Meaning can be like grains of sand. Merzouga, Morocco, photo by @hazardos

NLP & Lexical Semantics

The computational meaning of words

Finding out the computational meaning of word is an exciting process. However, I thought I would closer into one concept relating to how we can think about words. I write this article to learn more, and I will attempt to keep it as short as possible. This one concept mentioned is:

1. Lexical semantics.

What do words mean?

What do individual words mean?

That seems like an easy question if you talk to someone.

Say for example: forward.

Let us take two sentences.

“Forward into the ocean.”

“She was leaning forward.”

We can separate forward in both these cases.

‘Forward’ or ‘forward’ operates in two different contexts relating to other words.

Lexical semantics: is about individual words in context.

There are lexical relations, how meaning relate to each other.

According to Wikipedia, the study of lexical semantics looks at:

• “The classification and decomposition of lexical items
• The differences and similarities in lexical semantic structure cross-linguistically
• The relationship of lexical meaning to sentence meaning and syntax.”

“Forward into the ocean.” Is a sentence where the forward is referring to ‘ocean’ connected by into. “She was leaning forward.” This on the other hand refers to ‘she’ and a past tense action.

Clearly, one can tell based on English language that these are different, but how are these semantics represented? How can we recognise the meaning of these two?

One can see how other words relate to that one word. The following examples are taken from the Wikipedia page on lexical semantics.

Hyponymy and hypernymy: “…refers to a relationship between a general term and the more specific terms that fall under the category of the general term.”

Color (hypernym) → red, green, yellow, blue (hyponyms)

Then there are synonyms, and this may be more commonly known. Synonymy refers to: “…words that are pronounced and spelled differently but contain the same meaning.”

Happy, joyful, glad

Antonymy refers to words that are related by having the opposite meanings to each other. There are three types of antonyms: graded antonyms, complementary antonyms, and relational antonyms.

dead, alive
long, short

Homonymy is about the relationship between words that are spelled or pronounced the same way, but hold different meanings.

bank (of river)
bank (financial institution)

Lexical items can map onto events too.

There can be states, processes and transition.

a. The door is closed.
b. The door closed.
c. John closed the door.

“(1a) defines the state of the door being closed; there is no opposition in this predicate. (1b) and (1c) both have predicates showing transitions of the door going from being implicitly open to closed. (1b) gives the intransitive use of the verb close, with no explicit mention of the causer, but (1c) makes explicit mention of the agent involved in the action.”

If you are interested in reading more on the practical application within NLP I would recommend checking out this post by HuggingFace written by Thomas Wolf.

This is #500daysofAI and you are reading article 464. I am writing one new article about or related to artificial intelligence every day for 500 days.