| Project/Area Number |
16300043
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Japan Advanced Institute of Science and Technology |
|---|
Principal Investigator |
TU Baoho Japan Advanced Institute of Science and Technology, School of Knowledge Science, Professor, 知識科学研究科, 教授 (60301199)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TAKABAYASHI Katsuhiko Chiba University Hospital, Professor, 医学部附属病院, 教授 (90188079)
KAWASAKI Saori JapanAdvanced Institute of Science and Technology, School of Knowledge Science, Associate, 知識科学研究科, 助手 (40377437)
TUAN Nam Tran (TRAN Nam Tuan) 北陸先端科学技術大学院大学, 知識科学研究科, 助手 (60362018)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥9,600,000 (Direct Cost: ¥9,600,000)
Fiscal Year 2006: ¥2,200,000 (Direct Cost: ¥2,200,000)
Fiscal Year 2005: ¥2,900,000 (Direct Cost: ¥2,900,000)
Fiscal Year 2004: ¥4,500,000 (Direct Cost: ¥4,500,000)
|
|---|
| Keywords | HEPATITIS / DATA MINING / MULTIPLE RESOURCES / TEMPORAL ABSTRACTION / MEDICAL LITERATURE / EXPERT EVALUATIONS / INTEGRATED APPROACH / hepatitis / temporal abstraction / text mining / expert evaluation / 時区間関係 / 医療データマイニング / 医療文献マイニング / ウィルス性肝炎 / 専門知識 / 背景知識獲得 |
|---|
| Research Abstract |
1. Developed a novel method of temporal abstraction abased on temporal logic that exploits temporal relations. This is a new method for analyzing medical temporal data by extracting all kinds of possible temporal relations existing in the data, and discovering rules using the relations to solve the three hepatitis problems.
2. Developed text mining methods to exploit huge sources of textual data (using advanced machine learning techniques such as conditional random fields, tolerance rough set model, etc.) and apply them to MEDLINE database to find knowledge for two purposes: (a) finding back ground knowledge to narrow the search space in the learning phase and (b) finding knowledge to interpret discovered rules by data mining methods.
3. Developed in integrated approach to medical data mining that based on mutually combination of learning methods, medical literature and expert inspection. The approach showed its efficiency is much higher than just using each method.
4. We succeeded in some other research issues such as fundamental research (on kernel methods, visualization, rule induction, optimization), bioinformatics (protein interaction). These results allow us to continue study medical data with advanced learning methods and the relations between clinical data and disease genes.
|
