A kernel method for multi-labelled classification

By Andre´ Elisseeﬀ, Jason Aaron Edward Weston. 2001

1,523 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Molecular Biology

How researchers see this paper

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

In Artificial Intelligence

In Molecular Biology

Citations in the corpus

Transductive Multi-View Zero-Shot Learning (2015, background): “However, this does not exploit the multi-label correlations and thus has very limited expressive power [5, 33].”
Kernel methods in machine learning (2007, positive): “Weston et al. [153] proposed an algorithm which allows one to extend standard regularized LS regression models, as described in Section 3.3, to cases where Y has complex structure.”
Structured Prediction Energy Networks (2015, background): “Modeling improvements use max-margin or ranking losses that directly address the multi-label structure (Elisseeff & Weston, 2001; Godbole & Sarawagi, 2004; Bucak et al., 2009).”
On label dependence and loss minimization in multi-label classification (2012, methodology): “Rank-SVM is an instantiation of SVMs that can be applied for this type of problems (Elisseeff and Weston 2002).”
A Survey of Audio-Based Music Classification and Annotation (2011, methodology): “When the SVM classifier was applied to the same features (e.g., [16], [19]), it produced lower accuracy rates.”
Correlated Label Propagation with Application to Multi-label Learning (2006, background): “Another group of approaches toward multi-label learning is label ranking [5, 7 , 18].”
Random k -Labelsets: An Ensemble Method for Multilabel Classification (2007, background): “There exist multilabel extensions of decision tree [2], support vector machine [13, 14], neural network [15,”
General Multi-label Image Classification with Transformers (2020, background): “Multi-label image classi-ﬁcation is a visual recognition task that aims to predict a set of labels corresponding to objects, attributes, or actions given an input image [15, 46, 48, 50, 6, 32, 9].”
Efficient max-margin multi-label classification with applications to zero-shot learning (2012, methodology): “A hierarchical loss is deﬁned and optimised using the ranking method of Elisseeff and Weston (2001).”
Recent advances in feature selection and its applications (2017, methodology): “Moreover, a multi-label feature selection method is also described in [67], which is based on the minimization of correlation regularized loss of label ranking [68].”
On the Stratification of Multi-label Data (2011, neutral): “We first observe that iterative stratification achieves the best performance in terms of LD in all datasets apart from Scene, Yeast and TMC2007, where the labelsets-based method is better.”
Ensemble application of convolutional and recurrent neural networks for multi-label text categorization (2017, neutral): “For algorithm adaptation, ML-KNN [19], ML-DT [20], and Rank-SVM [21] adapt popular classification algorithms, e.g., k-nearest-neighbors, decision trees, support vector machines, to solve multi-label classification problems.”
mlDEEPre: Multi-Functional Enzyme Function Prediction With Hierarchical Multi-Label Deep Learning (2019, background): “To solve the problem, Elisseeﬀ and Weston (2002) proposed an excellent idea to incorporate the information of each single data point into the threshold, which replaces the constant with a linear function t ( x i ) = w ⊺ · l ( x i ) + b , where l ( x i ) is the output of the network on the instance…”
Addressing imbalance in multilabel classification: Measures and random resampling algorithms (2015, methodology): “In the algorithm adaptation approach there are proposals of multilabel C4.5 trees [14], algorithms based on nearest neighbors such as ML-kNN [15], multilabel neural networks [2] [16], and multilabel SVMs [17].”
A Literature Survey on Algorithms for Multi-label Learning (2010, methodology): “in [18] preprocessed the data and retained the functional classes structured into hierarchies up to 4 levels deep and thats how this data is most commonly used in multi-label problems.”
Multi-label learning by exploiting label dependency (2010, background): “To name a few, in text categorization, each document may belong to several topics, such as Shanghai World Expo, economics and even volunteers [14, 19]; In bioinformatics, each gene may be associated with a number of functional classes, such as metabolism, transcription and protein synthesis [7]; In ”
Combining instance-based learning and logistic regression for multilabel classification (2009, methodology): “…well-established methods for conventional classiﬁ-cation have been extended to the multi-label case, including support vector machines [Godbole and Sarawagi, 2004; Elisseeff and Weston, 2002; Boutell et al. , 2004], neural networks [Zhang and Zhou, 2006], and decision trees [Vens et al. , 2008].”