Introduction to Machine Learning: Part 1

Introduction

 
Hello, world! This is my first article on C# Corner. The topic I will be explaining is Machine Learning. Yes! You read it right! Machines do learn! After formal discussion. I will be moving to Microsoft Azure Machine Learning. Knowingly and unknowingly we are all using machine learning every day, or should I say every hour? How? Questions galore! So here is a list of a few applications using machine learning.
  • When using a search engine, search engines use learning algorithms for ranking the pages so that you can get the most relevant page for your query.
     
  • Whenever you upload a photo of a group on Facebook or any similar sites then your photo is recognized and gives you suggestions for tagging by recognizing the faces of your friend.
     
  • Email Classification! Based on various factors you get your email in an INBOX or SPAM folder, another fantastic application of machine learning and so the list goes on.

Machine learning

 
In general, as do human beings learn by example, from our mistakes similarly a machine can also learn by example. In technical terms, we say training the algorithm by using huge data volumes and then use that to predict certain patterns, trends and classifications, and various other outcomes.
 
Mathematical definition (Mitchell 1997):
 
A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at the tasks improves with the experiences
 
So the next question in your mind might be, how to make machines learn?
 
In general, if we write a program, provide input and after processing we get output.
 
Traditional Programming paradigm
 
1.1 Traditional Programming paradigm
 
In machine learning, we provide data and the desired output and we get programs, yes programs!
 
Machine Learning Programming paradigm
 
1.2 Machine Learning Programming paradigm
 
To be continued.