Journal of Science & Technology
Not a member yet
224 research outputs found
Sort by
Optimizing Diabetes Prediction with Machine Learning: Model Comparisons and Insights
No full textThis study aims to predict diabetes using various machine learning models and compare their performances. The dataset utilized contains health indicators and lifestyle factors from a diverse population. The models evaluated include Random Forest, Logistic Regression, Support Vector Machine (SVM), and Gradient Boosting. Results indicate that Gradient Boosting outperforms other models in terms of accuracy, precision, and recall, making it a robust choice for diabetes prediction. The analysis provides insights into the most significant features contributing to diabetes prediction and highlights the potential of machine learning in medical diagnosis
Risk Management Practices in Nigeria Construction Sector and Impact on Project Performance
No full textThe construction industry in every nation is essential to the growth of the economy and its development. The industry is characterized by complex projects, unpredictable variables, and is dependent on many different stakeholders, which subject projects in this sector to a variety of risks that can have a significant impact on project delivery. Risks in construction projects in Nigeria can lead to cost overruns, schedules not being met, and low quality of the project. The concept of risk management involves the management of identified risks which reduces the likelihood of their occurrence and the consequences of adverse risks. A project may be delivered or deemed successful, but not meet the project performance objectives. The Nigerian construction industry has undergone incredible growth and progress; however, there have been numerous challenges that have prevented project performance, such as a lack of expertise, knowledge gaps, and a drought of experienced personnel. The study examines the various risk management practices and their impact on construction project performance. Project performance implies that the project objectives are met in terms of cost, schedule, and quality. A survey research method was adopted using a questionnaire as the primary instrument for data collection on a five-point Likert scale. The questionnaire was designed to collect data relating to risk management’s impact on project performance. The questionnaires were also designed to collect data on risk identification as a risk management practice and its impact on project performance. A total of 85 questionnaires were administered and analyzed. The results from the study indicate a strong relationship between risk management practices and project performance
Time Complexity Analysis of Graph Algorithms in Big Data: Evaluating the Performance of PageRank and Shortest Path Algorithms for Large-Scale Networks
No full textThis paper delves into the time complexity analysis of two prominent graph algorithms, PageRank and shortest path algorithms, with a focus on their performance in large-scale networks commonly encountered in big data systems. The need to process extensive network data efficiently has led to an increased emphasis on understanding the computational complexity of algorithms applied to graph-based structures, especially in scenarios where the size of the data becomes a critical factor in performance evaluation. As the volume of network data grows exponentially, algorithms designed for tasks such as ranking web pages or finding optimal paths between nodes must be assessed not only for their accuracy but also for their scalability and efficiency in terms of computational resources.
PageRank, a foundational algorithm for ranking web pages, operates on the principle of recursively measuring the importance of nodes within a network based on their connectivity. The algorithm’s time complexity is dependent on both the number of nodes and edges in the graph, as well as the convergence criterion used. This paper evaluates the iterative nature of PageRank, examining its time complexity with respect to various parameters such as network size, convergence tolerance, and damping factor. Furthermore, the paper explores how different optimization techniques, including parallel and distributed computing, affect the performance of PageRank when applied to large-scale networks. Special attention is given to the algorithm\u27s behavior in both static and dynamic network environments, where the underlying graph structure may evolve over time. The paper aims to provide a comprehensive understanding of how PageRank\u27s computational complexity grows as the scale of the network increases, and how this growth can be mitigated through algorithmic and infrastructural optimizations.
Similarly, shortest path algorithms, such as Dijkstra\u27s algorithm and the Bellman-Ford algorithm, are analyzed with respect to their time complexity in the context of large-scale graphs. These algorithms are crucial for applications that require determining the optimal path between nodes, a common requirement in network routing, transportation logistics, and social network analysis. The performance of these algorithms is evaluated based on different graph structures, such as sparse versus dense graphs, and under various constraints, such as edge weights and graph directionality. The paper discusses how the choice of algorithm impacts the overall time complexity, especially in cases where real-time computation is critical. It also examines the role of heuristics, like A*, in reducing the computational overhead for certain types of networks.
To provide a holistic view, this paper integrates empirical analysis with theoretical evaluations, comparing the worst-case, best-case, and average-case time complexities of PageRank and shortest path algorithms. Through the use of experimental simulations, the paper showcases how these algorithms perform in practice when applied to datasets containing millions or billions of nodes and edges. The results of these simulations highlight the practical limitations of these algorithms when used in large-scale networks, and suggest possible improvements, including algorithmic enhancements and hardware-accelerated implementations.
