Impact of Preprocessing Using Substitution on the Performance of Selected NER Models - Methodology

This paper investigates the effect of preprocessing, specifically word substitution by pseudo words, on the performance of selected named entity recognition (NER) models. The study focuses on explaining the methodology used during the experimental process. The paper comprehensively describes the dataset used, the process of word substitution with pseudo words, the process of model training, the process of executing the test scenario, the performance evaluation criteria and the limitations of the experiment. This paper contributes to the evolving area of Natural Language Processing by providing a comprehensive examination of the impact of preprocessing using substitution strategy on the performance of selected NER models.