Feature Discovery in Temporal Data(Artificial Intelligence III)
スポンサーリンク
概要
- 論文の詳細を見る
In mining time series data, the graph similarity of the data can be used as an effective tool. However, when the time series has missing data, the utility of graph similarity in analyzing time series data is limited. The present study investigates experimentally the impact of missing values in time series data on dynamic time warping, a method that is commonly used in determining graph similarity. Based on the results of the investigation, we propose a new method by which to treat time series data having missing values. The proposed method uses point similarity rather than graph similarity. Experiments were conducted in order to evaluate the performance of the proposed method, and the results indicate that the proposed method is effective for finding features in time series data.
- 一般社団法人情報処理学会の論文
- 2004-12-04
著者
-
Numao Masayuki
The Institute Of Scientific And Industrial Research Osaka University
-
ICHISE RYUTARO
Intelligent Systems Research Division, National Institute of Informatics
-
Ichise Ryutaro
Intelligent Systems Research Division National Institute Of Informatics
関連論文
- SBSOM : Self-Organizing Map for Visualizing Structure in the Time Series of Hot Topics(Text Mining I)
- SBSOM : Self-Organizing Map for Visualizing Structure in the Time Series of Hot Topics(Text Mining I)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)
- Visualization of Damage Progress in Solid Oxide Fuel Cells
- Inductive Logic Programming for Multiple-Part Data : Applications on Structure-Activity Relationship Studies
- Partial Rule Weighting Using Single-Layer Perceptron(Artificial Intelligence III)
- Multiple-Instance Learning Based Heuristics for Mining Chemical Compound Structure(Scientific Data Mining)
- Partial Rule Weighting Using Single-Layer Perceptron(Artificial Intelligence III)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)
- Multiple-Instance Learning Based Heuristics for Mining Chemical Compound Structure(Scientific Data Mining)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)
- A Category-based Framework of a Self-improving Instructional Planner
- Feature Discovery in Temporal Data(Artificial Intelligence III)
- Preprocessing Planning for Data Mining(Artificial Intelligence I)
- Preprocessing Planning for Data Mining(Artificial Intelligence I)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)
- Feature Discovery in Temporal Data(Artificial Intelligence III)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)
- A Category-based Framework of a Self-improving Instructional Planner