International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
Exploring Stock Market Forecasting through Improved Machine Learning Methodology
Get PDFThis paper investigates the enhancement of machine learning methodologies for stock market forecasting, an area critically important for financial analytics and investment strategies. The study systematically compares traditional and advanced machine learning techniques to identify the most effective methods for predicting stock prices. Key components of the research include the utilization of ensemble methods, feature engineering, and deep learning algorithms, particularly Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM) networks, known for their proficiency in handling sequential data. The methodology encompasses a comprehensive preprocessing stage where financial data, including historical prices and volume, are cleansed and transformed into a machine-learnable format. Feature engineering is emphasized to extract and select temporal and technical indicators that significantly impact predictive accuracy. The research further explores the integration of ensemble methods that combine the strengths of various simple models to improve prediction reliability and accuracy
Genetic Twin Support Vector Based Movie Recommendation System
Get PDFMovie recommendation systems have become increasingly popular in recent years, as they can help users discover new movies that they are likely to enjoy. However, existing recommendation systems often suffer from limitations such as sparsity, cold start and overfitting. In this paper, we propose a new recommendation system called GATWSVM, which combines genetic algorithm with twin support vector machine to overcome these limitations. GATWSVM works by first constructing a user-movie rating matrix from a set of user-generated ratings. Then, it uses a genetic algorithm to evolve a population of classifiers, each of which is a twin support vector machine. The classifiers are trained to predict the ratings that users would give to movies that they have not yet rated. Finally, the system recommends movies to users by selecting the movies that have the highest predicted ratings.We evaluated GATWSVM on the MovieLens dataset and compared its performance to several state-of-the-art recommendation systems. Our results show that GATWSVM outperforms all of the other systems in terms of recommendation accuracy and precision
Study on Intelligent Power Electronics Dominated Grid Via Machine Learning Techniques
Get PDFIntelligent power electronics and machine learning algorithms are gradually reshaping the context of the more progressive electrical power grids today’s world. This abstract looks into how intelligent power electronics can be incorporated into grid systems and how control using machine learning can be applied. By using of complex algorithms, real-time analysis, these technologies improve temporal flexibility of the grid, its ability to prevent disruptions, and optimize the usage of renewable sources of energy. Intelligent power electronics can join forces with machine learning to provide completely new ways of managing the much-needed grid stability and low energy losses. The rapid emergence and evolution of power electronics has presented various challenges and opportunities in modern electrical grids. These include their ability to enhance grid flexibility and efficiency, but also their potential to introduce complex stability and control issues. This paper proposes a framework for addressing these issues using machine learning. The paper presents a comprehensive review of the current state of the art in machine learning and its potential to improve the stability and control of electrical grids. It proposes a framework that will help facilitate the transition to a more resilient and smart electrical system
Survey on Face Detection & Recognition Techniques
Get PDFFace detection is an extensively investigated subject in the realm of computer vision and holds considerable significance in diverse applications, encompassing human-computer interfaces, video surveillance, security access control systems, video surveillance, and image database management. Numerous face detection methods have already been devised, like Viola-Jones, RCNN, SSD among others. This paper discusses some additions that have been done on the existing models and systems in a bid to produce better outputs, using the standard datasets like WIDERFACE, FDDB etc. The face detection and recognition techniques discussed here employ the following approaches: (i) Mask R-CNN (ii) MTCNN, (iii) Local Binary Pattern Histogram (LBPH), (iv) PCA with Eigenfaces (iv) Weighted Kernel PCA, and (v) VGG architectures such as Siamese-VGG
Comparative Analysis of Primavera P6 and Microsoft Project: Optimizing Schedule Management in Large-Scale Construction Projects
Get PDFThis report presents a comparative review study of Primavera P6 and Microsoft Project tools focusing on their application in managing schedules for construction projects. These two are widely used in the construction industry to deal with time and resources and to track projects. The two platforms are compared in the context of this research with a focus on usability, features, cost, and compatibility. The goal is to comprehend the strengths of one software and the weaknesses of another and how construction managers could make the right decision on which software best fits a particular project type
Optimizing Security and Efficiency in Fog Computing: A Trust Management System Driven by Quality Matrix
