Category Archives: Uncategorized

Authors: I.Sravani, D. Lakshmi, M.Ushaswini, L.Aswini, C. Subramanyam

Abstract: Malware is any type of program that is intended to wreak havoc to the computer system and network. Examples of malware are bot, ransomware, adware, keyloggers, viruses, trojan horses, worms and others. The exponential growth of malware is posing a great danger to the security of confidential information. The problem with many of the existing classification algorithms is their low performance in term of their ability to detect and prevent malware from infecting the computer system. There is an urgent need to evaluate the performance of the existing Machine Learning classification algorithms used for malware detection. This will help in creating more robust and efficient algorithms that have the capacity to overcome the weaknesses of the existing algorithms. This study did the performance evaluation of some classification algorithms such as J45, LMT, Naïve Bayes, Random Forest, MLP Classifier, Random Tree, REP Tree, Bagging, AdaBoost, KStar, SimpleLogistic, IBK, LWL, SVM, and RBF Network. The performance of the algorithms was evaluated in terms of Accuracy, Precision, Recall, Kappa Statistics, F-Measure, Matthew Correlation Coefficient, Receiver Operator Characteristics Area and Root Mean Squared Error using WEKA machine learning and data mining simulation tool. Our experimental results showed that Random Forest algorithm produced the best accuracy of 99.2%. This positively indicates that the Random Forest algorithm achieves good accuracy rates in detecting malware.

DOI: https://doi.org/10.5281/zenodo.18921650

Published by: vikaspatanker

Authors: Shruti Dhage, Heena Barach, Sanakruti Jadhav, Vaishnavi Shivsharan, Shravani Pichake, Suchita Barkund

Abstract: Authentication is a critical component of digital systems, ensuring that only authorized users gain access to sensitive information and services. Traditional text-based password mechanisms, while widely used, suffer from vulnerabilities such as weak password selection, reuse across platforms, and susceptibility to brute-force and phishing attacks. To address these issues, this research presents the Graphical Password Authentication System, a web-based platform designed to enhance security by combining conventional password hashing with graphical pattern verification. The proposed system is developed using Java Server Pages (JSP), Servlets, MySQL database, HTML, CSS, and JavaScript, and deployed on the Apache Tomcat server. It includes features such as secure user registration, SHA-256 password hashing, graphical password setup and validation, OTP-based password recovery, and session management with duplicate login prevention. By introducing a dual-layer authentication mechanism, the system reduces risks of impersonation and unauthorized access while providing a user-friendly interface. The implementation demonstrates how graphical authentication can strengthen digital identity management and improve usability in academic, corporate, and community environments.

Published by: vikaspatanker

Authors: Ms.S.Kanimozhi, P.Guruprasaath, S.Lingaraj, S.Kanagaraj

Abstract: In modern healthcare, patient medical records are often distributed across multiple hospitals, resulting in repeated medical tests, delayed diagnosis, and poor continuity of care. This project proposes a Reference Electronic Health Record (EHR) System that provides secure and centralized access to patient records across hospitals. The system includes two login modules: users (patients) can view their medical history, disease descriptions, and prescriptions, while hospitals manage multiple doctors who select their own name to add new records for specific users. All medical records are maintained in a reference based manner, ensuring that previous information remains preserved, while doctors from different hospitals can view earlier records as reference to support accurate diagnosis and treatment. The system promotes better coordination among healthcare providers by maintaining a consistent and complete patient medical history. By offering structured storage and role-based access to sensitive medical information, the system enhances continuity of care, minimizes redundancy in medical testing, and improves overall efficiency within the healthcare process.

