I have been thinking to write about Machine Learning for quite a while so I'm finally writing my first blog using whatever I learned for the past few months. This will be a series of blogs covering all the fundamental topics you need to get start with Machine Learning.
Pre-Requisite
To get most out of this blog series, you need to have the understanding of Python.
IDE
For this series, We will be using Google Colab that allow you to write and execute arbitrary Python code from your browser without worrying about the computing resources.
Problem Statement
Through this series, We will be trying to build a real time NIDS solution using Machine Learning. However, we can deviat from this to get a better understanding of a specific concept.
Dataset
The dataset used for this series is CSICS-2017 which you can easily download and find information from here.
Supervised Learning
We will be using Supervised Learning to solve this problem. Supervised Learning algorithm are trained using labeled examples, such as an input where desired output is known.
For example:
We want to create a model which tells us whether a given Email is SPAM or HAM. The model is trained with dataset which contain millions of email each tagged with either HAM or SPAM. This is the simple example of supervised Learning.
In a nutshell, our Machine Learning models receive a set of input (emails) along with cerrosponding outputs (HAM or SPAM), and the algorithm learns by comparing its actual/predicted output with correct outputs to find errors. It then modifies the model accordingly in order to get the desired accuracy.
Let's continue our journey with the next blog.
Top comments (0)