论文信息 - An Agent-Based Algorithm for Data Reduction

An Agent-Based Algorithm for Data Reduction

The paper proposes an agent-based algorithm for data reduction, implemented using JABAT (JADE Based A-Team) environment designed for solving a variety of computationally hard optimization problems. The approach aims at reducing the original dataset in two dimensions including selection of reference instances and removal of irrelevant attributes. Several agents representing different local-search based strategies are employed in parallel to achieve a synergetic effect. The paper includes also computational experiment results.

Piotr Jedrzejowicz | Ireneusz Czarnowski

[1] Jacob Zahavi,et al. Using simulated annealing to optimize the feature selection problem in marketing applications , 2006, Eur. J. Oper. Res..

[2] Miroslav Kubat,et al. Selecting representative examples and attributes by a genetic algorithm , 2003, Intell. Data Anal..

[3] Frans Coenen,et al. Research and Development in Intelligent Systems XX , 2004, Springer London.

[4] Tony R. Martinez,et al. Reduction Techniques for Instance-Based Learning Algorithms , 2000, Machine Learning.

[5] Francisco Herrera,et al. On the combination of evolutionary algorithms and stratified strategies for training set selection in data mining , 2006, Appl. Soft Comput..

[6] Piotr Jedrzejowicz,et al. JADE-Based A-Team Environment , 2006, International Conference on Computational Science.

[7] Piotr Jedrzejowicz,et al. An Approach to Instance Reduction in Supervised Learning , 2003, SGAI Conf..

[8] Huan Liu,et al. Feature Selection for Classification , 1997, Intell. Data Anal..

[9] Christopher J. Merz,et al. UCI Repository of Machine Learning Databases , 1996 .