Presentation information

Interactive Session

[4Rin1] Interactive 2

Fri. Jun 12, 2020 9:00 AM - 10:40 AM Room R01 (jsai2020online-2-33)

[4Rin1-37] Clustering method for data with mixed parametric structures

〇Kohei Tamura1, Koutarou Tamura2,3 (1.Aioi Nissay Dowa Insurance Co., Ltd., 2.NRI digital, Ltd., 3.Nomura Research Institute, Ltd.)

Keywords:Machine Learning, Clustering, Survival Analysis

Regarding the structural complexity of data, stratification of the data using a clustering method is important for constructing a model based on more detailed characteristics of the data. In particular, the data such as product failures shows there are various processes causing failures. This implies it is necessary to stratify the data for more precise analysis. However, these stratification factors are rarely acquired as data, and it is often difficult to grasp them. In this research, we propose a clustering method for extracting data structures that follow a Weibull distribution event occurrence data in a certain market product.

Authentication for paper PDF access

A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.