International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
Modified EPPXGBOOST for Effective Data Stream Mining in Cloud
Get PDFIn today’s technology-driven landscape, the perva- sive use of online services across diverse domains has led to the generation of vast datasets, necessitating advanced data mining techniques for meaningful insights. The advent of data streams, characterized by continuous and dynamic data flows, presents a significant challenge, prompting the evolution of data stream mining. This field addresses issues such as rapid changes in streaming data and the need for quick algorithms. To tackle these challenges, an innovative approach named (Effective Privacy Preserving eXtreme Gradient Boosting) EPPXGBOOST is proposed, combining Adaptive XGBOOST for continuous learning from evolving data streams with PPXGBOOST for privacy preservation
Exploring Sentiment Analysis in Social Media: A Natural Language Processing Case Study
Get PDFSocial media plays an integral role in our daily lives, influencing and reflecting global perspectives through the consumption and creation of content. Platforms like YouTube are incredibly active, with a constant influx of video uploads, views, and comments. While the YouTube app allows us to browse videos and comments, it offers only a limited glimpse into the interests and trends of others. Analysing this vast data pool, encompassing diverse language styles, presents a significant challenge. This article delves into the YouTube Data API and its application in Python for accessing raw data. The process involves data cleaning using advanced Natural Language Processing (NLP) techniques, harnessing Python-based machine learning to explore social media interactions, and automating the extraction of trends and influential factors. The journey towards trend analysis is meticulously detailed, featuring examples that leverage a variety of open-source Python tools
Evaluating the Efficacy and Security of Steganography Techniques in Cloud Computing
Get PDFCloud computing has revolutionized the handling, access, and storage of data. However, ensuring the security and privacy of data within cloud environments remains a significant challenge. This research offers a comprehensive examination aimed at developing a dependable steganographic technique, evaluating its effectiveness, determining its impact on cloud computing performance, and exploring potential vulnerabilities along with solutions. The primary objective is to assess how steganography influences the performance of cloud computing systems. Through performance evaluations and benchmarks, the study investigates the effects of steganography on system performance. It utilizes various file formats and sizes to simulate real-world conditions. The research also identifies potential weaknesses of the proposed steganographic method and explores strategies to mitigate these vulnerabilities. An analysis of security vulnerabilities, including potential attacks and detection techniques, leads to the formulation of effective countermeasures. The findings of this research contribute to the advancement of steganography-based data security in cloud environments. By highlighting the strengths, weaknesses, and areas for improvement of the proposed steganographic approach, the study offers insights into its impact on cloud computing systems and supports the development of robust security measures
A Mobile Application Framework to Classify Philippine Currency Images to Audio Labels Using Deep Learning
Get PDFThis research presents a mobile application framework designed to empower visually impaired individuals in Legazpi City by providing real-time audio feedback for currency identification. Leveraging deep learning techniques, the proposed framework employs a robust model trained on a comprehensive dataset of Philippine currency images. The deep learning model is capable of accurately classifying various denominations of bills and coins, enabling the development of an inclusive solution for the visually impaired community. The researcher employed a qualitative approach in this study, which included a focus group discussion. Respondents were chosen using purposive sampling. Among those who responded were masseuses, chiropractors, herbal street vendors, and students. Through an online meeting, the selected participants contributed to the focus group discussion. In addition, an in-depth informal interview was conducted to gather additional information for the development of an architectural framework. Based on the result of this study, it was discovered that by implementing this architectural framework, these groups would be able to more easily identify money, increasing efficiency and reducing errors in cash transactions. The use of audio labels is particularly helpful for visually impaired individuals, as it provides an accessible way for them to independently handle and identify money
Prediction of Heart Disease Using Machine Learning Techniques
Get PDFA potential strategy in the healthcare industry is the prediction of cardiac disease using machine learning algorithms. Worldwide, heart disease continues to be one of the major causes of death, and successful treatment and prevention depend greatly on early identification. Large volumes of patient data may be analyzed using machine learning algorithms to find patterns and risk factors that might lead to the onset of heart disease. These algorithms use supervised learning, unsupervised learning, and ensemble approaches to assess a variety of data sources, including clinical test results, patient demographics, and medical records. Machine-learning algorithms may be trained on historical data from a variety of patients to discover complicated associations and generate precise predictions about a person's risk of acquiring heart disease. Our objective is to create a machine-learning technique that reliably predicts heart disease and is computationally effective. Feature selection is a crucial step in the creation of prediction models as it permits the identification of the most significant risk factors for heart disease. Machine learning methods including logistic regression, support vector machines, decision trees, random forests, and neural networks are often used to predict cardiac disease. By examining extensive patient data, machine learning algorithms show considerable potential in the prediction of cardiac disease. In the battle against heart disease, their capacity to spot patterns and risk factors may result in early identification, individualized therapies, and better patient care
A Deep Learning Approach to Video Classification for Indoor and Outdoor Environments
Get PDFThis research paper explores the application of deep learning techniques for video classification, specifically focusing on distinguishing between indoor and outdoor environments. We present a comprehensive analysis of different deep learning models and methodologies used for this classification task, evaluating their performance and effectiveness. Our study includes a detailed exploration of feature extraction methods, model architectures, and training strategies tailored to indoor-outdoor video classification. Through extensive experimentation and evaluation on benchmark datasets, we demonstrate the efficacy of our proposed approach, achieving significant accuracy rates and outperforming existing methods in this domain. The findings from this research contribute valuable insights and advancements in video classification using deep learning, with potential applications in various real-world scenarios such as surveillance, robotics, and environmental monitoring
Integration of Artificial Intelligence in Academic Teaching Practice: An Analysis in the University Environment.
