IKH

Parts-of-Speech

Let’s start with the first level of syntactic analysis: POS (parts-of-speech) tagging. A word can be tagged as a noun, verb, adjective, adverb, preposition etc. depending upon its role in the sentence. Assigning the correct tag such as noun, verb, adjective etc. is one of the most fundamental tasks in syntactic analysis.  

Let’s say you ask your smart home device a question – “Ok Google, where can I get the permit to work in Australia?”. Now, the word ‘permit’ can potentially have two POS tags – noun and a verb. In the phrase ‘I need a work permit’, the correct tag of ‘permit’ is ‘noun’. On the other hand, in the phrase “Please permit me to take the exam.”, the word ‘permit’ is a ‘verb’.

Assigning the correct POS tags helps us better understand the intended meaning of a phrase or a sentence and is thus a crucial part of syntactic processing. In fact, all the subsequent parsing techniques (constituency parsing, dependency parsing etc.) use the part-of-speech tags to parse the sentence.

Prof. Srinath will explain the commonly occurring parts of speech tags.

Note

At 1:59, Professor has given the example of Noun (NN) as People, Animals etc. But once we assign particular value to these group then it becomes Proper noun (NNP). For example People has assigned value Ronnie so now it becomes Proper noun.

Note

You do not need to remember all the POS tags except for a few which are listed later on this page. You’ll pick up most of these tags as you work on the problems in the coming segments, but it’s important to be aware of all the types of tags. Now, let’s look at some other tags.

There are 36 POS tags in the Penn Treebank in NLTK. The file given below enlists the most commonly used POS tags. It is recommended to remember these common tags. It’ll help you avoid the trouble of looking up meanings of tags in the upcoming segments, so you can focus on the core concepts.

Note that the set of POS tags is not standard – some books/applications may use only the base forms such as NN, VB, JJ etc without using granular forms, though NLTK uses this set of tags. Since we’ll use NLTK heavily, we recommend you to read through this list of tags at least once.

Additional Reference:

Report an error