NLP in a Nutshell

NLP basics and different AI terminologies

Hey You!

Today I am sharing some basics of NLP. If you are planning to learn LLMs, this will be a brief and best start for you.

Happy Learning!

Table of Contents

Introduction

If you want to understand how ChatGPT works on a technical level we need some fundamentals.

Our goal is to understand how NLP works, and how we can bridge the gap between human language and machine language.

NLP's objective is to create a machine that can replicate how humans interact in their natural language.

Check this Venn diagram →

All terminologies of AI and Data Science

AI is an umbrella term for all the fields you ever heard about.

We are trying to replicate human intelligence into machines.

Machine learning and deep learning are the techniques used to achieve artificial intelligence.

We need to give the capability of human natural language to a machine so that it understands our language, this can be achieved by processing our natural language thus NLP.

You can use ML and DL techniques to solve problems in NLP.

Not all the tasks need ML and DL for working on text data in NLP, that can be solved using Python scripts.

2 branches of NLP →

1 → NLU (Natural Language Understanding)

It works on context and intent which is semantic analytics, semantic search(ex: Google search)

Word-to-word translation is not happening here. It translates based on the context of the text.

Context

Let’s say I am complimenting you with this, “Your T-shirt is killer”

As humans we know, this is a positive statement. But for a machine the word “killer” is negative hence it will consider the sentence as negative.

We need the machine to understand the context of it, not only focus on individual words.

Intent

Let’s say you go to the vegetable seller and say, “My mom gave me money to buy 1kg of tomatoes otherwise she will be angry”

For the vegetable seller will understand that you want 1kg of tomatoes, other information is irrelevant to him.

His only concern is to know your intent.

2 → NLG (Natural Language Generation)

As the name suggests, you are going to generate text. It works on generating the next word in a sentence.

When I was pursuing my master's in Data Science & AI, there was a subject, Statistical Modeling.

Generative Modeling is a branch of statistical modeling, in which you predict the next number or word based on the previous number or word, using probability.

When you see any generative AI project, it is generating text which is nothing but the prediction of the next word in a sentence based on the probability, given the previous words.

The objective of NLP

If you want to give instructions to a machine, you can use any programming language like Python.

There is a gap between you and the machine.

The objective is to reduce this gap in which you don’t need to learn programming to give instructions to a machine.

This will only be possible by giving this capability to the machine to understand human natural language.

You don’t need to write code to get an output from ChatGPT, you can do it by using human natural language.

NLP Applications

  • Sentiment analysis

  • Text Summarization

  • Machine Translation

  • Information retrieval

  • Email spam detection

  • Question and answering bots

  • Text generation - autocomplete, chatbots,

  • Toxicity classification - threats, insults, hatred

  • NER - Named Entity Recognition - name, organization, location or quantities

Socials

Be part of 50,000+ like-minded AI professionals across the platform

LinkedInTwitterInstagramMediumWhatsApp 

Please reply to this email with your requirements or suggestions on what you want in future newsletter content.

PS: build your newsletter, → Here

Reply

or to participate.