5. Machine Learning
● In Algorithmic Problem Solving, we use Algorithms to solve problems
● Example
○ Sorting
○ Graph Search.
■ Depth First Search
■ Breadth First Search
● In Machine Learning, we still use Algorithms
○ But traditional algorithms are not sufficient to solve the problem.
● Example - Real time Traffic Prediction.
● Example - Real time food order delivery
● Example - Real time Taxi Cab Calling
■ We need algorithms that are constantly changing without much
human interaction.
● In other words, Machine Learning
6. Supervised Learning
● Supervised learning is used whenever we want to predict a certain
outcome from a given input.
● We have examples of input/output pairs.
● Like a Supervisor / Teacher.
○ Input and Output Examples are given to Machine to Study
○ Machine Studies Input and Output Examples with the help of
algorithms taught by Supervisor.
● Supervisor gives Tests
○ Tests help the Machine to know if it can answer questions correctly
○ If machine gets a test wrong, it studies again.
○ If machine gets test right, it takes harder questions
Using Supervised Learning - Machine creates a
Model to answer unseen problems !
7. Unsupervised Learning
● No Supervision.
● Learning algorithm is just shown the input data and asked to extract
knowledge from this data.
● For certain kinds of problems, data helps automatically.
○ Like differentiate between Land and Water
23. Student Performance
● With this data set,
student performance
is predicted for new
student by the
computer.
24. References
● Introduction to Machine Learning with Python by Sarah Guido;
Andreas C. Müller Published by O'Reilly Media, Inc., 2016
● https://towardsdatascience.com/a-brief-introduction-to-supervised-lear
ning-54a3e3932590
● Senthil Kumaran Georgia Tech Machine Learning Course Report.