Journal of Science & Technology (JST)
Not a member yet
967 research outputs found
Sort by
Enhancing Oracle Cloud HR Reporting Through AI-Driven Automation
Get PDFThe purpose of this study is to focus on AI, including ChatGPT, that is being used in Oracle Cloud HR reportingsystems. The system takes care of report generation and supports easier access for users to make critical HR decisions.The research points out the present imperfections of the system, the positive side of using conversational AI, and themajor difficulties in adopting it. It seems AI adoption enhances the HR reporting functions by increasing efficiency,better experiences, and quicker reactions in the cloud
Blockchain-Assisted Federated Learning for Cybersecurity: Combining Isolation Forest, Variational Autoencoders, and Differential Privacy
Get PDFThe complexity of the cyber threats dictates the need for strong, privacy-preservingmechanisms for anomaly detection. This paper introduces a new framework called BAFL, anintegration of Isolation Forest and Variational Autoencoders combined with DifferentialPrivacy, for safe and scalable solutions in cybersecurity applications. Federated Learningallows distributed training across numerous clients without exposure of sensitive information,while the blockchain technology introduces trust and integrity in model updates
Use of human hepatocytes from the second generation of the upcyte® brand in investigations of CYP inhibition and induction
Get PDFHistologic features shared by primary human hepatocytes are maintained by proliferating hepatocytes from the human upcyte® line. The use of four donors' second-generation upcyte® hepatocytes in inhibition and induction tests with a variety of reference inhibitors and inducers was thoroughly evaluated. Reproducible inhibition of CYP1A2, CYP2B6, CYP2C9, and CYP3A4 occurred at concentrations ranging from very low to very high, and the IC50 values computed for each chemical accurately designated them as powerful inhibitors
Privacy and Data Protection: Ensuring Compliance with Federated Learning in the Digital Age
Get PDFToday's digital age has made privacy and data protection a major concern-generally, with the kind of technologies that are turning things around and bringing everything to the cloud. FL will most likely provide a solution to the distance and make things clear in collaboration without exposing raw information from a consortium to boost its privacy. However, existing FL solutions include such challenges as increased overhead communication, risk in leaking data, and even the inefficiency of secure aggregation. To mitigate these constraints, this research proposes the Autoencoder-Based Federated Learning framework by integrating prevailing techniques such as differential privacy and homomorphic encryption that safeguard both the security and efficiency of the model. This method does not only steal model ideas for autoencoders to compress before sciences transmission but hugely reduces the transmission bandwidth and possibly minimizes gradient leakage. However, adaptive normalization is used to handle institutional heterogeneity to maintain better performance for the model. Conclusion of experimentation indicated that this framework could significantly reduce communication overhead while retaining high federated learning accuracy and even better security. Further, the trust-based client evaluation mechanism is presented to detect malicious behavior and improve reliability regarding federated aggregation. The experiment showed that Autoencoder Based Federated Learning was a scalable, secure, and privacy-efficient solution to applications tailored for healthcare, finance, and other sensitive data environments
Fortifying Cloud Security with Advanced Data Encryption Technique
Get PDFThe rapid growth of cloud computing has introduced several challenges regarding the securing of sensitive data.Traditional encryption methods such as those proposed by AES and RSA are unable to efficiently perform withthe scale of large cloud environments as they have high computational cost. Recent advancements that have beenmade in encryption methods, especially concerning homomorphic encryption, appear to unravel an unprecedentedpotential since they provide capabilities for performing computations on encrypted data without the need todecrypt it thus assuring the privacy and integrity of data. However, they are still associated with addingcomputational overhead, and that will definitely pose various challenges for real-time cloud data processing. Thesetup proposed in this paper is a complete framework that integrates homomorphic encryption within a cloudsecurity environment. It evaluates the effectiveness of homomorphic encryption in the cloud for aspects pertainingto performance and security, especially in terms of scalability as well with processing huge amounts of sensitivedata while ensuring much efficiency in performance. Further, the framework includes some prior processing likenormalization so as to optimize efficiency in encryption performance. A comprehensive security analysis isundertaken toward measuring the resistance of such encryption under numerous attack scenarios, and the effectof quantum computing applications on the proposed method of encryption is also discussed in this regard. Thispaper presents a thorough study of performance in conjunction with security trade-offs and, the overalldevelopment of a secure and efficient cloud data processing model
HISTORIC MILESTONES ON THE PATH OF THE BIRTH OF MICROSCOPY FROM LIGHT TO ELECTRON MICROSCOPY, AN INTERDISCIPLINARY TECHNIQUE. PART1 LIGHT MICROSCOPY
