Difference between AI, ML, Deep learning and Data Science

Hello everyone. In this video, we will be discussing the most fundamental thing in data science, that is the difference between artificial intelligence, machine learning, deep learning, and where data science fits in all these. So in this video, we will be discussing about all of them in detail so we can understand them better. At first, we'll be talking about AI, which means artificial intelligence. Artificial intelligence enables the machine to think, which means without any human intervention, the machine will be able to take its own decision. This is what AI means and a self-driving car is a great example of artificial intelligence. Now let us move towards machine learning. So basically, machine learning is a subset of artificial intelligence. So what does machine learning do? It provides us statistical tools to explore the data and understand data. When I talk about machine learning, it has three approaches. The very first is supervised machine learning. Then we have unsupervised machine learning, and at three, we have reinforcement or semi-supervised machine learning. Let me give you a good example of supervised machine learning. So here, we will be having some past data, and with this data, we will be able to do future prediction. Suppose I have two features in a data, that is height and weight, and I have to classify that whether the person belongs to fit or unfit, that is obese category. So in this, I will already have the data with me before I begin the tasks. So learning and using the past data to make future predictions is supervised machine learning. Now in second category, when I talk about unsupervised learning, I'll not be having any labeled data. That means in my data set, I'll be not knowing what my output is. There will be no past data as such supervised machine learning. So which kind of problem does we solve in unsupervised learning? So in this category, we'll be solving clustering kind of problems. When I say clustering, what it does is that, based on similarity of the data, it will try to group the data together. And here, there is some mathematical concepts used like Euclidean distance, et cetera. Now in reinforcement or semi-supervised technique, some part of the data will be labeled, and later on, some part of data will not be labeled. So the machine will learn slowly seeing the past data. So to sum up or to conclude, machine learning provides statistical tools to explore and analyze data. This is the work of machine learning, that is, providing statistical tool to explore and analyze the data. So just like machine learning, in deep learning also, we have various techniques. One is ANN, that is, Artificial Neural Network. Second is CNN, Convolutional Neural Network. And the third one is RNN, which means Recurrent Neural Networks. So data which is actually present in the form of numbers will be solved by ANN, that is, Artificial Neural Network. Suppose our input is in form of images, then we will use CNN. And if our input is in the form of time series, then we will use RNN. So for numbers, we use ANN, for images, we use CNN, and for time series kind of data, we'll use RNN. Apart from all these techniques, we have some advanced techniques like transfer learning as well. Here, understand that you first have to use deep learning and machine learning, and then try to derive AI, which means Artificial Intelligence application from them. Remember, we start from deep learning and machine learning, and our main goal is to achieve and to reach artificial intelligence. So like by using deep learning and machine learning, I want to create self-driving car, which is an example of AI. So by using deep learning and machine learning, we try to create applications for artificial intelligence, just like I gave you example of self-driving car, which is an application of artificial intelligence. Now the question arises, where data science fits in all these techniques? So data science is a technique which tries to apply all these techniques, that is, artificial intelligence, machine learning, deep learning, and also, it uses some mathematical tools like statistics, probability, linear algebra, etc. and this is what data science means, that is, using these mathematical tools and all these techniques, artificial intelligence, machine learning, and deep learning, and then solving the problems of the data. So these are all the things, techniques, and algorithms which one should learn if you wish to become a data scientist, that is, he must learn deep learning, machine learning, artificial intelligence, and enough knowledge about data science. Then only he can become a successful data scientist. I hope you have got enough knowledge regarding all these, that is, the difference between machine learning, deep learning, artificial intelligence, and where data science fits in all these. I hope you have understood all things very well. Thank you so much for watching this video till the end.