DOI: https://doi.org/10.5281/zenodo.18920808

Published by: vikaspatanker

Authors: Shravani Mali, Yukta Koli, Mayuri Mohite, Nilam Patil

Abstract: This study presents an AI-driven allergy checker designed explicitly for skincare. It reviews the user's allergies, skin type, and skin conditions and suggests a product and skincare routine based on those factors. The random forest model is used to classify skin type while the Light GBM model evaluates the skincare routine recommendations. Then a K-Nearest Neighbors (KNN) algorithm uses the allergy information the user provides to make the recommendations. A YOLOv8 model also analyzes the image the user provides and determines if there are skin conditions visible to the naked eye. In review, the system developed is able to provide appropriate personalized data-driven recommendations for skincare products and routines with a lower likelihood of allergic body complications, while also allowing for informed selections of skincare according to allergenic history.

Published by: vikaspatanker

Authors: Mrs. Meenakshi, Dr. Brij Mohan Goel

Abstract: Artificial Intelligence (AI) and Machine Learning (ML) have significantly transformed the field of healthcare, particularly in cancer detection, diagnosis, and treatment. With the advancement of digital pathology, large clinical datasets, and powerful computational techniques, AI has become a crucial tool in oncology research and clinical practice. Deep learning algorithms can analyse high-resolution histopathology images, genomic data, and electronic health records to detect patterns that may not be visible to human experts. These technologies enable early cancer detection, risk prediction, accurate diagnosis, and personalized treatment planning. Additionally, AI-based approaches such as Natural Language Processing (NLP), radiomics, and biomarker discovery have enhanced the analysis of complex medical data. Cloud-based AI platforms further facilitate large-scale data processing and collaborative research. Despite these benefits, the integration of AI into cancer care also faces technical and ethical challenges, including data privacy concerns, lack of standardized datasets, algorithm bias, and interpretability issues. This paper explores the applications of AI in cancer prediction, diagnosis, and treatment while discussing the technical and ethical challenges associated with its implementation. The study highlights the future potential of AI-driven precision medicine and its role in improving cancer care outcomes.

DOI: https://doi.org/10.5281/zenodo.18919879

Published by: vikaspatanker

Authors: Mr. Prathmesh M. Sadafale, Mr. Pratik S. Date, Mr. Raj S. Kharate, Prof. Ravindra R. Solanke

Abstract: Abstract – This research presents the design and development of an Embedded Intelligent System for Automatic Speed Regulation in sensitive areas such as school zones, hospitals, residential areas, and accident-prone locations. The main objective of the system is to improve road safety by automatically controlling vehicle speed without relying only on driver awareness. The proposed system uses embedded technology, sensors, and wireless communication to detect designated speed-control zones. When a vehicle enters a sensitive area, the system automatically limits its speed to a predefined safe level. Once the vehicle exits the zone, normal speed control is restored. The system operates in real time and reduces the risk of over-speeding. By minimizing human error and ensuring consistent speed regulation, the system enhances road safety, reduces accidents, and supports smarter transportation infrastructure.

 

 

Published by: vikaspatanker

Authors: Deepak Tomar, Kismat Chhillar

Abstract: Diabetes mellitus has emerged as a major global health concern, necessitating early detection and effective predictive mechanisms to support timely medical intervention. Machine learning techniques have increasingly been employed in healthcare analytics to improve diagnostic accuracy and assist clinicians in decision making. Among these techniques, the Support Vector Machine (SVM) algorithm has demonstrated strong performance in classification problems involving medical datasets. This study explores the application of SVM for predicting the likelihood of diabetes using patient health indicators such as body mass index, blood glucose level, age, and family medical history. By analyzing patterns within clinical data, the model classifies individuals into diabetic and non-diabetic categories. The predictive capability of SVM allows the identification of individuals who may be at risk of developing diabetes, thereby enabling preventive healthcare measures. Empirical findings from related studies indicate that SVM-based models can achieve high predictive accuracy, making them a reliable approach for diabetes prediction and early risk assessment in medical decision support systems.

