When I joined for my B. Tech in the branch of Artificial Intelligence, I don’t even know what is a programming language. Because of the hype, the fact that there are a lot of job opportunities made me take this decision. Once I joined, I was even more confused about everything, and had a lot of questions in my mind like, What is Machine Learning? How is it different from regular coding? Where do we use it? How to start learning? What are the steps involved in a Machine learning project? Does everyone need to follow the same procedure? How is it different for Data Science ( You may have heard this term too ) ? and many more unanswered questions. The only thing I know is we have the biggest tool for getting the information and knowledge i.e GOOGLE. But you know, when you search for something on google. It throws a bunch of related links and websites onto your screen and you will become even more confused. The difficulties that I have faced while learning makes me write this article. I will try to follow up series of articles on this topic in the next few days. Here it goes.

By now, most students, irrespective of their domains have heard the words “ Machine Learning and Artificial Intelligence”. There’s a lot of hype for it these days due to advancements in the field of computer science and the fact that Machine learning can be used in every other domain as long as the data is present.

Machine Learning, This term has become more common as “breathing” in recent years. Because of the increase in the amount of data that is being generated, the technology is being updated to handle the data, using it for the development and advancements in various fields. Wherever the data is present on large scale, Machine learning can be applied.

Can we apply machine learning to every case?

Yes, we can, but it entirely depends on the feasibility of the use case. For business purposes, we need to understand the feasibility of using ML. Most of the time, using statistics for analyzing the data solves the problems. Feasibility includes the cost of production of the product, availability of resources, etc.

Is it the same as regular coding/programming?

Well, it uses some of the regular programming concepts but it’s not the same as traditional programming. In traditional programming, we give the instructions to the machine/computer and expect the results, whereas, in Machine learning, the machine learns from the previous data of the same problem or the use case and gives the results on future scenarios of the same case by using statistics. To be precise, Machine learning is the combination of computer science, math, and Statistics. To be good in Machine learning, understanding statistics is very important.

How do we start learning?

As previously said, Machine learning is the combination of computer science, mathematics, and statistics, one should have basic knowledge of these three.

Computer Science :

The majority of the Machine learning is being done with python and R. Python is recommended for beginners. Basic concepts of python, using of oops, using various libraries in python like Numpy, and pandas for manipulating the data are required for machine learning. For understanding the data visualizations are made using various libraries like Matplotlib, seaborn, etc.

Mathematics and statistics :

Mathematics includes calculus and linear algebra and statistics play a major role in Machine learning. Calculus is used to create the functions that are best suitable for the data and statistics helps in finding the patterns and trends from the data. Statistics include central tendency, skewness, variance, and bias( which will be explained in detail in later parts).

In the next series of articles, I will be explaining what are the steps involved in a Machine learning project in the way I learned and understand.