Learning Latent Tree Models
主 题: Learning Latent Tree Models
报告人: Nevin L. Zhang (The Hong Kong University of Science & Technology)
时 间: 2007-05-30 上午 9:00 - 10:30
地 点: 理科一号楼 1418
Traditional Chinese medicine (TCM) is an important avenue for disease prevention and treatment for the Chinese people and is gaining popularity among others. However, many remain skeptical and even critical of TCM because a number of its shortcomings. One key shortcoming is the lack of a solid foundation. We endeavor to alleviate this shortcoming and use machine learning techniques to establish a statistical foundation for TCM.
When viewed as a black box, TCM diagnosis is simply a classifier that classifies patients into different classes based on their symptoms and signs. A fundamental question is: Do those classes exist in reality? To seek an answer from the machine learning perspective, one would naturally use cluster analysis. Previous clustering methods are unable to handle the complexity of TCM. We have therefore developed a new clustering method in the form of latent tree models.
In this talk, we provide a concise summary of research work on latent tree models, discuss a learning algorithm, and present a case study to show how latent tree models can help in establishing a statistical foundation for TCM.