https://onlineresource.ucsy.edu.mm/handle/123456789/2229 2025-11-21T11:49:33Z https://onlineresource.ucsy.edu.mm/handle/123456789/2814 An Analytical System for Lifelong Learning Achievements: Integrating EDP-Means Clustering and Edu-ETL Processes Mhon, Gant Gaw Wutt In the field of education, analyzing academic performance is vital for understanding student learning behaviors, identifying areas needing enhancement, and developing targeted interventions to improve educational outcomes. Traditional assessment methods typically depend on simple metrics like grades or standardized test scores; which often fail to capture the complexities of student proficiency and behavior. To overcome these limitations, educational researchers have increasingly adopted advanced data mining techniques and machine learning algorithms for a more granular and comprehensive analysis of academic performance data. This research proposes an Enhanced Dirichlet Process Means (EDP-Means) clustering algorithm combined with Educational Extract, Transform, Load (Edu-ETL) processes to evaluate academic performance across various educational levels. The proposed approaches aim to offer greater assurance and clarity in evaluating and supporting student achievements throughout their educational journey. The integration of Edu- ETL processes ensures data quality and consistency, preparing educational datasets for thorough analysis. The architecture of the proposed system utilizes the EDP- Means clustering algorithm, an improvement over the original DP-Means, for enhanced clustering performance. While both algorithms assign data points to clusters based on distance and threshold, EDP-Means introduces iterative optimization steps for improved accuracy and stability. In the original DP-Means algorithm, the number of clusters and the threshold parameter were typically fixed or set based on heuristic choices. In EDP-Means, these parameters are dynamically adjusted based on the data characteristics and clustering quality, leading to more accurate and reliable clustering results. This study demonstrates that EDP-Means performs better and is comparable to traditional K-Means and original DP-Means algorithms in clustering educational data. To validate and prove the performance of EDP-Means, datasets from different fields were used to further experiment EDP-Means and ensure its effectiveness. Furthermore, the analysis of the PySpark environment underscores how the utilization of PySpark enhances the scalability and efficiency of EDP-Means, particularly in processing large-scale datasets. 2024-06-01T00:00:00Z https://onlineresource.ucsy.edu.mm/handle/123456789/2813 A Dependency Analyzing System for Communication Activities in Network Construction Exercises using Trema Htet, Hlwam Maint Nowadays, networking and virtualization technology has been developing in momentum. Software Defined Networking (SDN) has been popular for research and innovation. Universities and research labs are the basic points for innovation because innovation by academia and research organizations can accelerate the rate of change in industries. SDN construction exercises have been developed in e-Learning. Software- Defined Networking (SDN) is a networking approach that decouples the control plane from the data plane, allowing centralized network management. It remains popular in the research field for its benefits that researchers continue to explore various aspects such as: network security, traffic management, network virtualization, edge computing, machine learning and so forth. SDN's flexibility and programmability keep it relevant for emerging technologies and innovative network solutions. When performing network construction exercises, novice learners cannot understand the behavior of their network and fail to satisfy the requirements for the network reachability of communication data. In this system, learners construct SDN network construction exercises by using Trema and OpenFlow Protocol is used for communication between controllers and switches. Here, some learners cannot find their bugs from their settings due to the reasons such as ping cannot find delivery routes including switches, switches have no function to log rules used for choosing output ports for packets, and Trema cannot find execution statements used for setting rules to switches. To satisfy these problems, learners need help and the system will provide analysis results for learners in visual way so that they can narrow down executed statements that cause incorrect communication. This dissertation presents a Dependency Analyzing System for Communication Activities in Network Construction Exercises using Trema. It includes four main modules: constructing Software Defined Network (SDN) Construction Exercises Using Trema, collecting data packets from constructed virtual network, collecting executed statements in controller program, and giving the analysis results to learners so that they can narrow down their visualizing packet location and executing statement information in chronological order. 