In addition to providing a detailed complexity analysis, the paper also addresses the trade-offs involved in the design and deployment of these algorithms in distributed computing environments. With the rise of big data platforms such as Hadoop and Apache Spark, the scalability of graph algorithms has become an increasingly important area of research. The paper examines how these distributed platforms handle the execution of PageRank and shortest path algorithms, focusing on the communication overhead, load balancing, and fault tolerance issues that arise when processing large-scale networks. The interplay between algorithmic complexity and distributed system architecture is discussed, highlighting the need for fine-tuning both the algorithm and the infrastructure to achieve optimal performance in big data contexts.
Furthermore, the paper addresses the practical implications of these time complexity analyses in real-world applications. For instance, the application of PageRank in search engine optimization and social media influence measurement, and the use of shortest path algorithms in logistics, transportation, and telecommunication networks, underscore the importance of understanding the computational limitations and scalability challenges of these algorithms. The findings presented in this paper will be relevant not only to researchers in the field of graph theory and big data but also to practitioners who must choose appropriate algorithms for handling large-scale network data.
Overall, this paper contributes to the field by providing a comprehensive analysis of the time complexity of PageRank and shortest path algorithms in the context of big data. By combining theoretical insights with empirical evaluations, the paper offers a robust framework for understanding the scalability challenges of these algorithms when applied to large-scale networks. Additionally, the paper identifies key areas for future research, including the development of more efficient algorithms for large-scale graph processing, the optimization of existing algorithms for distributed environments, and the exploration of new graph-theoretic approaches for handling the increasing complexity of big data networks
Predictive Analytics in Dental Health: Leveraging Data for Early Detection and Prevention
No full textThis paper delves into the application of predictive analytics in the realm of dental health, aiming to discern patterns, trends, and risk factors associated with oral diseases. Leveraging extensive datasets and advanced analytical techniques, predictive analytics emerges as a potent tool for early detection and preventive interventions in dental care. Through the integration of diverse data sources, including patient records, imaging studies, and demographic information, predictive models can forecast oral health outcomes with remarkable accuracy. By identifying individuals at heightened risk of developing oral conditions such as caries, periodontal disease, and oral cancer, healthcare providers can tailor personalized interventions and allocate resources efficiently. Furthermore, predictive analytics facilitates the optimization of treatment plans, enhancing patient outcomes and minimizing healthcare costs. This paper underscores the transformative potential of predictive analytics in revolutionizing dental care delivery, fostering a proactive approach towards oral health management
Real-Time IoT Data Analytics for Smart Manufacturing: Leveraging Machine Learning for Predictive Analytics and Process Optimization in Industrial Systems
No full textThe rapid evolution of the Internet of Things (IoT) has revolutionized smart manufacturing, enabling industries to harness real-time data for predictive analytics and process optimization. In this research, we delve into how IoT and machine learning (ML) technologies can be synergized to provide actionable insights, allowing for the optimization of manufacturing systems and the enhancement of predictive maintenance in industrial environments. The integration of IoT devices into manufacturing processes generates an unprecedented volume of data, which, when processed in real-time, has the potential to drive significant improvements in efficiency, cost-effectiveness, and decision-making. However, the challenge lies in the effective handling, analysis, and interpretation of this vast data, which is where machine learning algorithms play a pivotal role.
This paper explores various machine learning models, including supervised, unsupervised, and reinforcement learning techniques, and how they are employed in real-time IoT data analytics for smart manufacturing. The discussion extends to the architectures and frameworks needed to process and analyze IoT-generated data streams efficiently. Real-time analytics, powered by machine learning, enables the continuous monitoring of key performance indicators (KPIs) and predictive analytics in manufacturing environments, thus ensuring that manufacturers can react to potential issues before they escalate into costly downtimes or defects in production. Predictive maintenance, a crucial component of smart manufacturing, is significantly enhanced by the combination of IoT data and machine learning models, as they can predict equipment failures and maintenance needs with high accuracy, leading to reduced downtime, optimized asset utilization, and cost savings.
Moreover, the paper presents an in-depth examination of process optimization through machine learning in manufacturing. Traditional manufacturing processes often rely on retrospective data analysis, which, while valuable, limits the ability to react dynamically to changes in the system. IoT-enabled systems, combined with machine learning algorithms, allow for real-time feedback loops where manufacturing processes can be adjusted on the fly to improve efficiency and product quality. The real-time capabilities of these systems are critical for industries striving to remain competitive in an increasingly digital and connected industrial landscape. This shift from reactive to proactive operations is made possible through advanced machine learning models that analyze sensor data in real time, enabling the detection of anomalies, the identification of inefficiencies, and the optimization of processes.
