What is layman-script?

 

layman-script /ˈlāmən skript/

A crude, non-domain specific English language parser.

It can parse English at a “level 1” reading level.

Step 1: Tokenization

Step 1: Tokenization

Mr. Bill went to the store.

Tokenization is the processing of splitting up a group of text.

ex) Paragraph -> Sentences, Sentences -> words.

It’s pretty simple as it essentially converting a sentence to words using a simple split() function.
But it’s not that simple

Step 1: Tokenization

For example if this is in a paragraph and you're splitting to sentences by ending punctuation. This sentence...

Mr. Bill went to the store.

Will split into this...

[“Mr.”, “Bill went to the store.”]

Step 1: Tokenization

Also if you have one sentence and you split every word. You will get something like this.

[“Mr.”, “Bill”, “went”, “to”, “the”, “store.”]

However, when it is split like this, their is no correlation between the words. If you split by the word, finding the subjects is easier. 

[“Mr. Bill”, “went”, “to”, “the store”]

Step: 2 Part Of Speech Tagging

Step 2: POS Tagging

The process of assigning the correct part of speech to each word in a sentence. While many more complicated methods to tag, at its basic level it needs a word list to compare against.

 

So I have a list of verbs, adverbs, prepositions to compare against. Then I will easily be able to correctly assign a POS to each word.

 

Right?

I wish...

Step 2: POS Tagging

(kinda ties to the main parsing)

Let’s take the example:

Mr. Bill went to the store.

After applying the POS tagger in its crude form:

(ORANGE= verb)

Mr. Bill went to the store.

Uh Oh! It counted Bill, and store as verbs even though they are the subject:

Step 2: POS Tagging

(kinda ties to the main parsing)

So currently this pass of tagging is very inaccurate, but it will do for now and it’s solution will be solved in the next step.

Step: 3 Finding Subject and Verb

(lexical analysis)

Step 3: Finding Subject and Verb

PREFACE: This could be way simpler if the tokenization in the beginning was done better.

Essentially we need to simplify the sentence to a SUBJECT VERB THING sentence.

 

George and Ansh like burritos     -------->     George likes burritos.

                                                                            Ansh likes burritos.                                                                           

Step 3: Finding Subject and Verb

But first we need to set some rules for what defines an assigning verb.

1. The verb should follow after the subject or have it’s own subject associated with it. 
   1. George ran to the pool
2. The verb must not have an infinitive
   1. George likes to swim. ← Doesn’t count
3. The verb can’t be in a list.
   1. George likes to swim, eat, and sleep.                                                                          

Step 3: Finding Subject and Verb

Anything before the assigning verb can be assumed to be a Subject Phrase. And after it can be assumed to be the Noun Phrase. 

Mr. Bill went to the store.

Here is an example of this in a more complicated scale. This is called Context-Free Grammar.

 

 

 

 

 

 

 

 

Source: Tutorials-Point

That's kinda the gist of how the parsing works.