This video explains one of the basic concepts in linguistics: defining what the input tokens of an AI model mean and how to represent them conceptually.
Foundation Models (OpenAI GPT-3, Google BERT) have taken NLP to a level never seen before. Yet, few data scientists or artificial intelligence specialists master linguistics enough to build a Conceptual Data Model(CDM) that reflects the reality of their business and is compatible with powerful AI NLP models.
It's a specialized video, so please do ask me questions here or on Linkedin if you wish.
https://www.linkedin.com/in/denis-rothman-0b034043/