To fully realize the potential of IoT in smart manufacturing, it is essential to address the challenges associated with real-time data analytics. These challenges include managing the massive scale of IoT data, ensuring low-latency processing, and maintaining the security and privacy of sensitive industrial information. This research outlines the latest advancements in edge computing and cloud-based analytics that mitigate these challenges, enabling manufacturers to process data closer to the source while still leveraging the computational power of the cloud for complex machine learning tasks. Edge computing, in particular, has emerged as a critical technology for reducing the latency of IoT data processing, allowing for real-time decision-making in manufacturing environments where even minor delays can lead to significant disruptions.
Additionally, the paper explores case studies and real-world implementations of IoT-driven smart manufacturing systems, providing a comprehensive analysis of the successes and challenges encountered. These case studies highlight the tangible benefits of real-time IoT data analytics, such as increased operational efficiency, reduced downtime, enhanced product quality, and more sustainable manufacturing practices. Furthermore, they underscore the role of machine learning in transforming raw IoT data into meaningful insights that drive continuous improvement in manufacturing processes. The ability to predict potential issues and optimize production in real-time represents a paradigm shift from traditional manufacturing practices, positioning IoT and machine learning as key enablers of the next industrial revolution, commonly referred to as Industry 4.0.
This research also touches upon the future directions of IoT and machine learning in manufacturing, including the integration of advanced artificial intelligence (AI) techniques, such as deep learning and neural networks, which hold promise for even more sophisticated predictive analytics and process optimization. The potential for AI-driven automation in smart manufacturing is vast, and as these technologies mature, their adoption will likely become more widespread, leading to further enhancements in efficiency, scalability, and adaptability. Furthermore, the research discusses the importance of developing standardized frameworks and protocols for IoT data in manufacturing to facilitate interoperability and ensure that different systems can seamlessly communicate and collaborate.
This paper provides a detailed exploration of the integration of IoT and machine learning technologies in smart manufacturing, focusing on real-time data analytics for predictive maintenance and process optimization. By leveraging IoT data and machine learning models, manufacturers can achieve significant improvements in operational efficiency, reduce downtime, and enhance product quality, ultimately leading to a more competitive and resilient industrial environment. The research identifies key challenges in the field, such as data management and latency, and proposes technological solutions, including edge computing and cloud-based analytics. As IoT and machine learning technologies continue to evolve, their role in shaping the future of smart manufacturing will become increasingly critical, driving innovation and efficiency in industrial systems
Sustainable Power Solutions: Renewable Energy & Storage Advancements
No full textThe pursuit of sustainable power solutions has become imperative in the face of global environmental challenges and the growing energy demand. This abstract explores the advancements in renewable energy and energy storage technologies, aiming to create a comprehensive understanding of their role in achieving a sustainable future. Renewable energy sources, such as solar, wind, hydro, geothermal, and biomass, have gained significant traction due to their abundance, low environmental impact, and decreasing costs. Breakthroughs in photovoltaic technology have increased solar energy efficiency, making it a more viable option for widespread adoption. Similarly, innovations in wind turbine design and materials have bolstered the efficiency and reliability of wind power generation. However, the intermittent nature of renewable energy sources poses a challenge to their integration into the grid, highlighting the critical importance of energy storage systems. Advancements in energy storage technologies, including lithium-ion batteries, flow batteries, hydrogen storage, and thermal storage, have been instrumental in mitigating the variability of renewable sources. These developments have enhanced the stability and flexibility of energy systems, enabling a smoother integration of renewables into existing grids. Predictive analytics and control systems are pivotal in maximizing renewable resources\u27 efficiency and storage units\u27 efficiency while ensuring grid stability and reliability
Governance and Quality in Open-Source Projects: Navigating the Path to Excellence
No full textThis review article delves into the intricate interplay between governance structures and software quality within open-source projects. Governance practices are the backbone of collaborative dynamics, decision-making processes, and the overall success of these projects. The exploration encompasses the foundational principles of effective governance, its profound impact on software quality, challenges encountered, and emerging trends poised to reshape the future of open-source project governance. Anyone can also add to the collective open-source knowledge ecosystem or knowledge commons by contributing ideas, designs, observations, experimental data, deployment logs, etc.
In the introduction, the foundational importance of governance is emphasized, setting the stage for an in-depth examination of the multifaceted relationship between governance structures and software quality. The principles of effective governance, including transparency, community inclusivity, and decentralization, are presented as the guiding ethos that shapes collaborative environments. Open-source software development is the next stage in the evolution of product development, particularly software products.
