论文信息 - First-Order Rule Mining by Using Graphs Created from Temporal Medical Data

First-Order Rule Mining by Using Graphs Created from Temporal Medical Data

In managing medical data, handling time-series data, which contain irregularities, presents the greatest difficulty. In the present paper, we propose a first-order rule discovery method for handling such data. The present method is an attempt to use graph structure to represent time-series data and reduce the graph using specified rules for inducing hypothesis. In order to evaluate the proposed method, we conducted experiments using real-world medical data.

Masayuki Numao | Ryutaro Ichise

[1] Masayuki Numao,et al. Learning First-Order Rules to Handle Medical Data , 2001 .

[2] Michael Spenke,et al. Visualization and interactive analysis of blood parameters with InfoZoom , 2001, Artif. Intell. Medicine.

[3] J. Ross Quinlan,et al. Learning logical definitions from relations , 1990, Machine Learning.

[4] Eamonn J. Keogh,et al. Scaling up dynamic time warping for datamining applications , 2000, KDD '00.

[5] Henrik Boström,et al. Learning First Order Logic Time Series Classifiers: Rules and Boosting , 2000, PKDD.

[6] Pieter Adriaans,et al. Data mining , 1996 .

[7] Nada Lavrac,et al. Active subgroup mining: a case study in coronary heart disease risk group detection , 2003, Artif. Intell. Medicine.

[8] Stephen Muggleton,et al. Relational Rule Induction with CProgol4.4: A Tutorial Introduction , 2001 .

[9] Heikki Mannila,et al. Rule Discovery from Time Series , 1998, KDD.

[10] 元田浩,et al. Active mining : new directions of data mining , 2002 .

[11] Shusaku Tsumoto,et al. Rule Discovery in Large Time-Series Medical Databases , 1999, PKDD.

[12] Igor Kononenko,et al. Machine learning for medical diagnosis: history, state of the art and perspective , 2001, Artif. Intell. Medicine.

[13] Hongxing He,et al. Feature Selection for Temporal Health Records , 2001, PAKDD.