Procedia Computer Science最新文献

{"title":"Information security requirements in monitoring e-government systems","authors":"Azam Gafurov ,&nbsp;Fayzullajon Botirov","doi":"10.1016/j.procs.2025.01.001","DOIUrl":"10.1016/j.procs.2025.01.001","url":null,"abstract":"<div><div>This article examines the requirements for information security in monitoring e-government systems. It discusses the information security framework for eGoverment system, a model for organizing multi-level network security based on layered protection. This research study also covers the establishment of information security zones with eGovernment. and the importance of ensuring the local information securityand the process of connecting information security to a corporate network. It addresses the security measures for remote access and the use of automated control systems, such as VPN connections.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 424-429"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data-Driven Insights into Social Media Behavior Using Predictive Modeling","authors":"V. Selvakumar ,&nbsp;Nadipi Keerthana Reddy ,&nbsp;R. Sree Vardhini Tulasi ,&nbsp;Kunchala Rohit Kumar","doi":"10.1016/j.procs.2025.01.007","DOIUrl":"10.1016/j.procs.2025.01.007","url":null,"abstract":"<div><div>This study proposes a statistical machine learning approach to predict social media usage across various demographic categories in India. The dataset comprises twenty-six features, including demographic attributes (age, gender, education, location), social media engagement metrics (number of followers, posts, time spent on platforms), and device-related information. It reflects real-world social media behavior on platforms such as WhatsApp, Facebook, and Instagram, capturing distinct patterns of weekday and weekend usage. Key variables such as time spent on each platform, the number of Instagram posts and followers, and overall social media usage were analyzed in detail. It is identified that significant predictors of user status categories through feature engineering, including Sabbatical, Self-Employed, Student, and Working Professional. Multiple regression models—Linear Regression, K-Nearest Neighbors, Decision Tree Regression, Random Forest Regression, Gradient Boosting, Naïve Bayes, and Support Vector Regression—were employed to assess their performance in predicting user status. Comparative analysis revealed that the Gradient Boosting algorithm outperformed other models with the highest accuracy. The machine learning workflow encompassed data pre-processing, feature engineering, model training, and evaluation, all implemented using Python. This study significantly advances the field by elucidating the key predictors of social media engagement and providing a thorough evaluation of the importance of features alongside a comparative analysis of predictive models.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 480-489"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EM-ACO-ARM: An Enhanced Multiple Ant Colony Optimization Algorithm for Adaptive Resource Management in Cloud Environment","authors":"Prathamesh Lahande ,&nbsp;Parag Kaveri ,&nbsp;Harvinder Singh ,&nbsp;Sukhjit Singh Sehra ,&nbsp;Jatinderkumar R. Saini","doi":"10.1016/j.procs.2025.01.040","DOIUrl":"10.1016/j.procs.2025.01.040","url":null,"abstract":"<div><div>Ant Colony Optimization (ACO) is an intelligent algorithm ensuring optimal resource management in cloud environments. This paper proposes an enhanced version of the ACO algorithm called Enhanced Multiple Ant Colony Optimization for Adaptive Resource Management (EM-ACO-ARM). Our approach uses multiple ant colonies undergoing several iterations of optimizations to find the optimal Virtual Machine (VM) and adapt to the convergence uncertain-ties, unlike a single ant colony in the existing ACO, which can hinder Quality of Service (QoS)-based performance parameters. We conducted experiments in a cloud-simulated environment to evaluate EM-ACO-ARM in two phases. In the first phase, we computed real-time Montage tasks using the existing ACO algorithm on VMs across ten scenarios. To ensure an unbiased comparison, the same cloud configuration was maintained in the second phase, and the same tasks were computed using the proposed EM-ACO-ARM algorithm in all ten scenarios. The experimental results demonstrate that EM-ACO-ARM improves Execution Cost and Execution Time, leading to a 14.73% increase in Resource Utilization. This ultimately improves the management of cloud resources. Additionally, a stability evaluation was conducted using regression models, and it outputted EM-ACO-ARM to provide more stability than the existing ACO algorithm. The cloud can provide better QoS with the proposed EM-ACO-ARM algorithm while abiding by Service Level Agreements.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 796-805"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376862","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Metaheuristic Feature Selection for Diabetes Prediction with P-G-S Approach","authors":"Karuppasamy M ,&nbsp;Jansi Rani M ,&nbsp;Poorani K","doi":"10.1016/j.procs.2024.12.018","DOIUrl":"10.1016/j.procs.2024.12.018","url":null,"abstract":"<div><div>Diabetes mellitus is increasing in large numbers globally. It also leads to various complications ultimately leads to death. Increase in mortality due to diabetic complication is increasing. Early diagnosis of diabetes and its complication leads to decrease in mortality rate. Trending technologies like machine learning and deep learning are much helpful in diagnosing diseases. Advanced prediction models are built to find diseases earlier with efficient algorithms.Metaheuristic approach with particle swarm optimization is utilized. Stacking is one such method which helps in combining various weak algorithms and performs prediction with a meta-classifier. This proposed ParticleSwarm-Gridsearch-Stacking (PGS) approach encompasses the several methods for diabetes prediction and provides the best practices with stacking techniques. The results show that stacking with heterogenous machine learning algorithms with hyperparameter tuning and logistic regression as meta-classifier yields 96.7 % accuracy, 94 % precision and 92.7% recall. This work highlights the importance of feature selection, grid search and stacking which adds overall improvement in machine learning model.