Support vector machine learning for interdependent and structured output spaces

DMCA

Support vector machine learning for interdependent and structured output spaces (2004)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Ioannis Tsochantaridis , Thomas Hofmann , Thorsten Joachims , Yasemin Altun

Venue:	In ICML
Citations:	449 - 20 self

BibTeX

@INPROCEEDINGS{Tsochantaridis04supportvector,
    author = {Ioannis Tsochantaridis and Thomas Hofmann and Thorsten Joachims and Yasemin Altun},
    title = {Support vector machine learning for interdependent and structured output spaces},
    booktitle = {In ICML},
    year = {2004}
}

OpenURL

Abstract

Learning general functional dependencies is one of the main goals in machine learning. Recent progress in kernel-based methods has focused on designing flexible and powerful input representations. This paper addresses the complementary issue of problems involving complex outputs suchas multiple dependent output variables and structured output spaces. We propose to generalize multiclass Support Vector Machine learning in a formulation that involves features extracted jointly from inputs and outputs. The resulting optimization problem is solved efficiently by a cutting plane algorithm that exploits the sparseness and structural decomposition of the problem. We demonstrate the versatility and effectiveness of our method on problems ranging from supervised grammar learning and named-entity recognition, to taxonomic text classification and sequence alignment. 1.

Keyphrases

output space support vector machine complementary issue sequence alignment multiclass support vector machine learning main goal structural decomposition grammar learning general functional dependency complex output optimization problem machine learning recent progress powerful input representation named-entity recognition plane algorithm text classification multiple dependent output variable kernel-based method

DMCA