|
Mining methods for biomedical data and signals classification and clustering| old_uid | 12286 |
|---|
| title | Mining methods for biomedical data and signals classification and clustering |
|---|
| start_date | 2013/04/04 |
|---|
| schedule | 10h30-12h |
|---|
| online | no |
|---|
| location_info | salle de réunion de l’UFR |
|---|
| summary | In this presentation we will discuss some classification and clustering methods and their use for biomedical data and signals analysis. In a first part, the main concepts of knowledge discovery from data process will be introduced. Then we will focus on one particular step of this process: the Data Mining. A commonly adopted definition of Data Mining is: “The non-trivial process of identifying valid, novel, potentially useful and ultimately understandable patterns in data” [O. Fayyad et al., 1996]. Following an overview of state of the art data mining principles, we will focus on two main methods: classification and clustering. The main algorithms used in these two methods will be presented and illustrated; extension of these algorithms for uncertain data will also be addressed using fuzzy logic concepts. Finally, some recent applications on biomedical data and signals fields of study of these two methods will be presented. |
|---|
| responsibles | Hoffmann, Marin |
|---|
| |
|