Adaptive learning resources search mechanism
Proceedings of the second ACM international workshop on Multimedia technologies for distance leaning
Batch document filtering using nearest neighbor algorithm
CLEF'09 Proceedings of the 10th cross-language evaluation forum conference on Multilingual information access evaluation: text retrieval experiments
Similarity learning in nearest neighbor and application to information retrieval
FDIA'09 Proceedings of the Third BCS-IRSG conference on Future Directions in Information Access
Test-data volume optimization for diagnosis
Proceedings of the 49th Annual Design Automation Conference
| Hi-index | 0.00 |
In this paper, we propose an algorithm for learning a general class of similarity measures for kNN classification. This class encompasses, among others, the standard cosine measure, as well as the Dice and Jaccard coefficients. The algorithm we propose is an extension of the voted perceptron algorithm and allows one to learn different types of similarity functions (either based on diagonal, symmetric or asymmetric similarity matrices). The results we obtained show that learning similarity measures yields significant improvements on several collections, for two prediction rules: the standard kNN rule, which was our primary goal, and a symmetric version of it.