What's the difference between text analytics and natural language processing?

Phrases like "text analytics/text mining", and "natural language processing" are often used interchangeably. While there are considerable topics of overlap between the subjects that come under these terms, they are not identical.

Text Analytics/Mining

• Getting structure and patterns out of textual data (a type of data mining; traditional data mining deals with structured data)
• Heavily uses natural language processing concepts & tools to extract structure
• Uses rules or statistics to impose or derive structure
• Tasks like natural language generation are usually not part of this analysis, being more of a synthesis task
• May use information retrieval concepts such as as tf-idf, bag of words etc.; thus text mining may do more shallow language processing rather than any deep linguistic analyses.

Natural Language Processing

• A mixture of linguistics, computer science, and maths
• Usually considered a division of artificial intelligence studies
• Both analysis and generation tasks are part of this
• Often described as getting computers to understand natural language