Learning is any process in which a system improves its performance from experience, but in general it is any activity through which a human, animal, or machine acquires new understanding, knowledge, behaviors, skills, values, attitudes, and preferences.
In Machines particularly, learning is of three types:
Here we are going to discuss further about Supervised Learning, as the name indicates it is a sort of learning but under a supervisor (Teacher). Basically, in supervised learning a machine is trained with a well labeled data, i.e., some of the data already contains the correct answer. After this, a new data is given to the machine and supervised learning algorithm analyzes the training data and generates the correct outcome using labeled data.
For example: a patient comes to check if he has a certain disease or not, then the algorithm refers to the training data to verify the description of the patient and then using labeled data containing the characteristics of all diseases and the name of the disease the algorithm determines whether the patient has a certain disease or not. In this example the Input and Output are:
- I/P: The description of the patient who comes to the clinic
- O/P: Whether the patient has a certain disease or not.
For a test data for Triangle, since the machine has already learned the things from previous data and this time must use it wisely. It will first classify the shape and color and would confirm the shape name as Triangle and put it in the Triangle category. Thus, the machine learns the things from training data and then applies the knowledge to test data (new Shape).
Since Supervised learning algorithms are trained using labeled data it is a simpler and highly accurate method.
Categories of Supervised Learning Algorithms
- Classification: A Classification Algorithm is used when the output variable is categorical. E.g.: “Red” or “blue”, “disease” or “no disease”
- Naive Bayes Classifier
- K-NN(K nearest neighbors)
- Decision Tree
- Support vector machine
- Regression: A Regression Algorithm is used when the output variable is continuous. E.g.: “Price”, “Weight”
- Linear Regression
- Multiple Regression
Advantages:
- Supervised learning allows collecting data and produces data output from previous experiences.
- Helps to optimize performance criteria with the help of experience.
- Supervised machine learning helps to solve various types of real-world computation problems.
Dis-Advantages:
- Classifying big data can be challenging.
- Training for supervised learning needs a lot of computation time. So, it requires a lot of time.