Popular concepts in AI #Basics #AIForDummies

Looking at the trends and future tech, I believe it has become mandatory for people in technology to at least understand the very basics of AI. I believe even the simplest software will be AI-driven in the future.

For organizations, artificial intelligence can take them to the next level. And it is high time for organizations/businesses to know which AI tool will help them grow more. The same thought has brought me to learn the basics of AI and then write this blog for me to refer to Later : )

If you see in current times, a well-designed AI system can help you better interact with customers and even can help you identify new markets!!

These systems can be helpful in predictions and if the algorithm is strong, it can also be designed to make real-time decisions. There is a huge amount of this potential with this tech. But with so many possibilities it can be a hard place to see from where to and how to start with it. So I started by exploring the high-level concepts as -

1. Machine Learning
2. Artificial Neural Network
3. Deep Learning
4. Natural Language Processing (NLP)

I will focus the most on the NLP, everything else just follows ...

Natural Language Processing (NLP)

Basically, Natural language refers to the way we, humans, communicate with each other. It is broadly defined as the automatic manipulation of natural language, like text or speech, by software. Common tasks include parsing, speech recognition, part-of-speech tagging, and information extraction. NLP focuses largely on converting text to structured data.

NLP process can be broken down into three subsets-

• SRS — Speech Recognition System (Identify)
• NLG — Natural Langauge Generation (Write)
• NLU — Natural Langauge Understanding (Read)

NLP Example:

To understand it better, let me give you an example of the above three terms:

So let’s say you click on the mic icon on your phone and speak something -

“Hi, what's the weather like today”?

It will go through the below process -

1. Take the Audio file

2. Upload it to the Speech Recognition system (SRS). Machine identifies spoken words.

3. Convert the Audio to the Text file (NLG). While processing this, it tries to find the Patterns in Data (Text Patterns), tries to map the most common words.

4. Apply the Natural language Understanding (NLU)

5. Convert it into something which the search engine understands (Search engine friendly terms)

For Machines and the NLP, algorithm unlike humans, It’s a data challenge and not a creative issue. Human writers have to be creative, while the machine doing the NLG, they need to work with the available data and the application of the algorithm on top of it. So for any Automated system, a huge amount of data can help in providing the most possible best output.

This is the simplest form of explanation I could offer. What I derived from it is that To build or buy an AI system, below are the things one should be at least be aware of.

⌘ — Natural Language Processing (NLP)

⌘ — Top NLP tools (like NLTK library, TextBlob, etc)

⌘ — Generate natural-sounding text (NLG)

⌘ — Understand text and speech (NLU); Popular Example is ‘Responsive Automated Customer service Bot’.

I think with the basics of AI, one must know and understand the Hidden Markov Model too, so few words about it -

Hidden Markov Model

There is an infinite number of ways that people can mix and match words and phrases. The way pronounces the same word. For this, the AI system often relies on something called the “Hidden Markov Model”. It helps the computer system to determines the next set of words based on the previous information (data and trend analysis) provided during the process of Speech Recognition process (SRS).

I hope you’ve enjoyed this blog about AI — artificial intelligence. Throughout this small blog, I have tried to cover some very high-level AI concepts. I am looking forward to learning more and I plan to post about interesting new things in the world of AI. Thank you for reading this. Please drop your comments in feedback!