The Path from Structured to Unstructured Data

Working with unstructured data opens a new world of NLP possibilities

For much of the history of enterprise ML, structured data has been used due to its ease and simplicity. Now, unstructured data creates a vast new class of ML tasks. From recognizing sentiment, sarcasm, and hashtag inference in raw text, to creating captions for images, the practical applications of NLP in making sense of immense volumes of data are limitless. Knowing the difference between structured and unstructured data is one step in understanding the power of NLP and how it can help your enterprise.