Tokenization is a crucial step in natural language processing (NLP), acting as the beginning point for preparing text data. Essentially, it involves dividing a larger block of content – like a sentence – into smaller units called “segments". These tokens can be terms, punctuation marks, or even sub-