WEKO3
アイテム
Development Of A Dynamic Sensor Data Model With Contexts For Data Mining From Monitoring Of Infrastructures
http://hdl.handle.net/10173/1829
http://hdl.handle.net/10173/18294e3f672c-bcb9-4e25-a2a0-2c289fcc22f8
名前 / ファイル | ライセンス | アクション |
---|---|---|
SMS10-161.pdf (456.9 kB)
|
|
Item type | 会議発表論文 / Conference Paper(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2018-02-06 | |||||
タイトル | ||||||
タイトル | Development Of A Dynamic Sensor Data Model With Contexts For Data Mining From Monitoring Of Infrastructures | |||||
言語 | ||||||
言語 | eng | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | data mining | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | sensor monitoring of infrastructures | |||||
資源タイプ | ||||||
資源タイプ識別子 | http://purl.org/coar/resource_type/c_5794 | |||||
資源タイプ | conference paper | |||||
著者 |
Yoshida, Yoshihiro
× Yoshida, Yoshihiro× Yabuki, Nobuyoshi |
|||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | Sensor monitoring is more and more important for maintenance of infrastructures and for prevention of disasters. Micro Electro-Mechanical Systems (MEMS) technology enables the cost of sensors to decrease rapidly, and wireless sensor networks can reduce the cost of cables significantly. Thus, more and more sensors are expected to be installed in various places for monitoring in the future. In order to find meaningful information and knowledge from a large amount of sensor data, data mining has attracted considerable attention. However, simple application of the data mining technique to sensor data may not be successful compared to our expectation from our experience. As sensors are installed for infrastructures, in order to discover meaningful knowledge, contextual information, which is the situation data of each sensor, would be necessary. Currently, sensor data models such as SensorML and building product models such as Industry Foundation Classes (IFC) have been being developed without any interaction. In addition, input data for most data mining algorithms are formulated as tables, which are different from data implemented in accordance with product and sensor data models. Therefore, building and sensor data must be converted into a table format for data mining. In this research, in order to discover meaningful information and knowledge from a large amount of sensor data for buildings, we developed a dynamic data model which employs building information as contextual data of sensors. Then, we applied experimental data to the data model for evaluation and validation. Finally, data mining was executed using data stored in a database. As a result, it was verified that the developed data model can represent the contextual data of sensors with flexible table structure and can be useful for discovering new knowledge by data mining. | |||||
書誌情報 |
Society for Social Management Systems Internet Journal 巻 6, 号 1, 発行日 2010-03 |
|||||
ISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 2432-552X | |||||
著者版フラグ | ||||||
出版タイプ | VoR | |||||
出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||
出版者 | ||||||
出版者 | Society for Social Management Systems |