Machine Learning Tutorial for Beginners

In general programming we (developers) write the code / logic to find out the end result, in Machine learning concept the result is produced by Machine (a generic set of algorithm) based on data we provide.

There are various machine learning algorithm to use depending on situation. below are the few commonly used machine learning algorithms.

• Liner Regression Algorithm
• Logistic Regression Algorithm
• Decision Tree Algorithm
• SVM Algorithm
• Native Buyers Algorithm
• KNN Algorithm
• K-means Clustering Algorithm
• Random Forest Algorithm
• PCA – Principal Component Algorithm

What is Machine Learning ?

In machine learning instead of writing entire code we provide data to get the output, and the output is decided by some predefined generic algorithm. (there are different types of algorithms; we can try each one to check the best expected output.)

Think of situation like online shopping, how online shopping sites recommend you things you to buy, they recommend based on your previous shopping data or based on product you click the most.

Machine learning mainly focuses on developing of programs that can access data dynamically then analyse it and learn it.

There are different types of machine learning algorithms; they are broadly classifieds into three types.

• Supervised machine learning algorithms

  Supervised learning is something where Machine will predict based on previous training dataset, which means it knows the approximate correct answer (learned from previous dataset example).

  In supervised learning we have input variable (a) and output variable (b) , and we use an algorithm to learn the mapping function from input to output (a to b).

  The goal is, when next time we get a new input (a1) , we should be able to predict the output b1 (based on previous learning)

  Supervised learning can be further classified into regression and classification.

  • Regression: A regression problem is when the output variable is a real value, such as "amount" or “height”.
  • Classification: A classification problem is when the output variable is a category, such as "green" or "red"

  Supervised machine learning algorithms are:

  1. Random forest for classification and regression problems
  2. Linear regression for regression problems
• Unsupervised machine learning algorithms

  Unsupervised learning is we only have input data (a) and no corresponding output variable.

  The Unsupervised learning goal is to understand the data structure or distribution pattern, so we can learn more about the data.

• Semi-Supervised Machine Learning

Logically yes, (but logic has limitation), actually some programming experience is helpful because you get familiar with things like writing different type of syntax, basic coding principal, different ways of writing conditional logic, those experience would defiantly help to understand machine learning faster, but, if you don’t have any previous experience then also you can start learning machine learning.

The most popular machine learning languages are Python and R Language, but there many other languages like Julia, Scala, Ruby, Octave, MATLAB,C, C++ are used for developing machine learning

Here are some machine learning implementation example.

• Machine Learning Tutorial in C#
• ML regression analysis using C#