Machine Learning - An Overview

Artificial Intelligence brings a new chapter of technology that helps us to analyze data. Rather, we can say Artificial intelligence gives a response like a human brain and can think like a human. But the question is,  how does it work? Let me tell you guys a few concepts of Artificial Intelligence and Machine Learning.

What is Machine Learning?

By the terminology of Machine Learning, you can understand that a machine can learn by itself, but how? It is a part of Artificial Intelligence. Here, we can train a machine by using the Dataset. So data is very important or you can say, a couple of statistical models can train a Machine by using Dataset like historical dataset of Stock Market, an experimental dataset of Medical research, Car sales Dataset of the cross financial year, etc. etc.

How can a machine learn from a Dataset?

There are the following standard steps that the industry is using for Machine Learning.
  1. Collecting the data.
  2. Preparing the data.
  3. Choosing the right model.
  4. Train the Model by using Statistical Diagram.
  5. Evaluation.
  6. Tuning.
  7. Prediction and result.

Collecting the data

Data collection is a very important phase of Machine Learning because you need to build a predictive model.

Preparing the data

After collecting the data we have to get ready for the next label because raw data might have lots of redundancy, that is why data cleanliness is very important in this stage.

Choosing the right model

The machine learning model is a Mathematical and statistical presentation that evaluate the data. There are a few algorithms that I am going the share with you.
  1. Linear Regression
  2. Logistic Regression
  3. Decision Tree
  4. SVM
  5. Naive Bayes
  6. kNN
  7. K-Means
  8. Random Forest
  9. Dimensionality Reduction Algorithms

Train the Model

In here collected data will go through the Mathematical algorithms or Model. The machine will experience those historical data and improve the prediction. In Supervised Learning, data is input and labels are output and the function maps between the input and output. In Unsupervised Learning, we do not have any idea of the labels or output. Train  the model by using the patterns.


In the phase of Evaluation, we generalize the accuracy of a model. But how to measure the accuracy of the data model? The purpose of using the Evaluation is to test the model.


After testing the model you need to set the Parameter tuning to get a more accurate result.
In here you don’t know which is the right value for a model. You need to go through the trial and error.
It can take time to get the right value for hyperparameters of the model to discover the parameters that will get the appropriate result.


The final stage is a Prediction. In this stage, we will get the final answer from the Machine; i.e., that value we actually need for making or taking the right decision.

Area of Application

Actually, it is broad. In this era, we can put Machine Learning everywhere like Daily life, Medical Science, Image Processing, Driverless car, Face recognition, etc. etc.