Loading...
Please wait, while we are loading the content...
Similar Documents
Probabilistic graphical models for climate data analysis
| Content Provider | ACM Digital Library |
|---|---|
| Author | Banerjee, Arindam |
| Abstract | The prominence and usage of probabilistic graphical models for data analysis have increased substantially over the past decade. Unlike traditional models in statistical machine learning, graphical models capture statistical dependencies between variables making them suitable for many problems. In this talk, I will discuss two applications of graphical models to climate data analysis problems, including progress and open questions. The first application is on abrupt change detection, with emphasis on detecting significant droughts in the past century. The second application is on predictive modeling of land variables based on ocean variables. Building on the second application, a framework for constructing statistical climate networks will be presented. Key challenges from both the computational and climate science perspective in realizing the potential of these methods will also be discussed. |
| Starting Page | 3 |
| Ending Page | 3 |
| Page Count | 1 |
| File Format | |
| ISBN | 9781450311816 |
| DOI | 10.1145/2110230.2110235 |
| Language | English |
| Publisher | Association for Computing Machinery (ACM) |
| Publisher Date | 2011-11-13 |
| Publisher Place | New York |
| Access Restriction | Subscribed |
| Subject Keyword | Climate data |
| Content Type | Text |
| Resource Type | Article |
