Instance-Based Learning Algorithms
Machine Learning
C4.5: programs for machine learning
C4.5: programs for machine learning
Advances in instance-based learning algorithms
Advances in instance-based learning algorithms
SOAP: Efficient Feature Selection of Numeric Attributes
IBERAMIA 2002 Proceedings of the 8th Ibero-American Conference on AI: Advances in Artificial Intelligence
Fast feature selection by means of projections
IEA/AIE'2003 Proceedings of the 16th international conference on Developments in applied artificial intelligence
On the use of meta-learning for instance selection: An architecture and an experimental study
Information Sciences: an International Journal
| Hi-index | 0.00 |
This paper presents a new approach to data set editing. Thealgorithm (EOP: Editing by Ordered Projection) has some interestingcharacteristics: important reduction of the number of examples fromthe database; lower computational cost (O(mn \log n)) with respectto other typical algorithms due to the absence of distancecalculations; conservation of the decision boundaries, especiallyfrom the point of view of the application of axis-parallelclassifiers. The performance of EOP is analysed in two ways:percentage of reduction and classification. EOP has been comparedto IB2, ENN and SHRINK concerning the percentage of reduction andthe computational cost. In addition, we have analysed the accuracyof k-NN and C4.5 after applying the reduction techniques. Anextensive empirical study using databases with continuousattributes from the UCI repository shows that EOP is a valuablepreprocessing method for the later application of any axis-parallellearning algorithm.