Classification problems occur in many fields including statistics, data science, machine learning and medicine. The purpose of a classification algorithm is to assign a class label, or the probability of being in a particular class, to an unclassified example. Decision tree classifiers are conceptually simple, making them a popular statistical learning method. This project considers using oblique decision trees in ensemble learning, where the ensemble is built using techniques from random forest and boosting. The proposed methods will be numerically tested on a wide range of datasets.
Supervisors
Supervisor: Blair Robertson
Does the project come with funding
No
Final date for receiving applications
Ongoing
Keywords
Algorithms; statistical learning; data science
