Xiantao Zhao is a PhD researcher in I-Form, based in Trinity College Dublin. He received dual B.Sc. degrees from Beijing Normal University (CN) and Colorado State University (US) majoring in Computer Science in June 2021. Now he is pursing Ph.D. degree in School of Computer Science and Statistics, Trinity College Dublin. His main research areas are data mining, functional data clustering and machine learning.
Research Interests (Lay Summary)
Xiantao started his PhD in 2021 and is working on functional data clustering in additive manufacturing data streams. Clustering is usually used as a preliminary step in data exploration, and its goal is to display the data so as to highlight various characteristics and study important sources of patterns and variation among the data. Instead of traditional data clustering, Xiantao’s focus is on the functional data, which is multivariate data with an ordering on the dimensions. This type of data consists of curves varying over a continuum, such as time, frequency, or wavelength. Xiantao’s goal is to find a general method or a targeted method for clustering such data.
Technical Summary
Functional data clustering often has difficulties such as large data dimensions, lager data size, and uncertain cluster centres. Therefore, traditional clustering methods are difficult to apply to functional data. In order to perform functional data clustering , the existing three mainstream methods are dimensionality reduction, the use of non-parametric methods and the use of special clustering model. The goal of this project is to find a general method or a targeted method for clustering functional data, using data from additive manufacturing as an example
Expertise
Artificial Intelligence, Data Analytics, Multi-variate Time Series Data Analysis, Real-time Data Analytics