00:13Welcome again, and peace be upon you.
00:16I am with you, Ruba Al-Dabbas
00:17Now we will begin our new course, which is machine learning.
00:22Machine learning is one of the important topics that led to the development of artificial intelligence.
00:29Now, machine learning is a process where you have algorithms.
00:34These algorithms have existed for a long time, but because computers were slow, they were no longer effective.
00:39And you didn't have data specialists or a lot of data.
00:47This Sunday was one of the reasons that prevented this issue from developing further.
00:53But during the time when we had the internet and fast computers
00:59You now also have the ability to store large data or small data.
01:05This led to the development of machine learning and the development of artificial intelligence.
01:10Are we in this course or in this introduction?
01:14What are we going to talk about? What is machine learning?
01:17Let's learn about machine learning.
01:18We will also learn about the steps you need to follow.
01:23Why would you reach and train the model and test it?
01:29And what else do you put in the production?
01:31Of course, these are the steps we are talking about during this introduction.
01:35She considers it to be her and every step she takes will be part of this course.
01:42And then we will also get important packages so that we can use machine learning.
01:49I'll tell you about it too during this introduction.
01:51I'll also tell you what the requirements are to start learning machine learning.
01:58Okay, let's start with our introduction now.
02:02Okay, now we need to know what machine learning is.
02:05But before we learn what machine learning is
02:07We need to know how it relates to artificial intelligence.
02:12Is machine learning a subfield of artificial intelligence?
02:17Of course, artificial intelligence is a broad topic with many fields.
02:21Machine learning is one of the high fields found in Artificial Intelligence.
02:26Just as we also have deep learning, which is a part or subfield of machine learning.
02:33Speffield of Artificial Intelligence
02:35good
02:36We need to know what machine learning is.
02:39Our machine learning works
02:41Focusing for what?
02:43In order to do
02:44For ergonomics
02:47We use it because
02:48Machine learning, it works
02:52He works
02:53Fox to develop
02:54Ergothym high
02:55Why? For what reason?
02:58To make computers
03:01Learn and work
03:05As I told you
03:07It was us
03:09Machine Learning Ergonomics
03:10These ergonomics have existed for a long time.
03:13But because the computers were slow
03:15What was in the data
03:16adequacy
03:18They don't know how to develop these ergorithms
03:21When they became the computers
03:23It's fast and you now have data.
03:25Many because of the internet
03:27And it was in our care
03:28We can store this
03:31The data means and became a data or data management algorithm.
03:38It has evolved and helped us to make the computer learn and become capable
03:44Bast works on this data, and all this talk happens without it.
03:51I mean, when I take this, I made this model for my own and I made it for it.
03:59Then I wanted to put it in production, so I put it in this production, so the model
04:05It will start working on its own; I don't need to program it to make it work.
04:10He started working on his own, taking the data that reached him, and he started working
04:14Or on this data, okay, now we want to get into how it works
04:24We need to see the steps that machine learning requires you to take in order to
04:30Let your machine learning model work
04:34The first or most important step in the topic of machine learning is
04:41Collect Data Why? Because the data is the foundation of machine learning
04:46That means if you didn't have data, as I told you, then the algorithm
04:50You will find machine learning available, but it's only because there wasn't enough data.
04:56They couldn't develop it further; the data is the foundation.
05:01The consequences of machine learning, of course, include data such as, for example, you have
05:05You might have stock prices and sales prices, for example.
05:09Do you have pictures, videos, audio, or text files?
05:19You have real-time sensors that take data from and collect it to perform [the task].
05:24If you want to create a model that provides real-time predictions or designs
05:28The other one is in Data Akter, which you can input into their models.
05:35This is machine learning algorithms and you can use them to create things
05:39Prescriptions or designs, and all of this is what it means to use
05:44This data is for training your model, and in the end, it will appear on your screen.
05:51Riesoltz, but of course there's one important thing we need to take into consideration:
05:56The quantity and quality of my data are important; the more you have
06:01More data and more data are needed because it makes your model better.
06:07And what else does quality do? The quality of this data makes
06:13Your model is better, so the data is the foundation of your
06:20Machine learning requires data; without it, we cannot train on this.
06:25The model, without the data being large in quantity or big data, or
06:29Large data and high quality follow, we can't get better models now.
06:37The second step is to process and prepare the data. What does that mean?
06:43This means we now need to get the raw data; the data is usually Messi.
06:47It needs cleaning and preparation so that we can put it in
06:52Our model now involves a number of steps that we follow. The first thing we need to do
06:57Data cleaning involves, for example, if it was in our
07:01Or the data is incorrect, or something like that, we need to do something about this data.
07:09Secondly, we work, for example, if we have certain numbers, numbers
07:16Large and small numbers. In order to make this data consistent, we must use
07:22Normalization and skilling – we'll talk about that in detail later.
07:25We've reached this step, and the last thing we need to do is encoding. Now, encoding.
07:31It's important because, for example, we might have data that could be strings.
07:36Or for example, it could be images, or videos, or audio. We don't have this data.
07:43We can enter them as strings, fetch photos, or pictures or videos. What is required?
07:50The data that our model is working with must be numbers, so what is encoding?
07:55It converts all this non-number data into numbers, and we'll talk about that too.
08:03We'll discuss it in more detail later. What is data splitting? That's a topic for you.
08:09It's important that when we finish all these things, we need to split the data.
08:15Hello, data splitting. We do it by splitting three sats, first thing, second thing.
08:23Second and last thing, the test is, you can choose whether to use it or not.
08:32You can put it in now if you have either small or large data.
08:38If you didn't have large data and the data you were using was small, you could
08:43Use whatever you want, but leave the training satellite and test satellites as they are. Now, training satellites are available.
08:49We provide you with this data that we use to create
08:52The model, as its name suggests, is used to test our model, and the test set is used to test this model and compare its results.
09:08Will the results be similar after we use the test?
09:13The akirusi (high) means there are no errors; this is the test we use, that's why these are the three six.
09:24We can use them in our model
09:26Okay, we'll stop here, and in the next video we'll continue with these steps to learn how to walk.
09:35Learning is working, thank you and see you in the next video