Get PDFMicroscopes drove theories by providing the tool needed to make advances in many fields of science. Observations through the microscope primarily determined what early scientists thought and advanced science to a great extend as we know it today across many fields of science, becoming a dominating technique. What is captured here are some of the major milestones on the path of the birth of microscopy. The following is covered: The beginning of the light microscope, Early Stereo Microscope development, The Ultra Microscope, Differential interference contrast (DIC). Inverted light Microscope, The Fluorescence microscope, Confocal Microscopy, Confocal Scanning Light Microscope (or ‘CSLM’), Transmission Electron microscope, Scanning Electron Microscope (SEM) and the Environmental scanning electron microscopy (ESEM). With more older manuscripts becoming available digitally, this overview fills in many gaps that exist in some of the previous work, covering a wider history in microscopy. It is also the single most recognised technique, which in its development and application, has produced more Nobel prize laureates than any other technique. Quite a few laurates are born and imbedded in the history of microscopy, some are covered here
Performance Evaluation of Red Onion (Allium Cepa) and Ginger (Zingiber Officinale) Extracts as Low-Dosage Green Inhibitors for Gas Hydrate Formation
Get PDFGas hydrates present a significant challenge for the production, handling, and transportation of natural gas. This study focuses on experimentally investigating selected plant extracts as eco-friendly alternatives to chemicals for inhibiting gas hydrate formation. Specifically, the research explores the effectiveness of red onion (Allium cepa) and ginger (Zingiber officinale) extracts as biodegradable and water-soluble Low Dosage Hydrate Inhibitors (LDHIs).Experiments were carried out using a Mini Flow Loop to evaluate the effectiveness of these natural extracts compared to traditional chemical inhibitors. The findings indicate that both red onion and ginger extracts significantly reduce hydrate formation rates and increase induction times, demonstrating promising properties as hydrate inhibitors. Performance comparisons show that these natural extracts are comparable to conventional inhibitors, highlighting their potential as viable, environmentally friendly alternatives.This research emphasizes the importance of utilizing sustainable and non-toxic solutions to mitigate hydrate formation, which could lead to reduced environmental impacts and lower operational costs in the oil and gas industries
Neuroprotective Potential of a Polyherbal Formulation in Enhancing Mental Clarity and Emotional Stability
Get PDFChagas is one of the most common illnesses in Latin America. Chagas and Trypanosomacruzi are treated with benznidazole and nifurtimoxis, but these treatments are unsuccessful in the chronic phase of epimastigote and El Salvador. Thus, there is an urgent need for the creation of new medications, particularly for the chronic stage of the illness, according to the author. We examined 114 plant species from the Salvadoran flora using the in-vitro anti-Marvin J. Núñez trypanosome assay, and 34 active plants against the stepimastigote of Trypanosomacruzi Research Scholar were foun
AUTOMATING E-GOVERNMENT POLICIES HAND WRITTEN DIGITS RECOGNITION AND TEXT & IMAGE BASED SENTIMENT DETECTION USING AI
Get PDFIn this paper, author describes the concept of predicting insurance policy charges and user opinion sentiment on policies by applying machine learning and artificial intelligence. Machine learning can automatically predict future values by analyzing past historical data, and artificial intelligence will take decision as human brain (as our brain help us in making decision as working hard if marks are less, or taking it easy). Additionally, by analysing male and female BMI index AI and machine learning can predict insurance policies and their charges. This AI and machine learning can also analyze users' opinions or reviews and then it will take decision as whether their opinion is positive, negative, or neutral.
 
Real and Fake Currency Detection using ANN
Get PDFAbstract: Currency is the main form of exchange in India and is essential to the country's financial system, social progress, and economic growth. In today's highly technologically advanced culture, counterfeit money is a major problem since paper money is easy to move and safe to have, but its face value is significantly higher than its actual value.
Acknowledging the significance of preserving economic advancement, the Indian government has implemented policies like the demonetization of the Rs. 1000 and Rs. 500 notes. Though it becomes a possible target for counterfeiters, the introduction of the Rs. 200 note and the revised design for the Rs. 500, Rs. 100, Rs. 50, Rs. 20, and Rs. 10 notes provide new problems. The main problem with the hardware-based methods now in use for detecting counterfeit notes is that they are challenging for average people to use, even with their competence.
This essay looks at the characteristics that set the new legal tender from the Reserve Bank of India apart and uses methods to identify and confirm that it is real. a hybrid method that combines an ANN with a ResNet model, some architecture, and ANN adjusted parameters to accurately identify counterfeit money. This method employs residual networks (ResNet 50) and artificial neural networks (ANN) to recognize counterfeit currency based on its width, color, and serial number. To distinguish authentic notes from fakes, the system goes through a number of processes, such as pre-processing, segmenting, comparing, and extracting attributes from pictures. Finding counterfeit money is still a difficult and complex process. General Terms: Image Processing, Feature Extraction, Detection