Collective multi-label classification

By Nadia Ghamrawi, Andrew McCallum. International Conference on Information and Knowledge Management, 2005

495 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/e18936ac394bd7bc169ee30e6b82f6b40fbf1a08

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 Information Systems

Citations in the corpus

Learning Semantic-Specific Graph Representation for Multi-Label Image Recognition (2019, methodology): “To achieve this, a series of works introduced graphic models, such as Conditional Random Field [8], Dependency Network [10], or co-occurrence matrix [29] to capture pair-wise label correlations.”
Multi-label learning by exploiting label dependency (2010, background): “• Second-order approaches: The task of multi-label learning is tackled by considering the pairwise relations between labels, such as the ranking between the proper label and the improper label of an example [7, 8, 19, 28], or the interaction between any pair of labels [9, 16, 24, 30].”
Structured Prediction Energy Networks (2015, background): “…techniques’ complexity, both in terms of the number of parameters to estimate and the per-iteration cost of algorithms like belief propagation, grows at least quadratically in the number of labels L (Ghamrawi & McCallum, 2005; Finley & Joachims, 2008; Meshi et al., 2010; Petterson & Caetano, 2011).”
Efficient multi-label ranking for multi-class learning: Application to object recognition (2009, background): “This is summarized into the following optimization problem. min {fk∈Hκ}Kk=1 1 2 K∑ k=1 |fk|2Hκ + C n∑ i=1 K∑ k,l=1 εk,li , (2) where κ(x, x′) : Rd × R → R is a kernel function, Hκ is a Hilbert space endowed with a kernel function κ(·, ·) and C is a constant parameter.”
Maximizing Subset Accuracy with Recurrent Neural Networks in Multi-label Classification (2017, methodology): “Aside from PCC, there has been another family of probabilistic approaches to maximizing subset accuracy [9, 16].”
A Literature Survey on Algorithms for Multi-label Learning (2010, methodology): “Figure 1: Graphical representation of CoL Model (Figure taken from [59]) A conditional random field (CRF) based model is presented in [21] where two multi-label graphical models has been proposed, both parameterizes label co-occurances.”
LEAP: Learning to Prescribe Effective and Safe Treatment Combinations for Multimorbidity (2017, methodology): “The label dependency can also be modeled by graphical models, such as Conditional Random Fields [12], Dependency Networks [15], and co-occurrence matrices [39].”
Mining Multi-label Data (2010, background): “Classification accuracy [20] or subset accuracy [22] is defined as follows:”
Multi-label Image Recognition by Recurrently Discovering Attentional Regions (2017, methodology): “To better consider the correlations between labels instead of treating each label independently, traditional graphical models are widely incorporated, such as Conditional Random Field [8], Dependency Network [11], and co-occurrence matrix [29].”
On label dependence and loss minimization in multi-label classification (2012, background): “This includes, for example, conditional random fields (CRF) (Lafferty et al. 2001; Ghamrawi and McCallum 2005), a specific type of graphical model that allows for representing relationships between labels and features in a quite convenient way.”
Probabilistic Multi-Label Classification with Sparse Feature Learning (2013, unknown): “, 2010], and probabilistic graphical models [Ghamrawi and McCallum, 2005; de Waal and van der Gaag, 2007; Bielza et al., 2011; Zaragoza et al., 2011; Guo and Gu, 2011].”
Recurrent Attentional Reinforcement Learning for Multi-label Image Recognition (2017, methodology): “Speciﬁcally, our model achieves a mAP of (Gong et al. 2013), on this dataset.”
An Introduction to Conditional Random Fields (2010, unknown): “Rather than learning an independent classifier for each category, Ghamrawi and McCallum [35] present a CRF that learns dependencies between the categories, resulting in improved classification performance.”
SGM: Sequence Generation Model for Multi-label Classification (2018, background): “Collective multi-label classiﬁer (CML) (Ghamrawi and McCallum, 2005) adopts maximum entropy principle to deal with multi-label data by encoding label correlations as constraint conditions.”
Minimum-Risk Training of Approximate CRF-Based NLP Systems (2012, methodology): “We follow previous work (Ghamrawi and McCallum, 2005; Finley and Joachims, 2008) and use a fully connected CRF to model all pairwise dependencies between labels.”