International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
Dataset Creation and Comparative Analysis of Machine Learning Models for Mangrove Classification in Coastal Maharashtra, India
Get PDFThis research paper presents a new dataset of Landsat 8 image tiles processed for mapping mangroves in the coastal region of Maharashtra, India using bandcombinations of Band 5, 6 and 4. The dataset includes labelled image tiles which can beused for binary classification of mangroves using Convolutional Neural Network (CNN) and Random Forest algorithms.The radiometric correction of images and creation of composite images were done usingArcGIS Pro, while the image tiles were extracted and labelled using the Geotile library in Google Colab. The performance of CNN and Random Forest algorithms were comparedfor the classification of mangroves. This dataset can be used for further research on mangrove mapping and monitoring using remote sensing techniques
Generative AI in Financial Fraud Detection
Get PDFthe objective of this exploration is to utilize AI calculations to recognize Visa misrepresentation. Because machine learning algorithms can examine enormous datasets and uncover patterns suggestive of fraud, they present a promising method for detecting fraudulent transactions. An experimenta design is used in the research approach to gather, prepare, and analyze data. The study's dataset, which included 23 variables linked to financial transactions, was sourced from Kaggle. Feature engineering, addressing missing values, and removing superfluous columns were all part of the data pre-processing step. Three machine learning models — the Random Forest Classifier, K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) — were prepared and evaluated. The models were evaluated utilizing performance measures such area under the curve (AUC), review, exactness, accuracy, and F1-score. The review's discoveries show that Random Forest and KNN beat SVM in the distinguishing proof of Mastercard fraud. With an exactness of 99.63%, accuracy of 99.53%, review of 99.45%, and F1-score of 99.80%, Random Forest performed very well. With an exactness of 99.45%, accuracy of 99.87%, review of 99.89%, and F1-score of 99.80%, KNN performed all around well. With a F1-score of 99.61%, an exactness of 99.58%, accuracy of 99.74%, and review of 100 percent, SVM performed all around well. As per the examination, machine learning calculations might be helpful in recognizing Mastercard fraud. The capacity of machine learning calculations to recognize Visa theft is shown by this review. The results exhibit that Random Forest and KNN perform preferable on this task over SVM. The outcomes offer critical viewpoints for financial foundations and endeavors seeking to further develop their fraud detection systems. To expand the exactness of fraud detection, future examination could focus on exploring elective machine learning calculations and systems
Digital Transformation with SAP Hana
Get PDFThere is a significant reliance on digital technologies for innovation within the pharmaceutical engineering, medical device manufacturing, and biotechnology/biomedical engineering subsectors of the life science business. The life sciences industry has been notoriously slow to accept and implement digital services across the whole value chain, beginning with research and development and continuing all the way through commercialization. The most recent experience during the COVID-19 pandemic emergency proved that the life science business is the most crucial and time sensitive. Now, though, SAP has taken this on as a challenge, and its S/4HANA platform helps businesses in the life sciences digitally modernize their processes. Through the utilization of digital technologies such as blockchain, cloud computing, artificial intelligence, and the internet of things, the fourth generation of pharmaceuticals (Pharma 4.0) aims to enhance both transparency and efficiency by connecting patients and systems. Through the utilization of the robust SAP NetWeaver platform, you have the potential to lower your total cost of ownership by combining SAP with applications that are not SAP. In the following paragraphs, we will discuss the benefits and drawbacks of SAP S/4 HANA, as well as the manner in which it is transforming the life science business. Smart cloud ERP can accomplish a great deal of change. Time to value, security, compliance, scalability, ongoing innovation, and the capacity to integrate with other technologies will also be highlighted. Automatic and continual updates will also be provided
Fringe Benefits and Its Effect on the Wellbeing of Divine Word College of Legazpi Employees During the Pandemic
Get PDFThe COVID-19 pandemic has had a significant impact on employment, economic activity, and our way of life. Fringe benefits are extras that companies provide their workers in addition to their salary. The purpose of this study is to determine the Divine Word College of Legazpi, School of Engineering and Computer Studies employees' fringe benefits during the pandemic and to assess how such benefits affect their wellbeing. The study was conducted out using a qualitative descriptive research methodology. As a result, the DWCL SOECS faculty received benefits from the college's administration like clothes allowance and internet allowance. The majority of the faculty are happy with the services they have received from DWCL and have no complaints to make about the organization. On the other hand, some faculty members had issues with and concerns about the compensation and work they had gotten from DWCL. The DWCL SOECS faculty, administration, and management must hold consultation meetings so that they can express their concerns in a professional manner in order to preserve the finest possible education for the students that DWCL can provide
Machine Learning Methods for Prediction of Brain Tumors and Pneumonia Diseases
