This study introduces a geospatially informed deep learning framework for defect detection and localization in PV panels from drone and satellite imagery. The framework incorporates an adaptive tiling mechanism that adjusts tile boundaries according to object size, reducing information loss and enhancing detection performance. In addition, coordinate transformation between image pixels and …
Published in Journal of Remote Sensing & GIS · Vol. 17, Issue 2, 2026 Read article
-
Machine Learning Based Sentiment Analysis of Student Feedback in Higher Education
Abstract: Educational institutions routinely collect feedback from students to understand their perceptions of academic programs, infrastructure, and campus facilities, to improve the overall quality of the college environment. In current practice, feedback is often gathered using numerical or grade-based rating systems, which tend to oversimplify student opinions and may overlook important details related to their level of satisfaction. In contrast, open-ended textual feedback allows students to clearly express their views, concerns, …
Published in International Journal of Data Structure Studies · Vol. 4, Issue 1, 2026 · pp. 01–10 Read article
-
Topology and Geometry in Data Science: Persistent Homology and Beyond
Abstract: In recent years, the interplay between topology, geometry, and data science has gained substantial momentum, offering powerful frameworks to analyze and interpret complex datasets. Traditional statistical and machine learning methods often rely on linear or metric- based assumptions, which may fail to capture the intrinsic structure of high-dimensional or nonlinear data. In contrast, topological and geometric methods provide shape-oriented, scale- invariant tools that focus on the continuity, connectivity, and global …
Published in Recent Trends in Mathematics · Vol. 1, Issue 2, 2024 · pp. 21–27 Read article
-
A Comparative Study of Deep Learning Methods for Depression Detection in Social Media Data
Abstract: With the rise of social media platforms like Twitter, Reddit, and Facebook, individuals increasingly share personal information about their moods, behaviors, and mental states. This trend provides a unique opportunity to leverage large-scale textual data for understanding and monitoring mental health conditions, particularly depression, a prevalent and challenging mental health issue. Traditional depression assessments are often confined to clinical environments and lack the capacity for real-time monitoring. In contrast, social …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 12, Issue 3, 2025 · pp. 55–65 Read article
-
Renewable Solar Energy Prediction in India : 2025-2030
Abstract: As India endeavors to realize its objective of achieving 500 GW of renewable energy capacity by the year 2030, the precise forecasting of solar power generation is rendered increasingly essential. This scholarly article conducts a comprehensive review of the utilization of machine learning (ML) methodologies in the prediction of solar energy across diverse Indian states, underscoring their potential to address the complexities associated with the variability of solar irradiance. Conventional …
Published in Journal of Power Electronics and Power Systems · Vol. 14, Issue 3, 2024 · pp. 41–46 Read article
-
Study on Single-Slope Solar Still for Experimental and Data-Driven Analysis for Improving Productivity with Different Basin Materials.
Abstract: This study investigates the single-slope solar still under the diurnal variation of water temperature and distillate yield under identical operating conditions. Experimental analysis was conducted to evaluate the performance enhancement through the incorporation of natural basin materials, namely hemp and sand. The water distillation process is focused on improving potable water productivity and thermal behaviour. The inclusion of hemp and sand in the basin leads to noticeable differences in productivity …
Published in Emerging Trends in Chemical Engineering · Vol. 13, Issue 2, 2026 · pp. 31–46 Read article
-
Refining Retinal Layer Segmentation in OCT Imaging with Advanced Techniques and Clinical Applications
Abstract: Segmenting retinal layers from Optical Coherence Tomography (OCT) pictures entails locating and separating different retinal layers to offer comprehensive anatomical and pathological information. Age-related macular degeneration, diabetic retinopathy, and glaucoma are among the retinal illnesses for which this procedure is crucial for diagnosis and follow-up. By utilizing preprocessing techniques to improve image quality and applying advanced algorithms—such as intensity-based, gradient-based, and texture-based methods—alongside deep learning approaches, clinicians can accurately measure …
Published in Trends in Opto-electro & Optical Communication · Vol. 14, Issue 2, 2024 · pp. 01–06 Read article
-
Artificial Intelligence and IoT Integration for Real-Time Violence Monitoring
Abstract: The peace and tranquility of any place can be affected greatly by the insurgence of violence and violent attacks that are perpetrated by individuals with malicious and nefarious intentions. These individuals terrorize the areas and can cause a lot of harm and damage to people and public property. The incidences of violence are undesirable and can be problematic to handle by the law enforcement agencies, as these acts are committed …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 13, Issue 1, 2025 · pp. 39–45 Read article
-
AbhiGyam: A Machine Learning Model-driven Research Platform for Assessing Accessibility Infrastructure in Indian Cities
