supervised learning|full knowledge

You’re here to learn about the wonderful world of supervised learning arguably the most important branch of machine learning but what is machine learning ?

What is machine learning?

Machine Learning

machine learning is the art of science of giving computers the ability to learn to make decisions from data without being explicitly programmed for example your computer can learn to predict whether an email of spam or not spam given content and sender.

the other example of your computer can learn to cluster say Wikipedia entries into different categories based on the words they contain it could then assign any new Wikipedia article to one of the existing clusters. That is machine learning.

What is supervised learning?

Supervised Learning

Supervised learning provides you with a powerful tool for classifying and processing data using machine language. With supervised learning you use labeled data, which is a data set that is classified, to approximate a learning algorithm. The data set is used as a basis for predicting the classification of other non listed data through the use of machine learning algorithms.

Supervised learning is a learning model designed to predict, giving an unexpected input example. A supervised learning algorithm takes the input data set and a known set of known responses to the data (output) to find out the regression / classification model. A learning algorithm then trains a model to generate a prediction for the response of new data or test data sets. it uses different classification method. The algorithms include Linear Regression, Logistic Regression, and Neural Networks as well as Decision Tree, Support Vector Machine (SVM), Random Forest, Bhole Bayes and K-Nearest Neighbor.

Types of Supervised learning

They are mostly two types


It is a technique that suppose to reproduce the output value. We can use it, for example, to predict the price of a product, such as the price of a house or stock of a specific city. If we want, there are many things that we can predict.


It is a technique that suppose to reproduce class assignments. It can estimate the response value and the data is separated into “classes”. Examples? Identifying a type of car in a photo, whether it is mail spam or a friend’s message, or what the weather will be like today.

How supervised learning works?

For example, you want to help to train a machine so that you can estimate how long it will take you to drive home from your workplace. Here, you begin by creating a set of labeled data. This data includes

Weather condition
Time of day
All these details are your input. Output is the amount of time it takes to return home on that specific day.

You know instinctively that if it is raining outside, it will take you longer to drive home. But the machine needs data and figures.

Let us now see how you can develop a supervised learning model of this example that helps the user to determine commute time. The first thing you need to build is a training set. This training set will include factors such as total commute time and weather, time etc. Based on this training set, your machine can see that there is a direct relationship between the time it takes to rain and the time it takes to get home.

So, it turns out that the more rain, the longer you will be driving to get back to your home. It can also look at the relationship between the time you leave work and the time you get on the road.

Around 10 in the evening. Now it will take you longer to reach home. Your machine may find some relationships with your labeled data.


  1. You can get very specific about the definition of classes, which means that you can train classifiers in a way that has a correct decision range to accurately differentiate different classes.
  2. You can specifically determine how many classes you have to take.
  3. After training, you do not need to keep training examples in memory. You can place the decision range as a mathematical formula and it will be sufficient to classify future inputs.
  4. It is a simple process to understand learning. In the case of untrained learning, we do not easily understand what is happening inside the machine, how it is learning, etc.
  5. After the entire training is complete, you do not need to keep training data in your memory. Instead, you can place the decision limit as a mathematical formula.


  1. If there are no examples in your training set that you want to keep in the classroom, the decision range can be extended.
  2. While training the classifier, you need to select a lot of good examples from each class.
  3. Classifying big data can be a real challenge.Supervised learning requires a lot of computation time for training.
  4. In the case of classification, if we give an input that is not from any class in the training data, the output may be labeled incorrectly. For example, suppose you have trained an image classification with data from cats and dogs. Then if you give an image of a giraffe, the output can be a cat or a dog, which is not correct.
  5. Typically, training requires too much computation time, so perform classification, especially if the data set is very large. It will also test the efficiency of your machine and your patience.

Leave a Reply

1 Comment threads
1 Thread replies
Most reacted comment
Hottest comment thread
2 Comment authors
Atul YadavSuraj Recent comment authors
newest oldest most voted
Notify of

Good article for beginner’s.