Degree Type

Creative Component

Semester of Graduation

Fall 2018

Department

Electrical and Computer Engineering

First Major Professor

Daji Qiao

Degree(s)

Master of Science (MS)

Major(s)

Computer Engineering

Abstract

Random forest is an important technique in modern machine learning. It was first proposed in 1995, and with the increasing of the compute power, it became a popular method in this decade. In this paper, we present the background knowledge of random forest and also have comparison with other machine learning methods. We also have some discussion of data preprocessing and feature extraction. Then we study the application of random forest to a real world problem - an authentication system which for continuous user authentication approach for web services based on the user’s behavior. The dataset we used for evaluate is given by Workiva company, and the result demonstrates that random forest is eligible for the system and achieve a fair classification accuracy

Copyright Owner

Tingwei Chang

File Format

application/pdf

Share

COinS