论文信息 - Topics in Artificial Intelligence

Topics in Artificial Intelligence

In this paper we present CAT-CBR a component-based platform for developing CBR systems. CAT-CBR uses UPML (Universal Problem-solving Methods Language) for specifying CBR components. A collection of CBR components for retrieval of propositional cases is presented in detail. The CAT-CBR platform guides the engineer using a case-based recommendations system to develop a configuration of components that satisfies the requirements of a CBR system application. We also present how to develop a runtime CBR application from the configuration resultant of the configuring process.

Elisabet Golobardes | M. Teresa Escrig | Francisco Toledo

[1] Rina Azoulay-Schwartz,et al. Bidding Mechanisms for Data Allocation in Multi-Agent Environments , 1997, ATAL.

[2] R. Olshen,et al. Almost surely consistent nonparametric regression from recursive partitioning schemes , 1984 .

[3] L. Cooper,et al. When Networks Disagree: Ensemble Methods for Hybrid Neural Networks , 1992 .

[4] Santiago Ontañón,et al. Learning When to Collaborate among Learning Agents , 2001, ECML.

[5] Barry Smyth,et al. Modelling the Competence of Case-Bases , 1998, EWCBR.

[6] Barry Smyth,et al. Remembering To Forget: A Competence-Preserving Case Deletion Policy for Case-Based Reasoning Systems , 1995, IJCAI.

[7] Yoav Freund,et al. Experiments with a New Boosting Algorithm , 1996, ICML.

[8] Lars Kai Hansen,et al. Neural Network Ensembles , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[9] Jerome H. Friedman,et al. On Bias, Variance, 0/1—Loss, and the Curse-of-Dimensionality , 2004, Data Mining and Knowledge Discovery.

[10] Zoran Obradovic,et al. Classification on Data with Biased Class Distribution , 2001, ECML.

[11] Ron Kohavi,et al. Bias Plus Variance Decomposition for Zero-One Loss Functions , 1996, ICML.