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 165-171"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient Miner Selection in Blockchain Based on Predicted Transaction Time","authors":"Manjula K Pawar ,&nbsp;Prakashgoud Patil ,&nbsp;Narayan D.G. ,&nbsp;Vasundhara Pandey ,&nbsp;Shorya Jain ,&nbsp;Priyanshu Kumar","doi":"10.1016/j.procs.2024.12.022","DOIUrl":"10.1016/j.procs.2024.12.022","url":null,"abstract":"<div><div>Blockchain’s decentralized, transparent, and immutable nature has revolutionized digital transactions by removing the need for central authorities. Ethereum stands out among blockchain platforms for facilitating secure peer-to-peer transactions via smart contracts. Despite its transformative potential, blockchain faces challenges, particularly with the PoW consensus algorithm, which demands high energy consumption and raises centralization concerns. This affects the scalability of Blockchain by reducing the throughput. This paper explores machine learning (ML) integration to address these challenges, specifically focusing on optimizing miner selection in the Ethereum blockchain based on predicted transaction times. The study compares the performance of various machine learning models, including ElasticNet, Lasso Regression, Multilayer Perceptron (MLP) Regression in optimizing miner selection for reduced transaction times on the Ethereum blockchain. This study advances the ongoing research on integrating machine learning with blockchain to address the shortcomings of traditional Proof of Work (PoW) systems. It emphasizes the potential of machine learning to propel future innovations in blockchain technology.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 202-211"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Privacy Protection Technique of Electronic Health Records using Decentralized Federated Learning with Consortium Blockchain","authors":"S.P. Panimalar ,&nbsp;S. Gunasundari","doi":"10.1016/j.procs.2024.12.023","DOIUrl":"10.1016/j.procs.2024.12.023","url":null,"abstract":"<div><div>This research study proposes a novel FedBlock model by integrating the federated learning and blockchain technologies to enhance data security and predictive accuracy in healthcare domain. Across diverse medical datasets with varying data distributions, the proposed model outperforms existing methods in preserving data privacy and maintaining high predictive accuracy. By integrating federated learning, FedBlock achieves superior AUROC scores compared to blockchain-based and ShareChain models, showcasing its effectiveness in data privacy preservation. Additionally, the proposed model demonstrates a comparatively higher F1-score and accuracy rate with AUROC score reaching up to 0.98 while processing the medical dataset. Through collaborative training and decentralized data management, FedBlock ensures progressive accuracy improvements while safeguarding data integrity, paving way for enhanced healthcare data security and predictive analytics.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 212-221"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive Deep Reinforcement Learning-based Resource Management for Complex Decision Making in Industry Internet of Things Applications","authors":"Niharika Karne ,&nbsp;Chandra Shekar Ramagundam ,&nbsp;Ranga Rao Patnala ,&nbsp;Shashishekhar Ramagundam ,&nbsp;Sai Nitisha Tadiboina","doi":"10.1016/j.procs.2024.12.025","DOIUrl":"10.1016/j.procs.2024.12.025","url":null,"abstract":"<div><div>An independent industrial system may transform into a connected network through the assistance of Industrial Internet of Things (IIoT). The deployed sensors in the IIoT maintain surveillance of the industrial machinery and equipment. As a result, safety and reliability emerge as the primary concerns in IIoT. This presents a variety of well-known and increasing issues related to the industrial system. The IIoT devices are exposed to a wide range of malware, threats, and assaults. To prevent the IIoT devices from malware effects, effective protection plans must be implemented. But adequate security mechanisms are not be incorporated in IIoT devices with limited resources. It is essential to ensure the accuracy and dependability of information gathered by IIoT devices. Decisions taken with incomplete or inaccurate data might be devastating. To overcome these difficulties deep learning with reinforcement learning for complex decision-making in industry applications is developed in this research work. In this developed model, an Adaptive Deep Reinforcement learning (ADRL)-based resource management is performed to reduce the operation cost associated with IIoT deployments. Energy efficiency is essential in IIoT ecosystem, particularly for the devices that run on batteries. Through dynamic resource allocation based on workload needs and energy limits, ADRL-based resource management optimizes the usage of energy. The reliability of the designed model is enhanced by fine-tuning the parameters from DRL using the Ship Rescue Optimization (SRO) algorithm. Thus, ADRL-based resource management systems make real-time decisions based on current environmental conditions and system requirements. This helps the IIoT systems to react quickly to change demands and optimize resource allocation. Finally, the experimental analysis is performed to find the success rate of the developed resource management system via various metrics. Throughout the validation, the statistical analysis of the developed model shows 15.3%, 8.3%, 5.4% and 11.1% enhanced than DO-ADRL, CO-ADRL, OOA-ADR and SGO-ADRL in terms of mean analysis. This performance shows the developed model offers superior performance when compared with the existing models.