Get PDFAs fog computing emerges as a natural extension of cloud computing, its decentralized nature brings numerous advantages, such as reduced latency and enhanced Quality of Service (QoS). However, this paradigm also introduces significant security and privacy challenges, particularly when fog nodes collaborate and exchange data. In this paper, we propose a robust trust management system that evaluates both Quality of Service (QoS) and Quality of Protection (QoP) metrics from direct and indirect interactions among fog nodes. Our approach helps mitigate security risks posed by potentially malicious nodes by incorporating a predictive trust evaluation system. The proposed system reduces malicious interactions by approximately 66% and enhances response times by reducing latency by around 15 seconds. The findings demonstrate that an effective trust management system is crucial for building secure and reliable fog computing environments
Optimizing Visual Content Representation Through Semantic Sparse Recoding
Get PDFThis study introduces a novel methodology for optimizing visual content representation through Semantic Sparse Recoding (SSR). By leveraging advanced sparse representation techniques and integrating a Global Dictionary Learning approach, the proposed system addresses limitations in conventional image fusion and content retrieval methods. The SSR framework improves the ability to preserve structural details and semantic features, particularly for multi-modal image datasets. Experimental results demonstrate the system's superior performance in terms of edge preservation, visual fidelity, and computational efficiency compared to state-of-the-art techniques. Applications span various domains including medical imaging, surveillance, and multimedia content management
A Systematic Review on Data Science and Artificial Intelligence Applications in Healthcare Sector
Get PDFThis systematic review focuses on the emerging domains of data science and AI in healthcare; the research question guiding the study was to establish the trends and gaps in literature. Over the past few decades, with the advancement of technology, various application areas of Artificial Intelligence in healthcare delivery has seen a drastic improvement from diagnosis to treatment and patient monitoring. Thus, the search included all the major databases in AI containing universal machine learning, natural language, and predictive analysis techniques. This review also divides these applications into several fields which include disease risk prognosis, medical image analyzing, and personalized medicine stressing that they will bring about improvements on the clinical diagnosis and patients’ care. Concerning the critical challenges it provides a response regarding such issues as ethical questionnaires, technical constraints, and prospects of spreading its application in spheres of healthcare. This issue is particularly relevant because ethical issues such as the protection of the patient’s identity, fairness in algorithm design, and the lack or appropriate regulatory measures become apparent. Another set of barriers relates to technical issues, including data conversion and the possibilities of using AI models in cancer diagnosis. Nevertheless, based on the aforementioned challenges, the review suggests potential directions of future studies and practical advancements that the authors and the healthcare community in general should consider in order to unlock the potential of data science and AI for improving worldwide healthcare systems
Analysis of Vision based Techniques for the Translation of Indian Sign Language
Get PDFSign language acts as a medium of communication among those of the hearing impaired and mute community. However, it cannot be easily understood by common people. Various research has been done to bridge this gap by developing Sign Language Recognition (SLR) methodologies. Studies say that 1 in every 5 deaf people is Indian.
In this paper, a thorough review of these methodologies has been done, to compare and contrast various aspects of them. This includes an overview on different preprocessing methods used like segmentation, image morphological processing, cropping, etc, feature extraction techniques like Fourier Descriptors, Image Moments, Eigen values, Mediapipe and others. This study also covered classification models spanning from Distance metrics to Kernel based approaches and feedforward neural networks, along with Deep Learning based methods such as CNNs, LSTMs, GANs, Transformers etc
A Survey on Deep Neural Network (DNN) Based Dynamic Modelling Methods for Ac Power Electronic Systems
Get PDFThis research work contains the assessment of a deep neural network (DNN) based dynamic modeling scheme for AC power electronic systems. The study is based on the premise of utilization of deep learning algorithms to derive models that are accurate and dynamic for capturing the aspects that are complex in AC power electronics systems. Nonlinear relationships and variability in operating conditions make it challenging to apply traditional modeling; thus, a possibility to apply DNNs is considered due to their data-driven learning compared to conventional feature-oriented techniques. It is a process of training and testing of the developed DNN models on the data sets, developed from the AC power electronic systems, under various operational conditions. Satisfaction is measured based on performance indicators that individuals employ, for instance, accuracy, resilience to different loads, and computational speed that justifies the proposed approach. As per the obtained results, the proposed DNN-based models for the four classes have better prediction accuracies compared to conventional techniques for real-time continuous control and fault diagnosis in AC power electronic systems. This work fits within the advancements of the field by offering a detailed evaluation of DNNs as a valid means for dynamic modeling within the scope of AC power electronics in order to critic and improve the performance and dependability of more practicable applications