This dissertation presents the design and implementation of a comprehensive management system tailored for postgraduate programs. The primary objective is to streamline administrative processes, enhance student engagement, and facilitate effective communication between stakeholders, including students, faculty, and administrative staff. The system incorporates modules for course registration, grade management, scheduling, and document submission, utilizing a user-friendly interface that promotes accessibility and efficiency. Through a combination of database management, web technologies, and user-centered design principles, the system addresses common challenges faced by postgraduate (...) institutions, such as data silos and inefficiencies in information retrieval. A case study approach is employed to evaluate the system's effectiveness within a specific postgraduate program, highlighting improvements in operational workflows and user satisfaction. Feedback from users indicates a significant reduction in administrative burdens and an increase in overall productivity. This research contributes to the field of educational management systems by providing insights into best practices for system design and implementation, ultimately supporting the ongoing evolution of postgraduate education in an increasingly digital landscape. (shrink)

Semantic similarity measures have shown promise in enhancing natural language understanding by quantifying the likeness between textual elements. This paper investigates the application of semantic similarity measures to improve chatbot response relevance. By leveraging word embeddings and similarity metrics, this study aims to bridge the gap between simple keyword-based responses and contextually rich, relevant answers. The proposed approach integrates both traditional lexical measures and advanced vector-based embeddings to enhance user intent interpretation and ensure a more suitable response generation. By refining (...) the candidate response set with these techniques, the chatbot is able to generate highly relevant and contextually accurate replies. The results demonstrate that combining deep learning-based embeddings with traditional semantic metrics can significantly enhance response relevance, leading to more meaningful and human-like user interactions. Additionally, a feedback loop is incorporated to continuously learn from user interactions, further improving the quality of responses over time. (shrink)

The exponential growth of high-frequency real-time financial transactions necessitates scalable machine learning infrastructures capable of processing and forecasting data in real time. This paper proposes a comprehensive design and implementation strategy for such infrastructures using distributed computing frameworks like Apache Spark and cloud services such as Amazon Web Services (AWS). Emphasizing technical specifics, the paper delves into architectural designs, implementation strategies, and optimization techniques that address critical challenges in data ingestion, real-time processing, model training, and deployment. A proof-of-concept implementation demonstrates (...) the feasibility of the proposed architecture on a small scale, highlighting its potential benefits. The findings suggest that implementing a scalable distributed machine learning infrastructure can enhance computational efficiency and significantly improve the accuracy and timeliness of financial forecasts. Future work will involve deploying the proposed architecture in large-scale industry settings to validate its effectiveness in real-world scenarios. (shrink)

This paper investigates how large language models (LLMs) can be applied to publicly available financial data to generate automated financial summaries and provide actionable recommendations for investors. We demonstrate how LLMs can process both structured financial data (balance sheets, income statements, stock prices) and unstructured text (earnings calls, management commentary) to derive insights, predict trends, and automate financial reporting. By focusing on a specific publicly traded company, this research outlines the methodology for leveraging LLMs to analyze company performance and generate (...) investor-focused summaries and recommendations. (shrink)

As machine learning (ML) models become increasingly integrated into mission-critical applications and production systems, the need for robust and scalable MLOps (Machine Learning Operations) practices has grown significantly. This paper explores key strategies and best practices for building scalable MLOps pipelines to optimize the deployment and operation of machine learning models at an enterprise scale. It delves into the importance of automating the end-to-end lifecycle of ML models, from data ingestion and model training to testing, deployment, and monitoring. Approaches for (...) implementing continuous integration and continuous deployment (CI/CD) pipelines tailored for ML workflows are discussed, enabling efficient and repeatable model updates and deployments. The paper emphasizes the criticality of implementing comprehensive monitoring and observability mechanisms to track model performance, detect drift, and ensure the reliability and trustworthiness of deployed models. The paper also addresses the challenges of managing model versioning and governance at scale, including techniques for maintaining a centralized model registry, enforcing access controls, and ensuring compliance with regulatory requirements. The paper aims to provide a comprehensive guide for organizations seeking to establish scalable and robust MLOps practices, enabling them to unlock the full potential of machine learning while mitigating risks and ensuring responsible AI deployment. (shrink)

In today's digital era, large enterprises face the daunting task of managing and analyzing vast volumes of financial data to inform strategic decision-making and maintain a competitive edge. Traditional data warehousing solutions often fall short in addressing the scale, complexity, and performance demands of modern financial analytics. This paper explores the architectural principles, technological strategies, and best practices essential for building scalable data warehouses tailored to the needs of financial analytics in large organizations. It delves into data integration techniques, performance (...) optimization methods, security measures, and compliance with regulatory standards. Through in-depth analysis and real-world case studies, the paper provides a comprehensive roadmap for practitioners aiming to design and implement robust, scalable, and secure data warehousing solutions. (shrink)

