2021 Fiscal Year Research-status Report
Exploring Novel Mathematical Models and Efficient Algorithms to Discover Periodic Spatial Patterns in Irregular Spatiotemporal Big Data
| Project/Area Number |
21K12034
|
|---|
| Research Institution | The University of Aizu |
|---|
Principal Investigator |
Rage Uday・Kiran 会津大学, コンピュータ理工学部, 准教授 (20874324)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
是津 耕司 国立研究開発法人情報通信研究機構, ユニバーサルコミュニケーション研究所統合ビッグデータ研究センター, 研究センター長 (40415857)
|
|---|
| Project Period (FY) |
2021-04-01 – 2025-03-31
|
| Keywords | Big data |
|---|
| Outline of Annual Research Achievements |
This year, we have first developed an ER-model to store the air pollution data. Next, we have developed a theoritical database model by applying normal forms. Later, we have populated our theoritical model using postGres database. The air pollution data used for populating our model has been taken from the Atmospheric Environmental Regional Observation System (http://soramame.taiki.go.jp/). Next, proposed a mathetical model for representing the air pollution data as a geo-referenced time series database. The work is going to appear in IEEE FUZZ 2022.
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
The reason is progressing smoothly. We were able to collect the data from the real-world.
|
| Strategy for Future Research Activity |
The space of items in a database gives raise to an itemset lattice. This lattice represents the search space for finding interesting patterns within the data. Thus, the size of the search space is (2 power n)-1, where n represents the total number of items in a database. In this research, we plan to explore novel pruning techniques to effectively reduce the search space. In particular, we plan to investigate upper bound measures to reduce the search space.
|
| Causes of Carryover |
We have transferred the balance to the Co-PI of this project, Koji Zettsu
|