Get PDFPneumonia and brain tumors are considered critical diseases due to the substantial challenges related to accurate prediction and diagnosis at an early stage. Machine learning (ML) methods are used in medical imaging processing to detect specific patterns and features within input images and automatically classify various medical conditions. This paper aims to predict and classify pneumonia and brain tumors diseases, to compare the ML performance of methods: Decision Tree (DT), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forests (RF), Logistic Regression (LR), and Naïve Bayes (NB), and to analyze the impact of dataset increasing size on the classification performance. This study reveals that the Random Forest algorithm achieves the best performance, with 90% accuracy in the brain tumors dataset and 79% accuracy in pneumonia disease prediction
Design and Study of Fractional Order Circuits and Systems
Get PDFThe use of fractional calculus in electronic design has revolutionized the way systems and circuits are built. It provides a more accurate representation of complex dynamic systems than traditional integer-order methods. This paper presents a detailed analysis of the fundamental concepts of fractional calculus, including its mathematical properties. This paper reviews the key types of fractional order components, such as amplifiers, filters, and controllers. They are examined in terms of their practical implementations, design methodologies, and simulation techniques. For instance, the frequency response of a fractional order filter is precisely controlled, while the stability of a fractional order oscillator is enhanced. On the other hand, the high bandwidth and linearity of a fractional order amplifier can be enhanced using PID controllers
Review on Quantum Methods to Measure the Performance, Security, And Privacy for the Iot Framework
Get PDFMillions of devices worldwide, including smart appliances for domestic use such as smart TVs, thermostats, and CCTV cameras, were connected by the evolution of the internet. It was allowing the entire human community to connect and access devices from all over the globe through cloud infrastructure. Increasing IoT devices weaken classical model computing performance, making the IoT framework a failure model. Another big problem for IoT devices is keeping personal information safe and private because attackers can get in while the devices are talking to each other. The safety measures we have now, like changing passwords often, keeping IoT devices up to date, using a backup network (VPN), and not using plug-and-play features, work up to a point but can't promise that everything is 100% safe. Because of new technology, the public key cryptography method (RSA), which is thought to be safe right now, is being used more and more. One Time Pad (OTP) is thought to be a good way to encrypt information securely in classical cryptography, but it takes too long to send the key between parties. The trade-off between adding more IoT devices and traditional security methods isn't fair, so we needed a new technology to fix the issue. Modern innovations called quantum computing is being developed. This will help find long-term solutions for the speed and security problems that affect millions of IoT devices. Quantum technology operates on the principles of quantum physics, as opposed to existing technology, which is based on conventional physics. While classical cryptography utilizes deterministic bits that can be hacked, quantum cryptography uses qubits, which are superpositions of 0 and 1. The measurement of an unknown qubit, which provides ½ probabilities of measuring in either bit '0' or bit '1', complicates the prediction of the data. Grover's algorithm searches an unsorted databases in O(?n), as opposed to O(n) in classical algorithms, and Shor's algorithm in quantum computing solves factoring a huge integer in polynomial time, compared to quadratic time in conventional models. The privacy problem and the need for high-level security can be addressed by applying the laws underlying quantum physics and such as the Heisenberg principle and the no-cloning theorem, to detect when an unauthorized user is involved in a current connection. Heisenberg's uncertainty principle says that you can't measure quanta that aren't known without upsetting them, which means that a third party has to be involved. It is impossible to create a copy of the unknown states due to another quantum characteristic called the No-cloning theorem. Therefore, an enemy cannot copy the quantum state in order to read the quantum information in safe communication. Communication in quantum computing can also happen through entanglement, in which a third party creates an entangled photon and sends a qubit to the parties talking to each other. If one party measures a qubit, it will match up with measurements made by the other party. This creates a safe key. When two parties communicate directly and securely without the use of a key, it's known as quantum secure interacting directly, or QSDC. Quantum and conventional security should be combined for high-level IoT security
UI Testing, Mutation Operators, And the DOM in Sensor-Based Applications
