Volume 13 | Issue 2 |

 Abstract

The exponential growth of data and the development of artificial intelligence (AI) have generated huge data lakes, which provide organizations big, unstructured data sets. Although data lakes have the promise to provide useful insights to organizations, organizations are not able to use these assets effectively for strategic decision-making. The objective of this study is to explore the application of generative AI on big data lakes to generate meaningful insights and create embeddings to enhance business intelligence (BI) operations. A significant research gap exists in the practical implementation of generative AI models to unstructured data in data lakes, especially with regard to efficiency in processing and scalability. Most existing methodologies focus heavily on traditional data processing techniques and do not have the attention of overcoming the complexity and size of data involved in data lakes. Moreover, although AI has been implemented with smaller and structured data sets with significant success, applying AI on big, heterogeneous data sets requires new strategies to facilitate proper interpretation of data and actionable results. This research provides a conceptual model incorporating generative AI models with data lakes to recover contextual embeddings and generate insights in support of better-informed business decision-making. Identifying the importance of feature extraction through automated methods, anomalies, and advanced predictive models, this research is an attempt to address the prevalent vacuum of awareness surrounding the cost-efficient large-scale applications of AI to analyze real-time data. Last but not least, this research is about enhancing decision-making capabilities, cutting costs, and promoting organizational response effectiveness in emerging data ecosystems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Generative AI, data lakes, business intelligence, insight generation, embeddings, unstructured data, predictive modeling, scalable AI solutions, decision-making optimization, data processing, anomaly detection, feature extraction.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In today's increasingly interconnected global marketplace, effective data management stands as a critical enabler for competitive supply chains. This study explores the transformative role of Master Data Management (MDM) in aligning and optimizing data quality and governance across international operations. By integrating Generative AI (Gen AI) into MDM frameworks, organizations can revolutionize their approach to data curation, error detection, and consistency assurance. The innovative application of Gen AI enhances data enrichment processes and automates validation routines, resulting in more robust, accurate, and real-time data flows that underpin strategic decision-making. In addition, the convergence of MDM and Gen AI contributes to improved transparency and traceability within supply chain networks. This synthesis not only streamlines operations but also minimizes risks related to data inconsistencies and regulatory non-compliance. Furthermore, the implementation of Gen AI fosters adaptive learning, enabling continuous improvement in data governance practices. The study underscores the potential benefits of this technology integration, including accelerated operational efficiency, better demand forecasting, and enhanced supplier collaboration. As businesses expand their global footprints, embracing advanced MDM solutions with Gen AI integration becomes imperative for maintaining a competitive edge and ensuring resilient supply chain performance. Through empirical analysis and case studies, this research offers insights into best practices and strategic considerations that drive successful digital transformation in supply chain management.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Master Data Management, Global Supply Chains, Data Quality, Data Governance, Generative AI, Digital Transformation, Supply Chain Optimization

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In today's data-intensive landscape, efficient management of vast and heterogeneous datasets has become paramount. Hybrid cloud storage architectures offer scalable, flexible, and cost-effective solutions by combining on-premises resources with public cloud services. This paper explores the integration of machine learning techniques to drive advanced data management strategies within hybrid cloud environments. By leveraging machine learning algorithms, organizations can automate the classification, indexing, and retrieval of data, thereby improving system performance and reducing latency. The approach focuses on predictive analytics to forecast data access patterns and resource requirements, ensuring optimal allocation and minimizing bottlenecks. Additionally, machine learning models can enhance security protocols by detecting anomalies and potential threats in real time. This fusion of intelligent automation with hybrid cloud infrastructure not only streamlines data operations but also paves the way for proactive system maintenance and cost optimization. Experimental results indicate significant improvements in data throughput, energy efficiency, and overall user satisfaction. The study highlights the potential challenges, including model training complexities, data privacy concerns, and the need for robust integration frameworks that can adapt to rapidly evolving technologies. Future research directions include refining algorithm accuracy, expanding the range of predictive insights, and developing hybrid solutions that balance performance with regulatory compliance. Overall, this work demonstrates that machine learning-driven data management represents a transformative strategy for modern hybrid cloud storage systems, offering sustainable benefits for enterprise data governance.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning; Data Management; Hybrid Cloud Storage; Predictive Analytics; Intelligent Automation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Disaster Recovery (DR) and High Availability (HA) are core features of modern database management systems, particularly within the SQL Server environment, because they are the backbone of business continuity in instances of system crashes, data corruption, or any other disaster. While numerous options exist, for example, geo-replication, Failover Clustering, and Always On Availability Groups, significant challenges exist to optimize these systems in a myriad of environments like hybrid cloud, big data, healthcare, and small and medium-sized enterprises (SMEs). The gap in existing literature is on integration of these technologies with new-age cloud-based technologies, the utilization of automated recovery processes, and the determination of cost-effective approaches for small-sized organizations that have limited IT infrastructures. Current research highlights the increasing demand for multi-cloud and hybrid disaster recovery solutions that reduce downtime and data loss. Although such solutions are scalable and cost-effective, they also pose challenges of performance, latency, and coordination of distributed systems. Additionally, most studies concentrate on large enterprises and do not consider the unique requirements of SMEs that need more cost-effective yet reliable disaster recovery solutions. Additionally, there is little research on fully automating disaster recovery processes, which can significantly impact reducing recovery time and human error. Moreover, the healthcare industry needs special consideration in regulatory compliance, data protection, and confidentiality in disaster recovery processes. This paper aims at filling these gaps through an analysis of the latest HA and DR technologies, such as their adaptability across industries, the integration of automated recovery solutions, and cost-effective approaches for small, medium, and large enterprises. Important findings will guide improved disaster recovery functions for SQL Server in a way that permits minimal downtime and data coherence across various operating environments.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

