2012 Fiscal Year Final Research Report
Study on Organizing a Knowledge-Creating Database and Achieving Efficient Processing
| Project/Area Number |
20500137
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Hiroshima City University |
|---|
Principal Investigator |
KITAKAMI Hajime 広島市立大学, 情報科学研究科, 教授 (50234240)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TAMURA Keiichi 広島市立大学, 情報科学研究科, 准教授 (80347616)
KUROKI Susumu 広島市立大学, 情報科学研究科, 准教授 (20225288)
MORI Yasuma 広島市立大学, 情報科学研究科, 助教 (60264959)
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
KUROKI Susumu 広島市立大学, 情報科学研究科, 准教授 (20225288)
MORI Yasuma 広島市立大学, 情報科学研究科, 助教 (60264959)
|
|---|
| Project Period (FY) |
2008 – 2012
|
| Keywords | テキストマイニング / 配列マイニング / 索引構造 / 正規表現 / 並列処理 |
|---|
| Research Abstract |
In the research related to inductive mechanisms for knowledge-creating databases, the problem of extracting regularity from a mismatch cluster that was defined as a set of similar subsequences returned by either Gibbs sampling or approximate query processing has been solved, where the regularity was represented as a set of minimum generalized patterns with a regular expression. Moreover, parallelization has been achieved for extracting the regularity from the mismatch cluster using Multicore PC clusters. In the authors’ research about deductive mechanisms for knowledge creating-databases, the problems have been solved to increase accuracy of approximate query processing in coordinate sequence databases. In the authors’ research related to abduction mechanisms for knowledge-creating database, various communications were considered.
|
