AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

Unable to understand meaning of tag NONE *-1 in Penn Treebank example
 
 

I am completely naive and new to NLP. I am following a course on NLP on coursera. As I was studying about parse tree, I came across following sentence to parse:

`Because the CD had an effective yield of 13.4 % when it was issued in 1984 , and interest rates in general had declined sharply since then , part of the price Dr. Blumenfeld paid was a premium—an additional amount on top of the CD ‘s base value plus accrued interest that represented the CD ‘s increased market value .`

And partial parse tree to this sentence is:

  (S
    (SBAR-PRP
    (IN Because)
      (S
      (S
        (NP-SBJ (DT the) (NNP CD))
        (VP
          (VBD had) (NP
          (NP (DT an) (JJ effective) (NN yield))
          (PP (IN of) (NP (CD 13.4) (NN %))))
              (SBAR-TMP
              (WHADVP-4 (WRB when))
              (S
              (NP-SBJ-1 (PRP it))
              (VP
                (VBD was) (VP
                (VBN issued)
                (NP (-NONE- *-1))
                (PP-TMP (IN in) (NP (CD 1984)))
                (ADVP-TMP (-NONE- *T*-4))))))))

I am not able to get meaning of **-NONE- *T*-4** and **-NONE- *-1**. Anyone can help me?

 

 
  [ # 1 ]

After going through a number of examples, I think I got meaning of *NONE* in parse tree output. Let’s take an example, suppose we have to parse following sentence:

Ricky is a boy and likes fruits.
In above sentence two sentences are joined with coordinating conjunction and.
Sentence 1: Ricky is a boy.
Sentence 2: likes fruits.

In first sentence subject is clearly given which is Ricky but in second sentence it is not explicitly written (although it’s clear we are talking about Ricky in second sentence too.). So when we parse such sentences, in the parse tree we get *NONE* for the NP (Noun Phrase).

 

 
  [ # 2 ]

That sounds about right. What I know of grammar parsers is that they go over a list of standard patterns, sort of templates, and give you the one that matches best. In this case it looks like the standard pattern for the verbal phrase (VP) contained a noun phrase like “was issued orders” or “was given a book”, but found none to fit that slot. ADVP-TMP may similarly have expected a temporal adverb to follow. In short: they are ignorable.

 

 
  login or register to react