時間：113年10月28日(星期一) 19:00-21:00

地點： 大仁樓301

主持人：張家銘老師

演講者 :   國立台灣大學  ,   蔡政安教授

演講題目： Clustering Analysis in Data Mining (資料探勘工具-分群分析)

演講摘要： Clustering is a technique used in unsupervised learning to group data points into clusters that share similar features. These features are commonly observed in continuous and categorical (ordinal and nominal) mixed data types. Conventional clustering algorithms may struggle to handle mixed-type data effectively because they often rely on distance metrics that are not suitable for all variable types or fail to capture the correlation structure between different types of variables. This talk will address these challenges and propose a distance-based approach to tailored for mixed-type data. Simulation experiments and real data analyses are used to evaluate the performance of our proposed method. Results show that our method can help address the challenges associated with capturing the correlation structure and effectively clustering observations with diverse feature types.