Developing an NLP-based information retrieval tool for analysing impact of AI on EEG studies

The integration of artificial intelligence (AI) into electroencephalography (EEG) research has significantly transformed how brain signals are analyzed, interpreted, and utilized across diverse applications such as neurodiagnostics, brain–computer interfaces (BCIs), and cognitive monitoring. This thesis presents the development of a natural language processing (NLP)-based information retrieval tool designed to systematically explore and analyze the evolving impact of AI on EEG-related scientific literature. By leveraging advanced keyword expansion, co-occurrence analysis, and semantic filtering techniques, the system dynamically identifies relevant articles, clusters key concepts, and ranks publications based on recency and topical relevance. The tool integrates arXiv’s open-access corpus and employs customized domain-specific keyword groupings (e.g., "pathology", "signal processing", "machine learning") to uncover trends, terminologies, and relationships between AI methods and EEG applications. This work contributes both a practical research utility and an analytical framework for understanding how AI has shaped the direction of EEG studies, offering valuable insights for researchers navigating this interdisciplinary domain