You need to analyze unstructured text from multiple sources, including EMR notes, social media posts, and other free-text records. The data varies a lot in writing style, terminology, length, and noise level, and different sources may support different business or research questions.
How would you use natural language processing methodologies to work with EMR data, social media data, and other unstructured data?