High availability, disaster recovery, SQL Server, Always On Availability Groups, failover clustering, hybrid cloud environments, multi-cloud approaches, automated recovery processes, big data analytics, healthcare systems, small and medium-sized businesses, data replication methodologies, business continuity planning, backup strategies, Disaster Recovery as a Service (DRaaS)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The integration of Large Language Models (LLMs) with SQL-oriented data pipelines is an emerging area that seeks to make databases more functional and usable based on the paradigm of natural language processing (NLP) methods. Although the impressive capabilities demonstrated by LLMs in the domain of text-to-SQL translation are well documented, the wider potential of LLMs for the domain of database systems is relatively unexplored. Existing academic contributions have been mostly focused on niche applications, such as query construction; however, concerns related to database schema understanding, query optimization, and scalability in dynamic environments are still open. There is a pressing need for well-tuned models with the capability to handle diverse domain-specific data, as well as the incorporation of LLMs in data preprocessing and real-time querying, which is an open research gap. Furthermore, existing solutions are not robust enough for large-scale, real-time applications and are usually beset with challenges of ensuring data privacy and security when handling sensitive data. This research effort seeks to address these gaps by suggesting an end-to-end system for the incorporation of LLMs in SQL-oriented data pipelines, with a focus on important considerations such as query construction efficiency, query optimization, and dynamism in heterogeneous domains. Through the exploration of pre-trained and well-tuned LLM approaches, this research seeks to close the gap between state-of-the-art NLP methods and real-world database management, thus improving the effectiveness and scalability of SQL-based systems for a range of real-world applications. The expected outcomes are expected to provide insights into the construction of more intelligent, autonomous database systems with reduced human query construction and enabling more natural interaction with data.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Large Language Models, SQL data pipelines, text-to-SQL translation, database integration, query optimization, schema comprehension, domain-specific models, NLP methods, query generation, real-time query, data privacy, database automation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The objective of this paper is to explore the critical elements of change management in Oracle Cloud implementations, focusing specifically on user training and adoption. With organizations increasingly relying on cloud-based systems to streamline operations and drive innovation, the successful integration of Oracle Cloud solutions requires a well-structured change management framework. The study examines the strategic planning and execution of user training programs designed to facilitate smooth transitions and optimize system utilization. It further investigates the challenges encountered during adoption phases, including resistance to change, varying levels of digital literacy among users, and the cultural adjustments necessary for embracing new technologies. This research employs a qualitative approach, combining case studies and expert interviews to derive practical insights into effective change management practices. Key factors, such as communication strategies, stakeholder engagement, and the customization of training modules, are highlighted as essential components to achieving high user acceptance and productivity gains. The findings suggest that when organizations invest in comprehensive training initiatives and proactive support mechanisms, the barriers to successful cloud integration can be significantly mitigated. In conclusion, the paper emphasizes the importance of a holistic approach to change management that not only addresses technical upgrades but also prioritizes user readiness and continuous learning. The insights gathered are intended to serve as a guide for practitioners and decision-makers aiming to leverage Oracle Cloud's capabilities while minimizing disruption and fostering a resilient, agile workforce. Ultimately, effective change management and robust training are crucial for enabling organizations to thrive in a competitive digital landscape.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Oracle Cloud, Change Management, User Training, Adoption, Cloud Integration, Digital Transformation, Training Programs, Organizational Change

