Textese is a form of text commonly used with typed messaging applications. It consists of a series of terse abbreviations that are used as a substitute for more verbose words or phrases, for example, BTW (short for by the way). Twitter feeds is another common place where textese is used. Due to its prevalence, it can be important to determine the POS for such message types.
In this recipe, we will be using the Stanford NLP API to demonstrate how to find the POS of textese. We will use the gate-EN-twitter.model model in conjunction with the MaxentTagger class to perform this task.