As organizations increasingly depend on data-driven decision-making, the complexity of Business Intelligence (BI) systems and their data pipelines has grown exponentially. This complexity introduces significant challenges in maintaining data quality, ensuring traceability, and guaranteeing system reliability. Unmanaged data dependencies in upstream BI components can lead to data inconsistencies, system failures, and compromised analytics. This paper presents a comprehensive framework for monitoring and managing data dependencies in upstream BI systems, with a primary focus on the Dependency Discovery Engine utilizing Static Code (...) Analysis. The framework provides a systematic approach to identifying, tracking, and managing data dependencies across the entire BI ecosystem using advanced static code analysis techniques. Detailed methodologies, algorithms, and implementation considerations for static code analysis are discussed. This framework offers a robust solution for organizations seeking to enhance data quality, system reliability, and operational efficiency in their BI systems. Future research will expand on dynamic runtime analysis and machine learning- based methods to further enhance the framework. (shrink)

Using speech act theory from the Philosophy of Language, this paper attempts to develop an ethical framework for the phenomenon of speech processing. We use the concepts of the illocutionary force and the illocutionary content of a speech act to explain the ethics of speech processing. By emphasizing the different stages involved in speech processing, we explore the distinct ethical issues that arise in relation to each stage. Input, processing, and output are the different ethically relevant stages under which a (...) spoken item or a speech navigates within the range of speech-processing modules. Employing the illocutionary force-content distinction, we specify and characterize the input-related ethical issues, the output-related ethical issues, and the processing-related ethical issues involved in speech processing. Together with illocutionary force-content distinction, we employ the data-information distinction to characterize the stage-wise ethical issues in the phenomenon of speech processing as the ethics of collecting (speech) data, the ethics of contextualizing (speech) data/information, and the ethics of releasing the con-textualized information (processed speech). Immediate ethical issues that arise from the range of speech processing modules are distinguished from distant ethical issues. We also indicate the nature of ethical issues that arise from Speaker Independent speech technologies. (shrink)

The integration of Robotic Process Automation (RPA) into Human Resources (HR) functions represents a significant advancement in organizational efficiency and effectiveness. RPA technology automates repetitive and rule-based tasks, allowing HR professionals to focus on strategic initiatives that add value to the organization. This paper provides a comprehensive introduction to RPA within HR, detailing its applications, benefits, implementation strategies, and how its principles can be transferred to other domains such as finance. By examining the transformative potential of RPA, organizations can better (...) prepare for the future of work, where automation and human ingenuity coexist to drive success. (shrink)

This paper addresses the critical role of automated data quality monitoring in Machine Learning Operations (MLOps) pipelines. As organizations increasingly rely on machine learning models for decision-making, ensuring the quality and reliability of input data becomes paramount. The paper explores various types of data quality issues, including missing values, outliers, data drift, and integrity violations, and their potential impact on model performance. It then examines automated detection methods, such as statistical analysis, machine learning-based anomaly detection, rule-based systems, and data profiling. (...) The integration of data quality monitoring into different stages of the MLOps pipeline is discussed, emphasizing continuous monitoring at data ingestion, pre-training validation, post- deployment drift detection, and feedback loops for model retraining. The paper also addresses key challenges in implementing automated data quality monitoring, including balancing precision and recall in anomaly detection, handling high-dimensional and unstructured data, managing false positives and alert fatigue, and adapting to evolving data distributions. By providing a comprehensive framework for automating data quality monitoring in MLOps pipelines, this paper aims to equip practitioners with the knowledge and strategies necessary to enhance the reliability and performance of machine learning systems in production environments. (shrink)

Due to the Coronavirus Disease-2019 (COVID-19) lockdown implemented by the government, we had to transform our classes into the online sphere. The most commonly used methods of online teaching in Government Medical College, Thrissur were, live online lectures, PowerPoint presentations with narrations, prerecorded videos and assignments. Aim: To assess the logistical aspects, merit and demerits of different online teaching-learning methods among phase-1 medical student in a tertiary care teaching hospital during COVID-19 lockdown Materials and Methods: This cross-sectional study was conducted (...) from July 2020 to September 2020, among phase-1 MBBS students of Government Medical College, Thrissur, Kerala, India. Data was collected from 161 students through a questionnaire consisting of two parts. The first part dealt with logistical aspects, like net connectivity, gadget, expenditure, financial burden, residence and eye strain with online teaching learning methods. The second part dealt with the merits and demerits of each type of teaching learning methods, according to the students. All statistical data was analysed using the Statistical Package for the Social Sciences (SPSS) software version 16. Qualitative data was expressed as a percentage. Results: A total of 97.5% population of students was able to frequently access the online classes. Online learning strained the eyes of 47.8% of students. Mobile phones were used by 92.5% of students for attending online classes internet connectivity was mainly through 4G (71.4%) followed by Wi-Fi (19.9%) connections. Their most preferred method of online learning was online live lectures (35.4%) followed by PowerPoint presentations with narrations (33.5%). Blended learning (online and traditional together) was best preferred method of learning (46.6%). Conclusion: Live online lectures and blended learning were preferred by the majority of students. Thus, the online classes can be continued along with the traditional teaching in future. (shrink)