New approaches to support vector ordinal regression

By Wei Chu, S. Sathiya Keerthi. International Conference on Machine Learning, 2005

317 citations recorded by the source corpus.

Subfields: Analytical Chemistry, Computer Vision and Pattern Recognition, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/23cb9cb90a31ca7c899beeae5dac22be0ffdaaf9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Analytical Chemistry

In Computer Vision and Pattern Recognition

In Statistics and Probability

Citations in the corpus

Gaussian Processes for Ordinal Regression (2005, background): “Shashua and Levin (2003) generalized the support vector formulation by finding multiple thresholds to define parallel discriminant hyperplanes for ordinal scales, and reported that the performance of the support vector approach is better than that of the on-line algorithm (Crammer and Singer, 2002).”
Rank consistent ordinal regression for neural networks with application to age estimation (2019, methodology): “Early work in this regard includes the use of perceptrons (Crammer and Singer, 2002; Shen and Joshi, 2005) and support vector machines (Herbrich et al., 1999; Shashua and Levin, 2003; Rajaram et al., 2003; Chu and Keerthi, 2005).”
Ordinal Regression by Extended Binary Classification (2006, methodology): “In addition, they found that SVOR-EXP performed better in terms of the classification cost, and SVOR-IMC preceded in terms of the absolute cost.”
Learning to rank for information retrieval (2009, unknown): “In recent years, learning to rank has become a very hot research direction in IR, and a large number of learning-to-rank algorithms have been proposed, such as [9, 13, 14, 16, 17, 26, 29, 33, 34, 47, 49, 59, 63, 73, 78, 90, 97, 99, 102, 114, 119, 122, 129, 134, 136].”
A Modified Finite Newton Method for Fast Solution of Large Scale Linear SVMs (2005, methodology): “To set up the SVM formulation we follow the approach given in Chu and Keerthi (2005) and use p−1 thresholds,bs, s= 1, . . . , p−1 to divide the scores intop bins so that the interval,(bs−1,bs) is assigned for examples which haveti = s.13 Let β denote the vector which containsw together with 13.”
LETOR: Benchmark Dataset for Research on Learning to Rank for Information Retrieval (2007, unknown): “Many methods have been proposed (e.g., [2], [5], [6], [7], [10], [16], [20], [22], [23]) and applied to IR applications (e.g., [3], [12], [27], [31], [32], [33]).”
Assigning Polarity Scores to Reviews Using Machine Learning Techniques (2005, methodology): “To create a data set with a uniform score distribution (the eﬀect of skewed class distributions is out of the scope of this paper), we selected 330 reviews per sp-score for Corpus A and 280 reviews per sp-score for Corpus B 5 .”
Support Vector Ordinal Regression (2007, methodology): “Full details of the SMO algorithm are given in our technical report (Chu and Keerthi, 2005).”
On the Consistency of Ordinal Regression Methods (2014, methodology): “Following [6], we will consider two variants of the margin-based loss function ψC for ` = `0−1 and ` = `A with φ = hinge loss.”
Support Vector Regression (2007, background): “Both approaches have guaranteed that the thresholds are properly ordered at the optimal solution [91].”
Ordinal Regression Methods: Survey and Experimental Study (2016, background): “Consequently, Chu and Keerthi [29], [91] proposed two different reformulations for the same idea, solving the problem of unordered thresholds at the solution.”