Mostakim, MoinRahman, Md. Arifur2024-06-052024-06-0520232023-10ID 20166040http://hdl.handle.net/10361/23147This thesis is submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science and Engineering, 2023.Cataloged from the PDF version of thesis.Includes bibliographical references (pages 102-107).This thesis is submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science and Engineering, 2023.The convergence of technology in mobile operator networks has ushered in a new era of interconnectedness and communication efficiency. As these networks become increasingly complex, ensuring their optimal performance and addressing customer concerns become paramount. This thesis delves into the realm of ”ML Based Performance Assurance and VoC Management of Highly Convergence Mobile Operator Network,” presenting a multifaceted approach to tackling these challenges through advanced machine learning (ML) techniques by obsolete traditional manual working approach of MNOs.Within the scope of ML-based performance assurance, this study places a strong emphasis on Forecasting Time Series and detecting Anomalies. Leveraging the power of predictive analytics, the research harnesses a spectrum of algorithms including ARIMA, XGBoost, LSTM, Dynamic Linear Model, Prophet, VAR, and GRU. This enables the anticipation of network behavior and facilitates proactive measures for optimization. Additionally, the study integrates sophisticated Anomaly Detection methods encompassing DBSCAN, Isolation Forest, Local Outlier Factor (LOF), One Class SVM, Elliptic Envelope, and Autoencoders. These techniques empower the system to identify and mitigate aberrations in real-time, safeguarding network statistics and ensure business growth. Extending the purview of the research, the study delves into Voice of Customer (VoC) Management within the context of highly converged mobile operator networks. By employing diverse algorithms such as SVM, CNN, GNB, MNB, and LR, the research addresses the critical task of understanding customer insights, preferences,thoughts and concerns. Through effective VoC analysis, operators can tailor their services to meet customer expectations, thereby enhancing overall satisfaction. This thesis contributes to the field by providing a all-encompassing structure for the enhancement and design of mobile operator networks. The amalgamation of ML-based performance assurance and VoC management techniques presents a holistic solution for network operators and service providers. By proactively forecasting network behavior and promptly addressing anomalies, operators can ensure seamless operations. Simultaneously, a holistic and customer centric approach driven by advanced ML algorithms enables the refinement of services based on customer feedback to obsolete traditional working approach of mobile network operators.108 pagesenBrac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.Machine learningTime series forecastingAnomaly detectionVoC managementAlgorithmsMobile communication systems.Thesis