Recall Systems: Effcient Learning and Use of Category Indices

Omid Madani, Wiley Greiner, David Kempe, Mohammad R. Salavatipour
Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, PMLR 2:307-314, 2007.

Abstract

We introduce the framework of recall systems for efficient learning and retrieval of categories when the number of categories is large. A recallsystem here is a simple feature-based intermediate filtering step which reduces the potential categories for an instance to a small manageable set. The correct categories from this set can then be determined using traditional classifiers. We present a formalization of the index learning problem and establish NP-hardness and approximation hardness. We proceed to give an efficient heuristic for learning indices, and evaluate it on several large data sets. In our experiments, the index is learned within minutes, and reduces the number of categories by several orders of magnitude, without affecting the quality of classification overall.

Cite this Paper


BibTeX
@InProceedings{pmlr-v2-madani07a, title = {Recall Systems: Effcient Learning and Use of Category Indices}, author = {Madani, Omid and Greiner, Wiley and Kempe, David and Salavatipour, Mohammad R.}, booktitle = {Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics}, pages = {307--314}, year = {2007}, editor = {Meila, Marina and Shen, Xiaotong}, volume = {2}, series = {Proceedings of Machine Learning Research}, address = {San Juan, Puerto Rico}, month = {21--24 Mar}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v2/madani07a/madani07a.pdf}, url = {https://proceedings.mlr.press/v2/madani07a.html}, abstract = {We introduce the framework of recall systems for efficient learning and retrieval of categories when the number of categories is large. A recallsystem here is a simple feature-based intermediate filtering step which reduces the potential categories for an instance to a small manageable set. The correct categories from this set can then be determined using traditional classifiers. We present a formalization of the index learning problem and establish NP-hardness and approximation hardness. We proceed to give an efficient heuristic for learning indices, and evaluate it on several large data sets. In our experiments, the index is learned within minutes, and reduces the number of categories by several orders of magnitude, without affecting the quality of classification overall.} }
Endnote
%0 Conference Paper %T Recall Systems: Effcient Learning and Use of Category Indices %A Omid Madani %A Wiley Greiner %A David Kempe %A Mohammad R. Salavatipour %B Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2007 %E Marina Meila %E Xiaotong Shen %F pmlr-v2-madani07a %I PMLR %P 307--314 %U https://proceedings.mlr.press/v2/madani07a.html %V 2 %X We introduce the framework of recall systems for efficient learning and retrieval of categories when the number of categories is large. A recallsystem here is a simple feature-based intermediate filtering step which reduces the potential categories for an instance to a small manageable set. The correct categories from this set can then be determined using traditional classifiers. We present a formalization of the index learning problem and establish NP-hardness and approximation hardness. We proceed to give an efficient heuristic for learning indices, and evaluate it on several large data sets. In our experiments, the index is learned within minutes, and reduces the number of categories by several orders of magnitude, without affecting the quality of classification overall.
RIS
TY - CPAPER TI - Recall Systems: Effcient Learning and Use of Category Indices AU - Omid Madani AU - Wiley Greiner AU - David Kempe AU - Mohammad R. Salavatipour BT - Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics DA - 2007/03/11 ED - Marina Meila ED - Xiaotong Shen ID - pmlr-v2-madani07a PB - PMLR DP - Proceedings of Machine Learning Research VL - 2 SP - 307 EP - 314 L1 - http://proceedings.mlr.press/v2/madani07a/madani07a.pdf UR - https://proceedings.mlr.press/v2/madani07a.html AB - We introduce the framework of recall systems for efficient learning and retrieval of categories when the number of categories is large. A recallsystem here is a simple feature-based intermediate filtering step which reduces the potential categories for an instance to a small manageable set. The correct categories from this set can then be determined using traditional classifiers. We present a formalization of the index learning problem and establish NP-hardness and approximation hardness. We proceed to give an efficient heuristic for learning indices, and evaluate it on several large data sets. In our experiments, the index is learned within minutes, and reduces the number of categories by several orders of magnitude, without affecting the quality of classification overall. ER -
APA
Madani, O., Greiner, W., Kempe, D. & Salavatipour, M.R.. (2007). Recall Systems: Effcient Learning and Use of Category Indices. Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 2:307-314 Available from https://proceedings.mlr.press/v2/madani07a.html.

Related Material