2022-4: Homogenising data density distribution for better pattern recognition
Date: Thursday 21 April
Time: 2.00pm – 3.00pm
Speaker: Dr Ye Zhu
Abstract: Distance-based and density-based algorithms have been widely applied in various industries for clustering and anomaly detection. However, these algorithms usually suffer from the long-standing issue of inhomogeneous cluster densities, since they implicitly assume that all clusters have approximately the same density. Many remedies have been suggested but do not address the issue satisfactorily. In this talk, Dr Zhu will present his latest work on a method of homogenising data density. This method is based on local data density estimation and achieved by using a new multi-dimensional Cumulative Distribution Function. He will show that simply transforming the data as a pre-processing can overcome key shortcomings of distance-based and density-based algorithms.
Bio: Dr Ye Zhu is a lecturer in IT at the School of Information Technology at Deakin University. He is also the Data to Intelligence (D2i) Research Centre HDR coordinator and Master of Data Science Course CPL Officer. He received a PhD degree in Artificial Intelligence with a Mollie Holman Medal for the best doctoral thesis of the year from Monash University in 2017. Dr Zhu joined Deakin University as a post-doc research fellow in complex system data analytics in July 2017 and then became a lecturer in Feb 2019. His research works focus on clustering analysis, anomaly detection, similarity learning and their applications for pattern recognition. He has secured 5 research grants of around AUD$150,000 in total for interdisciplinary and industrial research. Dr Zhu has published over 30 papers in top-tier conferences and journals, including SIGKDD, PAKDD, AAAI, AIJ, ISJ, PRJ, JAIR and MLJ. He has served as Program Chair and Program Committee for many top international conferences, such as SIGKDD, AAAI and IJCAI. He obtained both an Early Career Researcher Award and a Teaching and Learning Award at the School of IT in the last two years.