2015 Fiscal Year Final Research Report
Data Mining Methods for Discovering Statistically Significant Substructures of Graphs
| Project/Area Number |
26880013
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2014-08-29 – 2016-03-31
|
| Keywords | グラフマイニング / パターンマイニング / 仮説検定 / 多重検定 / 統計的有意性 |
|---|
| Outline of Final Research Achievements |
We have developed significant subgraph mining methods, which find substructures of graphs that are statistically significantly enriched in a class of transactions. In drug discovery, for example, the methods enable us to find substructures of chemical compounds which are significantly associated with a particular activity such as an anticancer activity. Our methods control the false positive rate, the probability of erroneous discoveries that actually do not have a particular effect, which is essential to provide reliable results in a number of application domains from drug discovery to social network analysis.
|
| Free Research Field |
機械学習・データマイニング
|
