Name
Abstract | ||
|---|---|---|
While hierarchical semi-supervised classification methods have been previously studied, we still lack an algorithm that can learn a non-predefined categorical hierarchy from multi-labeled data at various levels of specificity. Inspired by human psychology and learning experience, in this paper we propose a semi-supervised learning method that can classify multi-labeled data into a hierarchy based on the label's specificity level such that the separability between each class and its siblings is greater than the separability between each class and its parents. To build the hierarchy we show that a minimum spanning tree minimizes an upper bound on the pairwise Kullback-Liebler divergence between the true and approximated distributions. We show the effectiveness of our method using three types of data sets and draw a comparison between our learned hierarchy and one learned by human subjects using the same data set. We also show the effectiveness of our method compared to hierarchical clustering. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1109/MLSP.2011.6064565 | Machine Learning for Signal Processing |
Keywords | Field | DocType |
learning (artificial intelligence),pattern classification,pattern clustering,psychology,trees (mathematics),approximated distributions,hierarchical clustering,hierarchical semisupervised classification methods,human psychology,human subjects,minimum spanning tree,multilabeled data,nonpredefined categorical hierarchy,pairwise kullback-liebler divergence,semisupervised hierarchy learning,learning artificial intelligence,semi supervised learning,upper bound | Hierarchical clustering,Pairwise comparison,Semi-supervised learning,Pattern recognition,Computer science,Categorical variable,Artificial intelligence,Conceptual clustering,Hierarchy,Machine learning,Single-linkage clustering,Minimum spanning tree | Conference |
ISSN | ISBN | Citations |
1551-2541 E-ISBN : 978-1-4577-1622-5 | 978-1-4577-1622-5 | 0 |
PageRank | References | Authors |
0.34 | 4 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ailar Javadi | 1 | 0 | 0.34 |
| Alexander G. Gray | 2 | 990 | 80.16 |
| David V. Anderson | 3 | 418 | 75.23 |
| Visar Berisha | 4 | 76 | 22.38 |