Get PDFIn the era of widespread dependence on web applications, ensuring their stability is paramount for seamless digital experiences. While UI testing is acknowledged as crucial for delivering user satisfaction, the prevalent approach of manually creating web application UI test suites using Selenium-compatible technologies lacks a systematic method for evaluating their bug-finding capabilities. This study introduces a groundbreaking Test Case Coverage Model with Priority Constraints (TCCM-PTWA) to address the challenges of mutation testing in online applications. Diverging from conventional mutation testing that primarily targets source code, our approach operates within the Document Object Model (DOM) of web browsers. This innovative technique eliminates the need for source code alterations, ensuring compatibility across a diverse array of online applications. The incorporation of priority constraints in TCCM-PTWA enhances the testing procedure by ranking test cases based on their significance, optimizing resource allocation, and minimizing testing overhead. Additionally, we present a set of mutation operators tailored specifically for web applications, drawing inspiration from common web application flaws. These operators are designed to replicate real-world issues, thereby increasing the effectiveness of mutation testing in practical scenarios. Through an empirical review encompassing various sensor based applications, we demonstrate the efficacy of TCCM-PTWA in evaluating test suites and identifying faults, with priority constraints contributing to the overall reliability and resilience of online services. This study introduces a pioneering Test Case Coverage Model with Priority Constraints, focusing on UI testing, MAEWU (Mutation Analysis for Web Applications with Emphasis on UI), and the DOM. The methodology presented herein addresses the unique challenges posed by online applications, offering a comprehensive solution that improves the reliability and resilience of web applications in the digital age
Enhancing Healthcare Security with Chaotic Reversible Watermarking for Medical Images
Get PDFThe implementation and advancement of digital technologies in medicine have sparked a technical revolution, leading to new developments across various medical fields. Ensuring the confidentiality of Electronic Health Records (EHR) and maintaining patient privacy are critical security requirements for healthcare systems. In today’s digitally connected world, malicious tampering of digital images has become increasingly common, representing a serious violation of intellectual property rights. Consequently, protecting images by establishing rightful ownership has become a priority. Reversible watermarking (RW) techniques offer a promising alternative to conventional watermarking systems, particularly for safeguarding highly sensitive images. It is crucial to authenticate the source and origin of medical images and associated patient information to ensure they correspond to the correct patient. This work presents a method for protecting medical images and patient-related information in healthcare using a chaotic reversible watermarking technique. The performance of this method is evaluated using metrics such as Mean Absolute Error (MAE), Peak Signal-to-Noise Ratio (PSNR), Normalized Cross-Correlation (NCC), and Root Mean Square Error (RMSE)
Challenges in Implementing Machine Learning-Driven IoT Solutions in Semiconductor Design and Wireless Communication System
Get PDFhe integration of machine learning and Internet of Things has great potential to revolutionize several industries including semiconductor design and wireless communication techniques as it can. However, the application of ML in the IoT poses great problems such as compatibility with existing systems, real-time decision making, scalability and cost. This research seeks to investigate these challenges by identifying major technical and operational impediments that hinder IoT based ML application in these industries.
The current research adopted the quantitative research approach and the data was collected through an online questionnaire completed by the professionals in the semiconductor and wireless communication industries. Questionnaire was distributed and filled by 300 participants, to evaluate their experience about ML-IoT integration. The survey encompasses a number of aspects such as technical issues, practical problems, the questions of expansion of the usage and questions of data protection. Quantitative data were analyzed descriptively and inferentially employing Chi-square test, ANOVA, multiple regression analysis, correlation analysis. These techniques enabled the assessment of the issues and interactions between the factors of industry focus, company size and ML-IoT implementation.
The article concluded that the main challenges still persists and they are chiefly evidenced in the issues of legacy system integration whereby semiconductor design still finds itself in the lager of hauling old architectures that cannot support computation of today’s convolutional ML algorithms. The identified key issue in wireless communication environment was real-time decision-making because it could not afford a time delay in processing of large amount of data when required. The issue of scalability was identified to be widely affecting both industries as they attempt to handle the increasing amount of IoT data in efficient and performant manners. Further, the data privacy and security were reported to be slightly higher in the wireless communication system and the participants called for enhanced legal safeguards and privacy-preserving methodologies in the ML domain.
The regression analysis revealed that the difficulty level of the ML-IoT challenges was dependent on the size of the organization and the number of years spent on such projects: While larger organizations had more potential to come to terms with the issues of scalability of the projects, they required in order to advance, they still struggled with the costs involved in the projects and the matter of data privacy. The article suggests that the use of techniques like edge computing and federated learning can eliminate the challenges posed by real-time processing while cloud environments can also be the possibility of cost-saving. More emphasis must be placed on security solutions that facilitate data privacy such as privacy-preserving technologies and commitment to more technological advances for creating more scalable ML models to be used in IoT systems