Page 16 - AI

It is very important to recognize when bullying is happening. Sometimes, we may

say something which we consider to be harmless but it may offend the other person
and they may feel bullied depending on their experiences, culture, religion or family
values. You should be very careful not to hurt anyone’s sentiments and beliefs.

Sometimes, you might post something on social media or send an email which
might contain some words or phrases which could be interpreted as offensive or
bullying in nature. At times, even typos, grammatical errors or sarcasm may be

misunderstood as bullying.

Can AI help?

Yes, absolutely.

Let us look at the following AI tool:

https://mycomputerbrain.net/php/experiments/ai.experiment19b.php

It helps you train an AI application to recognize texts/posts and match them to
a particular output. No coding as such is required. You can enter a total of 12
expressions, including single words and sentences.

The instructions are on the right side of the menu and help you to:
• Create the training data

• Train the Artificial Neural Network (ANN) through the Start Learning button

(it will say Learning Completed when finished)

• Test the ANN by entering short sentences and phrases

On entering the inputs, ANN will be constructed in real time so you can analyze

how AI is giving the output in real life.

Note: Your input is not stored. When the browser window is refreshed, the experiment
is reset. So you can play with it as many times as you want.

Discuss your learning with your friends and come up with your own training data

to test it further.

• How does AI decide when good words and when mean words are used in
combination?

• Can AI pick up sarcasm in phrases?

• Can you improve the AI model further?

ArtificiAl intelligence 15