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This abstract addresses the sophisticated strategies that are involved in effective management and leadership of offshore and onshore Business Intelligence (BI) support teams. The paper emphasizes the paramount importance of reconciling varying work cultures, time zones, and communication channels to create an integrated team atmosphere. The paper touches on the unique challenges and opportunities that come with managing globally distributed teams, highlighting the value of vision, communication, and culturally sensitive leadership styles. From a review of best practices in project management, performance measurement, and stakeholder management, the research identifies essential strategies that enable leaders to develop operational effectiveness and drive innovation. The analysis is centered on how tailored leadership strategies can effectively overcome challenges like knowledge transfer, technology integration, and collaborative problem-solving, ensuring onshore and offshore teams work as a unified unit. Finally, the abstract promotes a light-footed and adaptive management style that leverages the strength of a global workforce to deliver positive BI outcomes, building a sustainable competitive advantage in an ever-changing business environment.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Effective Leadership, Offshore BI Management, Onshore BI Support, Global Team Integration, Cross-Cultural Communication, Strategic Decision-Making, Operational Efficiency, Business Intelligence Solutions

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid evolution of generative artificial intelligence (GenAI) has witnessed extensive application across industry sectors, bringing in new technologies and changes in business model deployment. Though its widespread applicability, more development of efficient pipelines to track the usage and performance of these tools is needed to enhance the integration of GenAI technology in organizational systems. This paper seeks to balance the need in current methodology and frameworks dedicated to tracking the performance metrics for GenAI systems with specific reference to real-time integration of data, accuracy, and scalability. Most current practices of measuring the performance of AI overlook the nature of GenAI application, i.e., its requirement to learn and adapt continuously, in addition to requiring multiple data inputs. Most importantly, the absence of measurable benchmarks for measuring GenAI output only makes measurement more challenging. This study proposes the development and deployment of data pipelines that enable the gathering, processing, and analysis of GenAI usage and performance metrics. The proposed pipelines are designed to provide comprehensive insights into system efficiency, output quality, user engagement, and computational resource usage. Through the application of cutting-edge data engineering techniques, such as automated data gathering and real-time performance monitoring, this study offers a framework for increasing the transparency and accountability of GenAI applications. The findings of this study will guide the development of resilient monitoring systems that can be integrated into various GenAI-powered platforms, guaranteeing optimal performance and well-informed decision-making. The findings of this study have the potential to guide future advancements in GenAI deployment and management, paving the way for more reliable and effective AI-powered solutions.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Generative AI, data pipelines, performance monitoring, real-time data integration, AI performance metrics, system efficiency, computational resource usage, data engineering, AI monitoring, performance optimization, scalability, real-time analysis, user interaction.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Due to its widespread use, English is commonly used by non-native speakers of other languages to communicate. Therefore, in order to be able to successfully convey themselves in a range of contexts, learners must possess essential abilities in it. To be proficient in the English language, one must possess much more than a knowledge of linguistic structures. One among the components is grammar which supports the learning of language with structural understanding. And, preposition usage is common among ESL students. This research focuses on exploring the efficacy of ESL students and to understand the preposition and its involvement in speaking English.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

ESL, linguistic structure.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Use of artificial intelligence (AI) in enhancing web accessibility for the disabled has generated enormous interest over the past few years. Even with enhanced assistive technologies, numerous problems still continue in offering a simple and effective online experience for users with vision, hearing, motor, and cognitive disabilities. AI-based systems such as predictive text, image recognition, and natural language processing have demonstrated extensive potential in offering customized, adaptive solutions to specific needs. However, existing literature has focused on isolated methodologies such as text-to-speech or speech recognition without exploring the entire, multi-modal web interaction in users with different disabilities. This research demonstrates profound shortcomings in existing research, specifically the need for more integrated artificial intelligence systems that incorporate multiple forms of assistance, including gesture recognition, speech recognition, and predictive action, thus offering a comprehensive solution for people with complex or co-occurring disabilities. Existing models are also plagued by a lack of capacity for real-time adaptation to the cognitive state of the user, limiting the potential for enhancing user participation and reducing cognitive load. Additionally, while many AI models are extremely proficient at single tasks, integrating these systems within existing web frameworks is a significant problem. Future work must concentrate on creating end-to-end AI-driven systems that are capable of perceiving user needs across various contexts and hence making interactions more natural and intuitive. In addition, it is critical to incorporate continuous learning processes that are capable of adjusting to changing user preferences and behaviors. By addressing these limitations, a line will be drawn to more accessible, inclusive, and personalized web experiences so that AI technologies can fulfill their potential of crossing accessibility divides for individuals with disabilities.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Web accessibility, image recognition, Predictive, disabled, vision, hearing, Artificial Intelligence

  License

Creative Commons Attribution 4.0 and The Open Definition