Abstract
Over the past decade or so, several research groups have addressed the problem of multi-label classification where each example can belong to more than one class at the same time. A common approach, called Binary Relevance (BR), addresses this problem by inducing a separate classifier for each class. Research has shown that this framework can be improved if mutual class dependence is exploited: an example that belongs to class X is likely to belong also to class Y; conversely, belonging to X can make an example less likely to belong to Z. Several works sought to model this information by using the vector of class labels as additional example attributes. To fill the unknown values of these attributes during prediction, existing methods resort to using outputs of other classifiers, and this makes them prone to errors. This is where our paper wants to contribute. We identified two potential ways to prune unnecessary dependencies and to reduce error-propagation in our new classifier-stacking technique, which is named PruDent. Experimental results indicate that the classification performance of PruDent compares favorably with that of other state-of-the-art approaches over a broad range of testbeds. Moreover, its computational costs grow only linearly in the number of classes.
| Original language | English (US) |
|---|---|
| Article number | 7069233 |
| Pages (from-to) | 2480-2493 |
| Number of pages | 14 |
| Journal | IEEE Transactions on Knowledge and Data Engineering |
| Volume | 27 |
| Issue number | 9 |
| DOIs | |
| State | Published - Sep 1 2015 |
Fingerprint
Keywords
- chaining
- label dependence
- Multi-label classification
- stacking
ASJC Scopus subject areas
- Computational Theory and Mathematics
- Information Systems
- Computer Science Applications
Cite this
PruDent : A Pruned and Confident Stacking Approach for Multi-Label Classification. / Alali, Abdulaziz; Kubat, Miroslav.
In: IEEE Transactions on Knowledge and Data Engineering, Vol. 27, No. 9, 7069233, 01.09.2015, p. 2480-2493.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - PruDent
T2 - A Pruned and Confident Stacking Approach for Multi-Label Classification
AU - Alali, Abdulaziz
AU - Kubat, Miroslav
PY - 2015/9/1
Y1 - 2015/9/1
N2 - Over the past decade or so, several research groups have addressed the problem of multi-label classification where each example can belong to more than one class at the same time. A common approach, called Binary Relevance (BR), addresses this problem by inducing a separate classifier for each class. Research has shown that this framework can be improved if mutual class dependence is exploited: an example that belongs to class X is likely to belong also to class Y; conversely, belonging to X can make an example less likely to belong to Z. Several works sought to model this information by using the vector of class labels as additional example attributes. To fill the unknown values of these attributes during prediction, existing methods resort to using outputs of other classifiers, and this makes them prone to errors. This is where our paper wants to contribute. We identified two potential ways to prune unnecessary dependencies and to reduce error-propagation in our new classifier-stacking technique, which is named PruDent. Experimental results indicate that the classification performance of PruDent compares favorably with that of other state-of-the-art approaches over a broad range of testbeds. Moreover, its computational costs grow only linearly in the number of classes.
AB - Over the past decade or so, several research groups have addressed the problem of multi-label classification where each example can belong to more than one class at the same time. A common approach, called Binary Relevance (BR), addresses this problem by inducing a separate classifier for each class. Research has shown that this framework can be improved if mutual class dependence is exploited: an example that belongs to class X is likely to belong also to class Y; conversely, belonging to X can make an example less likely to belong to Z. Several works sought to model this information by using the vector of class labels as additional example attributes. To fill the unknown values of these attributes during prediction, existing methods resort to using outputs of other classifiers, and this makes them prone to errors. This is where our paper wants to contribute. We identified two potential ways to prune unnecessary dependencies and to reduce error-propagation in our new classifier-stacking technique, which is named PruDent. Experimental results indicate that the classification performance of PruDent compares favorably with that of other state-of-the-art approaches over a broad range of testbeds. Moreover, its computational costs grow only linearly in the number of classes.
KW - chaining
KW - label dependence
KW - Multi-label classification
KW - stacking
UR - http://www.scopus.com/inward/record.url?scp=84939241890&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84939241890&partnerID=8YFLogxK
U2 - 10.1109/TKDE.2015.2416731
DO - 10.1109/TKDE.2015.2416731
M3 - Article
AN - SCOPUS:84939241890
VL - 27
SP - 2480
EP - 2493
JO - IEEE Transactions on Knowledge and Data Engineering
JF - IEEE Transactions on Knowledge and Data Engineering
SN - 1041-4347
IS - 9
M1 - 7069233
ER -