## The IMI Colloquium Report in July 10, 2024

2024.08.21

・Title: Statistical Distances with Mathematical Explanation

Speaker: Prof. GIM Minjung

National Institute for Mathematical Sciences/

Ajou University Department of Mathematics, Korea

・Title: Statistical Learning Models in Functional structure of Clinical data

Speaker: Prof. Soon-Sun Kwon

Ajou University Department of Mathematics, Korea

・Place: IMI Auditorium(W1-D-413) and Live streaming with zoom

Attendance: 50 (Students: 40; Staff: 10)

At the colloquium in July, we had two speakers, Prof. GIM from the National

Institute of Mathematical Sciences, Korea, and Prof. Kwon from Ajou University.

Prof. GIM introduced statistical distance, which is used in various machine

learning applications to measure the distribution of data and the differences

among samples. He also introduced several commonly used statistical distance

functions implemented in SciPy, and explained their operating principles and

properties with visual explanations. In addition, he showed examples of

application to actual examples. This enabled us to understand the

usefulness of each statistical distance introduced.

Prof. Kwon first introduced a case study in which he applied his proposed

clustering method using Frechet distance for multidimensional functional data to

multidimensional thyroid cancer data collected in Korea. Then, he introduced the

multivariate analysis method to investigate the relationship between

kinematic gait measurements and GMFCS levels based on gait data from both normal

and cerebral palsy (CP) patient groups with various Gross Motor Function

Classification System (GMFCS) levels. He also showed that the proposed method

enables prediction with good accuracy and provides discriminant functions with

easy interpretation.