2024-07-01T00:00:00Z https://onlineresource.ucsy.edu.mm/handle/123456789/2812 Deep Neural Ranking Models for Myanmar News Retrieval Oo, HAY MAN This dissertation focuses on enhancing Myanmar Information Retrieval (IR) system to generate more natural text for a given input text. Typical IR systems have two main components: text query (user needs or preferences) and text documents (related to text query). Both text query and documents are important for the clarity and effectiveness of the IR system. Therefore, this research is emphasized on both text query and documents in Myanmar IR system. In the contemporary era dominated by Information Technology (IT), search engines such as Google have become ubiquitous tools for individuals seeking access to a vast array of information. These platforms serve as indispensable resources, enabling users to effortlessly locate and acquire knowledge on a myriad of topics according to their needs and interests. Searching for News in English or Myanmar has become incredibly convenient, requiring a minimal effort to access a wealth of information. The structure of IR has been altered dramatically by the inclusion of neural models, facilitating a more refined analysis of textual data. The textual data for Myanmar News dataset has been prepared in this research. In this research, the Myanmar News dataset was collected from Myanmar News website. In this dataset, each document contains two parts: title and contents. The evaluations on different neural ranking models were conducted and so the results are thoroughly analyzed and discussed. A comprehensive analysis has started, with immersion in the use of various neural ranking models to comprehend intricate semantic connections, ultimately enhancing the effectiveness of IR systems. Pivotal neural ranking models such as DRMM, MP, Duet, KNRM, PACRR, CONV-KNRM, MZ-CONV-KNRM, which have left a profound impact on the field, are delved deep into, investigating their implications for enhancing the precision and efficiency of retrieval systems. Another evaluation was done using a fine-tuning approach with the pre-trained model, Vanilla-BERT. The superior performance of this model compared to baseline methods, showcasing improvements in MAP, MRR, P@1 and P@3 overall retrieval performance. The implications of these findings extend to retrieve the similarity score results, highlighting the potential for enhanced IR capabilities. 2024-07-01T00:00:00Z https://onlineresource.ucsy.edu.mm/handle/123456789/2811 REAL-TIME HUMAN MOTION DETECTION AND ACTIVITY RECOGNITION WIN, SANDAR One of the interest areas of computer vision is real-time human motion detection, tracking, and activity recognition. It has many applications in a variety of fields, including video processing, abnormally detection, behavior prediction, human- computer interaction, video surveillance, and content-based image retrieval systems. This technology is essential in the fight against crime, terrorism, and threats to public safety. Due to variations in human appearance, changes in illumination, and the volume of data generated, video-based real-time human activity recognition is a difficult and demanding task. Supporting a safe and secure environment for real-time motion detection, tracking, and activity recognition is the aim of this research. The system detects human body parts with skeleton and to define activity based on joint sequence movement and to extract more reliable manner for overlapping area and to solve similar pose with different activities. The goal of this proposed system is to enhance an automated video surveillance system that can identify and track people in both indoor and outdoor settings. The main step of the system involves motion detection, tracking and activity recognition through several steps: First, the system is designed to capture input video and extract region of interest for each frame. And generate features to estimate human and to detect 2D joint projected positions. Then, human detection is applied by using OpenPose detector and categorizes 2D joint sequence of body parts. The system recreates a human skeleton joint in three dimensions using spatial-temporal integration of human body parts. Finally, recognizes the activities such as standing, walking, sitting and running according to joint collection distance and displacement of skeleton joint position. With a deep learning framework, the proposed method operates a robust human skeleton model that is unaffected by changes in the environment or various circumstances. Using joint estimation and position recognition, the system builds a skeleton model from the data perception. The objective of this research is more robust and efficient approach in human detection and activity recognition system from training and testing of multiple data generation by using deep learning approach to recognize different human activities changes in real life environment. The system's total accuracy is 94%, and the proposed approach performs better than expected when it comes to 3D skeleton model-based human detection and activity recognition. 2024-06-01T00:00:00Z