Correlated Label Propagation with Application to Multi-label Learning

By Feng Kang, Rong Jin, Rahul Sukthankar. 2006

251 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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 Computer Vision and Pattern Recognition

Citations in the corpus

Transductive Multi-View Zero-Shot Learning (2015, background): “Evaluation metrics (a) Hamming Loss: it measures the percentage of mismatches between estimated and ground-truth labels; (b) MicroF1 [16]: it evaluates both micro average of Precision (Micro-Precision) and micro average of Recall (Micro-Recall) with equal importance; (c) Ranking Loss: given the rank”
Multiple instance learning: A survey of problem characteristics and applications (2016, background): “Thus, the observation of nature segments might help to decide if the image contains a cocktail or a bear [68].”
Multilabel dimensionality reduction via dependence maximization (2008, background): “…and Singer 2000; Yu et al. 2005; Zhang and Zhou 2006], scene classi.cation [Boutell et al. 2004; Zhou and Zhang 2007], image and video annotation [Kang et al. 2006; Qi et al. 2007], bioinformatics [Elisseeff and Weston 2002; Barutcuoglu et al. 2006], association rule mining [Thabtah et al.…”
Feature selection for multi-label naive Bayes classification (2009, background): “Besides these eager-style learning algorithms, researchers have also proposed several lazy-style approaches where no training phase is involved and labels of each test instance are predicted based on their similarity to training instances [4,22,46].”
Label Distribution Learning (2013, methodology): “Utilizing such correlation is one of the most important approaches to improve the learning process [16], [32], [29].”
Multi-instance multi-label learning (2008, methodology): “In addition to text categorization, multi-label learning has also been found useful in many other tasks such as scene classification [11], image and video annotation [38,48], bioinformatics [7,12,13,21, 27], and even association rule mining [50, 63].”
Transductive Multilabel Learning via Label Set Propagation (2013, methodology): “This criterion has been used in [19], and [22].”
Interactive shape co-segmentation via label propagation (2014, methodology): “Due to its simplicity and robustness, it has been used in processes such as patch labeling [24], image segmentation [25], and image annotation [26].”
A Review on Multi-Label Learning Algorithms (2014, methodology): “There are other ways to make use of lazy learning for handling multi-label data, such as combining kNN with ranking aggregation [7], [15], identifying kNN in a label-specific style [41], [101], expanding kNN to cover the whole training set [14], [49].”
Large Margin Metric Learning for Multi-Label Prediction (2015, background): “For example, Kang, Jin, and Sukthankar (2006) explicitly exploit high-order correlation between labels, but this involves an optimization problem with an exponential number of constraints.”
Large-Scale Object Classification Using Label Relation Graphs (2014, background): “[20]), but most prior work addresses contextual relations between co-occurring objects (e.”
Hypergraph spectral learning for multi-label classification (2008, background): “Various approaches have been proposed in the past to exploit the correlation information contained in multiple labels [15, 16, 20, 21, 22, 26, 27].”
Facial Age Estimation by Learning from Label Distributions (2010, methodology): “Utilizing such correlation is one of the most important approaches to improve the learning process [33], [18], [31].”
Random k -Labelsets: An Ensemble Method for Multilabel Classification (2007, background): “Multilabel classification is a challenging research problem that emerges in several modern applications such as music categorization [1], protein function classification [2–5] and semantic classification of images [6, 7].”