Get PDFAt the Universidad Autónoma de Coahuila, the integration of artificial intelligence has become essential in faculty academic research. This study examines how the introduction of this technology transforms the search process, highlighting its influence on the efficiency and quality of the resources discovered. The research focused on identifying the factors associated with the adoption of artificial intelligence, evaluating the frequency of its use through a case study at the Autonomous University of Coahuila, Mexico. A two-phase survey was conducted using a five-point Likert scale. The results reveal that the factors delineated in the Technology Acceptance Model (TAM) influence teachers' behavior regarding the adoption of artificial intelligence. These factors include lack of knowledge, insufficient training, resistance to change, and barriers to implementatio
Transforming Image Captioning: Refining Models with Advanced Encoder-Decoder Architecture and Attention Mechanism
Get PDFImage captioning involves the generation of textual descriptions that describe the content within an image. This process finds extensive utility in diverse applications, including the analysis of large, unlabeled image datasets, uncovering concealed patterns to facilitate machine learning applications, guiding self-driving vehicles, and developing software solutions to aid visually impaired individuals. The implementation of image captioning relies heavily on deep learning models, a technological frontier that has simplified the task of generating captions for images. This paper focuses on the utilization of encoder-decoder model with attention mechanism for image captioning. In classic image captioning model, the words usually describe only a part of the image, however with attention mechanism special attention is given to the low level and high-level features of the image. With the use of stable dataset and improvised encoder – decoder modelling, it is possible to generate captions having an accurate description of image with CIDEr score more by 16.52% of established models
Energy Efficient Wireless Sensor Activities in Computer Networks
Get PDFA Wireless Sensor Network (WSN) can be described as a sophisticated ensemble of interconnected devices that collaborate to relay information collected from a designated observation area. This network architecture enables the transmission of data across various nodes, ultimately converging at a gateway that integrates the data into larger networks, such as wireless Ethernet. Essentially, a WSN consists of base stations and numerous nodes equipped with wireless sensors. Modern iterations of these networks support bi-directional communication, not only facilitating the collection of sensor data but also allowing for the remote control and adjustment of sensor operations. Initially spurred by military needs for comprehensive battlefield surveillance, the utility of wireless sensor networks has expanded significantly. Today, they are integral to a variety of both industrial and consumer contexts, ranging from monitoring and controlling industrial processes to assessing the condition of machinery in real time
Automatic Writer Identification of Historical Kannada Handwritten Palm Leaf Manuscripts using AlexNet Deep Learning Approach
Get PDFAncient manuscripts have been a rich source of archeological information for decades, and some of the researchers have trying to develop Machine learning(ML) and Deep learning(DL) tools to restore the degraded information from ancient manuscripts. Today, the challenge lies in cataloging the manuscripts based on categories like subject, title, author, place, language, and script. The proposed study presents an automated deep learning model developed using the AlexNet CNN concept to classify and organize the historically significant Kannada handwritten manuscripts based on the various authors. Specifically, the AlexNet CNN approach is used to classify old Kannada handwritten manuscripts according to authorization. Old manuscripts present a unique set of challenges because they are historically significant, contain a variety of styles, and contain damaged text. The proposed research intends to present a strategy that employs deep learning techniques to attribute writing to older Kannada writings automatically. The proposed method shows promising results in several areas, i.e., the model's overall average of accuracy is 99.85%. The accurate assignment of publications to their respective authors. The classification model's performance for each class