WEBVTT

00:00.000 --> 00:05.000
Developing a basic AI model.

00:05.000 --> 00:10.000
In this video, we will understand the types of data sets used in AI.

00:10.000 --> 00:13.000
How data sets are split for training and testing.

00:13.000 --> 00:17.000
Steps to develop a simple AI model.

00:17.000 --> 00:20.000
Introduction to data sets.

00:20.000 --> 00:23.000
A data set is like a library for AI.

00:23.000 --> 00:26.000
It contains examples the AI uses to learn.

00:26.000 --> 00:31.000
To build a basic AI model, we first need to understand data sets.

00:31.000 --> 00:35.000
Types of data sets.

00:35.000 --> 00:38.000
Data sets come in different types.

00:38.000 --> 00:42.000
First, there are structured data sets.

00:42.000 --> 00:49.000
Meet, organize data, like spreadsheets and tables where everything fits into rows and columns.

00:49.000 --> 00:53.000
Then, we have unstructured data sets.

00:53.000 --> 01:01.000
Messy data like images, videos, and text documents that don't fit easily into tables.

01:01.000 --> 01:05.000
Data sets can also be labeled or unlabeled.

01:05.000 --> 01:13.000
Labeled data sets have clear answers or tags, like a photo labeled cat or dog.

01:13.000 --> 01:17.000
Unlabeled data sets don't have those answers yet.

01:17.000 --> 01:21.000
The AI has to figure them out on its own.

01:21.000 --> 01:24.000
Teaching AI to learn.

01:24.000 --> 01:29.000
When we teach AI to learn, we split the data set into two parts.

01:29.000 --> 01:32.000
The training set and the testing set.

01:32.000 --> 01:39.000
AI uses 75% of the data set as training data to learn patterns and make predictions.

01:39.000 --> 01:45.000
While the remaining 25% is used as testing data to evaluate the model's performance

01:45.000 --> 01:50.000
and check how well it makes predictions on new data.

01:50.000 --> 01:53.000
Steps to develop a simple AI model.

01:53.000 --> 02:00.000
Now, what are the steps to develop a simple AI model using a predefined data set?

02:00.000 --> 02:03.000
Split the data set into training and testing sets.

02:03.000 --> 02:06.000
Train the model using the training set.

02:06.000 --> 02:10.000
Test and evaluate the model using the testing set.

02:10.000 --> 02:14.000
Adjust parameters to improve its accuracy.

02:14.000 --> 02:17.000
We explored how data sets are used to train AI.

02:17.000 --> 02:20.000
Learned about different types of data sets.

02:20.000 --> 02:26.000
And understood the importance of splitting data into training and testing sets to build an accurate model.