Abnormal user detection using rule-based machine learning

• Main Author: Zhang, Jiawei
• Other Authors: Wang Lipo
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2019
• Subjects: Engineering; Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/136494
• Institution: Nanyang Technological University

In this dissertation, we implement a rule-based machine learning abnormal behavior detection system on a real-world database by using association rule learning method. The objective of our work is to detect the abnormal behaviors from a new database using rule-based machine learning techniques and evaluate its performance by comparing it with other machine learning methods. One-day behaviors pattern feature dataset would be derived from the real-world competition raw data. Apriori and FP-growth algorithm would be used to find the association rules in order to figure out the abnormal user. The experiment part verified that the system could precisely identify the abnormal user. It has similar performance compared with other machine learning method but is more intelligible than other methods since it could use {IF-THEN} expression to introduce the prediction process.