Abstract: This work presents AbhiGyam, a machine learning-driven research platform designed to streamline and automate the assessment of accessibility infrastructure in Indian cities. AbhiGyam leverages the Google Maps API to transmit street view images to the backend, where computer vision techniques are implemented using OpenAI's CLIP (Contrastive Language-Image Pre-training) model to identify objects such as ramps, sidewalks, crosswalks, and parking spaces. The accuracy of the model is validated using labeled data …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 11, Issue 2, 2024 · pp. 84–91 Read article
-
Digital Audio Watermarking: An Overview
Abstract: AbstractA real watermarking procedure is required for copyright security and confirmation of protected innovation. This work recommends an advanced watermarking procedure which makes utilization of concurrent recurrence veiling to conceal the watermark data. The calculation is grounded on Psychoacoustic auditory model and Spread spectrum hypothesis. It creates a watermark sign utilizing spread range hypothesis and implants it into the sign by measuring the covering edge utilizing modified Psychoacoustic auditory model. …
Published in Journal of Communication Engineering & Systems · Vol. 5, Issue 1, 2015 · pp. 1–9 Read article
-
Improving Reliability of Deep Learning System Under Distribution Shift Via Feature-Space Out-of-Distribution Detection
Abstract: Deep learning systems achieve strong predictive performance under independently and identically distributed conditions; however, their reliability can decline significantly when exposed to distribution shifts during deployment. This decline is often accompanied by overconfident predictions, which can introduce serious risks in safety-critical applications. In this work, we explore whether explicitly modeling the structure of the latent feature space can improve reliability through more effective out-ofdistribution detection. A statistically grounded framework based …
Published in International Journal of Bioinformatics and Computational Biology · Vol. 4, Issue 2, 2026 · pp. 66–76 Read article
-
Content-based Image Retrieval: Recent Trends and Techniques
Abstract: Due to the affordability of digital devices and the accessibility of internet technologies, a vast number of multimedia databases have been established for various applications. These image databases increase the need for efficient picture retrieval search strategies that meet user requirements. When compared to other systems, the content-based image retrieval (CBIR) system is one of the most widely used systems for retrieving images from enormous databases. Much work has been …
Published in Journal of Image Processing & Pattern Recognition Progress · Vol. 12, Issue 1, 2025 · pp. 01–32 Read article
-
Evolution of CSS Frameworks: From Bootstrap to Utility-First Design
Abstract: CSS frameworks have become integral to modern web development, offering developers predefined styles and components that simplify the design process, ensure visual consistency, and accelerate deployment. Over the years, these frameworks have evolved significantly: from traditional component-based architectures, such as those seen in Bootstrap and Foundation, to more flexible utility-first methodologies like Tailwind CSS. This study explores the evolution of CSS frameworks by highlighting the contrasting approaches of these widely …
Published in Recent Trends in Programming languages · Vol. 12, Issue 2, 2025 · pp. 40–48 Read article
-
IoT Sensors to Monitor Pipeline Pressure and Flow Rate Combined with ML-Algorithms to Detect Leakages
Abstract: In the field of fluid mechanics, pipelines are the lifeblood of industries, transporting everything from natural gas and oil to water and chemicals. Maintaining their integrity is paramount for safety, economic efficiency, and environmental protection. Traditional leak detection methods explained in fluid mechanics can be slow, expensive, and sometimes fail to identify small leaks early enough to prevent significant damage. However, the convergence of Internet of Things (IoT) and Machine …
Published in Recent Trends in Fluid Mechanics · Vol. 12, Issue 2, 2025 · pp. 40–48 Read article
-
Learning of Maximum Power Point Tracking Architecture with Various Algorithms for Photovoltaic Systems: A Review
Abstract: Currently, as the requirement on the Earth for ever more electricity grows, so too accordingly must demands upon renewable energy. These days, with the growth of renewable energy on all fronts, countries everywhere watch its development. Since demand for power generation goes up again, fossil fuels become less and less available, and expense is not coming down. When there is a rapidly changing irradiance, temperature, or partial shading, the output …
Published in Journal of Power Electronics and Power Systems · Vol. 16, Issue 2, 2026 · pp. 14–22 Read article
-
Accessavior: AI-Driven Real-Time Adaptive Web Interfaces for Intelligent User Experiences
Abstract: As the web has become more diverse, many traditional, one-size-fits-all web accessibility solutions that must be configured manually have proven inadequate in meeting the needs of each user. This paper introduces “Accessavior”, which is an intelligent, behavioural pattern-driven accessibility framework that can automatically customize the personalization of web interfaces based on the interaction patterns of users. The proposed system will continuously track cursor behaviours such as cursor movements, keystroke characteristics, …
Published in Current Trends in Information Technology · Vol. 16, Issue 2, 2025 · pp. 14–22 Read article