Minakshi Kaushik, the PhD student of the Department of Software Science, will defend her PhD thesis "Generalized Association Rule Mining – Dimensional Unsupervised Learning" on February 15, 2024, starting at 14:00. The defense will take place in room ICT-315 (Akadeemia tee 15a, ICT building of TalTech) and can be also followed via Zoom.
PhD thesis "Generalized Association Rule Mining: Dimensional Unsupervised Learning" propose a novel unsupervised learning technique for generalized association rule mining. The thesis focuses on developing novel measures for discretizing numerical attributes using order-preserving partitioning method.
The thesis explores the integration of numerical association rule mining and order-preserving partitioning methods to identify partitions of numerical attributes that highlight the substantial impact of an independent numerical attribute on a dependent one.
This research presents four substantial contributions to the domains of ARM, QARM, or NARM. These contributions are outcomes of three main research questions and seven sub-research questions answered in the thesis. The thesis follows design science research methodology to create innovative artifacts and methods, providing new insights to widen understanding of the domain under the research.
The first contribution is an in-depth analysis of existing research articles in the field of NARM, offering a comprehensive overview of the existing literature. The second contribution is an explanation of the need and importance of human perception in partitioning numerical attributes. The third contribution is the introduction of two novel measures designed for partitioning numerical attributes. The fourth contribution is an analytical evaluation of the introduced measures in contrast to the outcomes of human perception. The author argues that the impact of this research resonates across decision support systems, data analytics, and the broader landscape of Machine Learning.
The thesis is published in the Digital Collection of TalTech Library.
Supervisor: prof. Dirk Draheim, TalTech.
Oponents:
- Prof. Gillian Dobbie, University of Auckland, New Zealand;
- Prof. Dr. A Min Tjoa, Vienna University of Technology, Austria.
Meeting ID: 930 8658 9932
Passcode: 855917
Before the defence, starting at 10:00 a.m. Prof Gillian Dobbie will give a presentation on "Navigating the Social and Ethical Responsibilities of Computing" and at 11:00 a.m. Prof. Dr. A Min Tjoa will give a presentation on "Digital humanism as an enabler to address the urgent need for a holistic socio-technical approach to the latest developments in computer science and artificial intelligence". The presentations will take place in room ICT-A1, You are welcome to listen!