From the course: Machine Learning in Telecommunication: From Basics to Real-World Cases
Join today to access over 24,800 courses taught by industry experts.
Decision tree
From the course: Machine Learning in Telecommunication: From Basics to Real-World Cases
Decision tree
(soothing music) - [Instructor] A decision tree is a supervised learning algorithm that works like a flow chart, and it splits the data into the branches based on certain feature values so as to make a right prediction. It is widely used for both classification as well as for regression tasks and help us make data-driven predictions accurately. For example, it can answer certain questions like, is the network congested or not? By checking certain conditions such as utilization is greater than certain threshold, say 70%, or maybe throughput is less than 10 Mbps. And then each decision leads to a final outcome that either the network is congested or not. There is one important term which is widely used in decision tree, which is variance and variance measures how data is spread from their average value. So low variance means data points cluster closely around the mean value, whereas high variance indicates they're widely scattered. Imagine two datasets where the age of people in a city,…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.