Most of the NLP stuff I've been looking at does NER given a long blob of text (e.g., a news article). I am curious what the best method is when you have millions of short strings, say for example names:
Mr. Foo Bar John Doe, MBA, PhD
Say I want to create a model that recognizes the position of the word MBA, the fact that it is surrounded by commas, and so on, and tags based on that. Is NLP the right tool for the job? I could create a database of word+position, but that would be quite large and I'd like to distribute this open source without the end user having to download a large database.