| Abstract [eng] |
This master thesis explores the application of machine learning (ML) to customer risk assessment strategies in the aviation industry. The aviation sector, which is the backbone of global transport and trade, faces unique challenges such as regulatory compliance, reputation management and mitigating risk in high value transactions. Traditional approaches to managing these challenges are becoming increasingly inadequate due to the complexity of global networks, dynamic regulatory frameworks and vast amounts of data. Therefore, advanced technological solutions are needed to ensure operational and reputational integrity. The study focuses on the integration of ML models into processes such as Know Your Customer (KYC) and Continuous Due Diligence (CDD), with an emphasis on reputational risk assessment. Building on methodologies used in the financial sectors, the study applies techniques such as Natural Language Processing (NLP) to analyse large datasets including social media, publicly available data and sanctions lists. Models such as GPT, BERT, RoBERTa and VADER are being used and developed to meet the specific needs of the aviation industry and provide greater efficiency in sentiment analysis, reputation assessment and risk classification. The study describes a new ML-based framework for dynamic and efficient reputational risk assessment for aviation companies. The developed models demonstrate improved accuracy in risk identification and are able to adapt to rapidly changing data patterns and regulatory requirements. The implementation of these solutions aims to improve compliance, maintain stakeholder confidence and strengthen industry resilience in a competitive and interconnected global environment. Finally, this research contributes to the academic and practical understanding of the use of ML to address critical issues in the aviation industry. The results not only highlight the potential of AI-driven tools, but also provide useful insights on how to improve KYC and risk management processes to ensure sustainable and safe operations. |