The impact of governance on software quality is explored, illustrating how governance decisions influence critical aspects such as code review processes, quality assurance practices, and feature prioritization. Well-governed projects are shown to prioritize quality, user satisfaction, and the early detection of issues. Software quality is a critical factor in ensuring the success of software projects.
Challenges in open-source project governance, including maintaining consistency across contributors and balancing centralization and decentralization, are addressed. Considerations for overcoming these challenges are provided, highlighting the importance of clear guidelines and hybrid governance models.
The article also discusses emerging trends in open-source project governance, such as the exploration of blockchain for transparent governance, the increasing adoption of Decentralized Autonomous Organizations (DAOs), and the integration of artificial intelligence into governance tools. These trends are poised to redefine decision-making processes, enhance transparency, and introduce data-driven governance practices.
In conclusion, the review underscores the pivotal role of governance in shaping collaborative environments and ensuring the delivery of high-quality software. Navigating the complex relationship between governance structures and software quality is crucial for the continued success and innovation of open-source projects. As the landscape evolves, effective governance remains a cornerstone, guiding projects towards excellence and meeting the evolving needs of the software development community
Navigating the Software Development Landscape: A Comprehensive Review of Performance Testing Tools, Monitoring Solutions, and Agile/DevOps Practices
No full textThis review article provides an in-depth analysis of key components in the modern software development ecosystem, focusing on performance testing tools, performance monitoring solutions, and the integration of Agile and DevOps practices into quality assurance processes. software engineering is a discipline that undergone many improvements that aims to keep up with the new advancements in technologies and the modern business requirements through developing effective approaches to reach the final software product, agile software development is one of these successful approaches. The exploration centers around Apache JMeter, LoadRunner, and Gatling as performance testing tools, unraveling their features and impact on identifying and addressing performance bottlenecks in software applications. Concurrently, it evaluates prominent performance monitoring solutions such as New Relic, AppDynamics, and Dynatrace, offering insights into their real-time monitoring capabilities and contributions to application behavior and resource utilization.
The article also delves into the paradigm shift in quality assurance practices driven by Agile and DevOps methodologies. It explores how continuous testing, shift-left testing, collaboration, and test automation have become integral components of the development pipeline, facilitating faster feedback loops and enhanced software quality. Software quality is a critical factor in ensuring the success of software projects. Numerous software quality models have been proposed and developed to assess and improve the quality of software products. In the face of global competition, businesses across various industries have increasingly turned to lean methodologies to enhance their production processes and remain competitive.
By synthesizing these elements, the review offers a holistic perspective on how these tools and practices converge to shape the ever-evolving landscape of software development. The reader is guided through a comprehensive understanding of the tools\u27 functionalities, their applications in diverse development scenarios, and their synergy with Agile and DevOps practices. The article serves as a valuable resource for developers, QA professionals, and stakeholders navigating the complexities of the modern software development lifecycle, providing insights that contribute to the creation of high-quality, scalable, and resilient software solutions
AI Integration with ServiceNow and CMDB: Revolutionizing Industries and Society
No full textThe integration of Artificial Intelligence (AI) with Configuration Management Database (CMDB) and ServiceNow, a leading IT service management platform, represents a pivotal advancement in organizational efficiency, data-driven decision-making, and transformative innovation. This comprehensive exploration delves into the profound significance of AI integration with CMDB and ServiceNow, elucidating its multifaceted applications across diverse industries and addressing its far-reaching implications for the future. Through a detailed analysis spanning 20,000 words, this article endeavors to provide an exhaustive understanding of this transformative synergy and its potential to reshape the technological landscape, drive organizational excellence, and foster societal progress
A Survey on Malware Detection and Analysis
No full textMalware, or malicious software, poses a significant threat to the security and functionality of computer systems globally. This survey provides a comprehensive analysis of current malware detection and analysis methods, focusing on data mining methodologies. The study categorizes malware detection techniques into signature-based and behaviour-based approaches, highlighting their respective strengths and weaknesses. It explores heuristic techniques enhanced by artificial intelligence, including neural networks and genetic algorithms, to improve detection accuracy. The literature review examines host-based and network-based intrusion detection systems, hybrid systems, and virtual machine introspection. The paper also discusses static and dynamic analysis methods, emphasizing the importance of analysing malware in controlled environments. Through detailed examination, this survey aims to present a thorough understanding of contemporary malware detection strategies and their applications, offering insights for future advancements in the field