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 231-240"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stock market time series forecasting using comparative machine learning algorithms","authors":"Samraj Gupta ,&nbsp;Sanchal Nachappa ,&nbsp;Nirmala Paramanandham","doi":"10.1016/j.procs.2025.01.050","DOIUrl":"10.1016/j.procs.2025.01.050","url":null,"abstract":"<div><div>Stock market trends prediction and their assessment have always been leading topics due to the market’s extreme volatility and chaotic nature. Determining the stock market possesses an abstract representation of possession over enterprises and organizations, more commonly known as “stock.” This kind of assessment is generally accepted as the foundation of market conduct is not provided with as many shares as needed admitted to financial failure. Foreseeing the performance of a single business on stock markets is a precarious one, as values of stocks are subjected to constant change. Nonetheless, viewing the stock market’s behavior in retrospect and distributing investor expectations regarding future stock market values is a beneficial approach. Different models have been proposed in the last few years, and there is a saturation of work to compare efficiency, accuracy or robustness for identifying which model works best on different applications. In this paper, a detailed comparative study of different machine learning algorithms for stock market time series prediction has been shown. It provides baselines with widely used algorithms such as Linear Regression and Support Vector Machines, to state-of-the-art methods including Long Short-Term Memory networks, Convolutional Neural Networks and Transformer-based architectures.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 893-904"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A pyDAOS Approach for Enabling Efficient Data Handling in DAOS","authors":"Chandrika Prasad ,&nbsp;Geetha J Dr ,&nbsp;Vinay M ,&nbsp;Manoj V L ,&nbsp;Bharath Vamsi ,&nbsp;D. Malavika Dileepa","doi":"10.1016/j.procs.2025.01.053","DOIUrl":"10.1016/j.procs.2025.01.053","url":null,"abstract":"<div><div>High-Performance Computing (HPC) systems require efficient storage solutions to handle vast amounts of data. Distributed Asynchronous Object Storage (DAOS) is designed for such environments, offering scalable and high performance storage. Our study initially explores the capabilities and performance of DAOS using pyDAOS when integrated with a key-value store (KV store). The existing methodologies are analyzed to identify inefficiencies, particularly in handling large I/O operations. This paper proposes an enhanced approach that incorporates chunking for large I/O operations, automated resource selection, and simplified system administration via server discovery and by fetching configuration details. This results show significant improvements in efficiency, scalability, and usability of DAOS using pyDAOS, making it a robust solution for modern data intensive applications.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 922-933"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Smart Platform Connectivity Interface: Train detection and Distance Prediction Using IoT And Machine Learning","authors":"Muhammed Tufayl Dalvi ,&nbsp;Aditi Narkar ,&nbsp;Sujata Kadu Dr.","doi":"10.1016/j.procs.2025.01.029","DOIUrl":"10.1016/j.procs.2025.01.029","url":null,"abstract":"<div><div>This research study proposes an intelligent IoT system interface to handle automated and motorized horizontal passenger transfers from one railway platform to another while increasing the overall efficiency of the railway systems. The proposed system employs NodeMCU as the main controller with a motorized rolling stage (interface) for motion control; various LEDs are used for displaying status and alarm signals and one screen display for displaying the incoming train information. Through computerized opening and closing of the platform edges, the passengers are able to have a smooth and secure passage between the platforms. The railway train detection module is implemented in Python with the help of OpenCV and YOLOv5 object detection model. The proposed design also accommodates wheelchair users by providing an accessible transfer platform and enhancing mobility and safety for all passengers. To enable the recognition of trains under diferent environmental conditions, a custom dataset of train images with annotations was developed specifically for training. When the approaching train is detected, the platform interface movement mechanism is activated, and the information on the display together with the signal lights is changed.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"252 ","pages":"Pages 692-701"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}