DOI: https://doi.org/10.5281/zenodo.18899892

Published by: vikaspatanker

Authors: Rekha Agarwal, Rajesh Kumar Mishra, Divyansh Mishra

Abstract: Recent solar cycles, particularly the ascending and peak phases of Solar Cycle 25 (2020–2025), have been characterized by heightened solar activity, including X-class flares, fast coronal mass ejections (CMEs), and complex interplanetary shocks. These transient events strongly modulate galactic cosmic rays (GCRs) and produce solar energetic particles (SEPs), thereby altering the flux, energy spectrum, and anisotropy of high-energy charged particles in near-Earth space. This paper synthesizes observational and theoretical advances concerning the effect of recent solar events on high-energy cosmic ray particles (>100 MeV to multi-GeV), with emphasis on Forbush decreases, shock acceleration, magnetic cloud interactions, and ground level enhancements (GLEs). We discuss observations from neutron monitor networks and space-based detectors such as Parker Solar Probe, Solar Orbiter, ACE, and GOES, highlighting case studies from 2021–2024. Quantitative comparisons reveal cosmic ray depressions of 3–20% during major CME passages and episodic enhancements up to GeV energies during extreme SEP events. The broader implications for space weather, atmospheric ionization, and radiation risk are examined.

 

 

Published by: vikaspatanker

Authors: Jose Gnana Babu, Lata Khani Bisht, Visaga Perumal, Vineeth Chandy

Abstract: In recent years, artificial intelligence (AI) has emerged as a strategic catalyst in the field of drug discovery, revolutionizing one of the most complex and resource-intensive areas of the pharmaceutical industry. AI introduces innovative methodologies that enhance efficiency and precision across multiple stages of drug discovery and development, including—though not limited to—virtual screening, target identification, lead optimization, and clinical trials. This review provides an in-depth examination of current AI-driven tools, programs, and platforms that are reshaping modern drug discovery. Beyond presenting the present state of AI applications in this domain, it also explores future directions, existing challenges, and emerging opportunities. The traditional drug discovery process is often constrained by its high cost, long timelines, and substantial attrition rates. However, the integration of AI and machine learning (ML) has introduced transformative solutions, making drug development more rapid, cost-effective, and data-driven. Leveraging vast biological and chemical datasets, AI and ML employ advanced computational techniques—such as neural networks, natural language processing (NLP), and reinforcement learning—to enhance prediction accuracy and streamline decision-making throughout the drug discovery pipeline. These technologies facilitate the identification of novel therapeutic targets, accurate efficacy and safety predictions, and the optimization of clinical trial design, thereby significantly shortening development cycles and reducing overall expenditures. Real-world case studies further illustrate AI’s contribution to groundbreaking therapies in fields such as oncology, neurodegenerative disorders, and rare genetic diseases. Despite these remarkable advancements, notable challenges remain. Concerns surrounding data quality, model transparency, algorithmic bias, and regulatory compliance continue to pose barriers to widespread adoption. Moreover, ethical issues related to data privacy, accountability, and the interpretability of AI-driven decisions demand critical attention. Looking ahead, emerging paradigms such as multi-omics data integration, quantum computing, and precision medicine are expected to redefine the landscape of AI-assisted drug discovery. Achieving this vision will require interdisciplinary collaboration, technological innovation, and the establishment of robust ethical and regulatory frameworks. Collectively, these efforts will pave the way for a new era of patient-centric, precision-driven pharmaceutical development, fully harnessing the transformative potential of AI and ML in drug discovery.

DOI: https://doi.org/10.5281/zenodo.18899332

Published by: vikaspatanker

Sriram Ghanta
Affiliation	Senior Java Full Stack Developer
Email-Id:	shriram.gh@gmail.com
SUMMARY:- Results-oriented Senior Java Full Stack Developer with over 14 years of experience in designing, developing, and integrating enterprise-level applications across Retail, Finance, Healthcare, and Entertainment domains. Expertise in Java, Spring Boot, Microservices, RESTful APIs, and modern front-end technologies such as Angular. Proven ability to build scalable, cloud-native applications using AWS and containerized environments. Strong experience in system design, performance optimization, and implementing resilient architectures. Adept at working in Agile environments and delivering high-quality software solutions

Published by: vikaspatanker