Parametric Mixture Models for Multi-Labeled Text

By Naonori Ueda, Kazumi Saito. Neural Information Processing Systems, 2002

384 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/d5388a48970e0fbea9dfcb23b607578e95537da6

How researchers see this paper

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

Citations in the corpus

Collective multi-label classification (2005, background): “[16] take a similar approach in that each word in each category is generated from a multinomial distribution over vocabulary words.”
A Literature Survey on Algorithms for Multi-label Learning (2010, methodology): “Several generative modeling based approach had been proposed [32, 56, 59] especially for modeling the generative process of text based multi-label data.”
Statistical topic models for multi-label document classification (2011, methodology): “comparison, we present results from three more commonly used benchmark multi-label datasets: the RCV1-v2 dataset of Lewis et al. (2004) and the Arts and Health subdirectories from the Yahoo! dataset (Ueda and Saito, 2002; Ji et al., 2008), all of which have signiﬁcantly fewer labels, and more exampl”
A k-nearest neighbor based algorithm for multi-label classification (2005, methodology): “One year later, using independent word-based Bag-ofWords representation [11], Ueda and Saito [3] presented two types of probabilistic generative models for multi-label text called parametric mixture models (PMM1, PMM2), where the basic assumption under PMMs is that multi-label text has a mixture of ”
A Review on Multi-Label Learning Algorithms (2014, background): “1 Early researches on multi-label learning mainly focus on the problem of multi-label text categorization [63], [75], [97].”
Learning author-topic models from text corpora (2010, background): “…approaches to modeling document content based on the idea that the probability distribution over words in a document can be expressed as a mixture of topics, where each topic is a probability distribution over words [Blei et al. 2003; Hofmann 1999; Ueda and Saito 2003; Iyer and Ostendorf 1999].”
Efficient multi-label ranking for multi-class learning: Application to object recognition (2009, background): “Another limitation of these approaches is that they are unable to capture the correlation among classes, which is known to be important in multilabel learning [22].”
Labeled LDA: A supervised topic model for credit attribution in multi-labeled corpora (2009, methodology): “As a baseline, we use a set of multiple one-vs-rest SVM classiﬁers which is a popular and extremely competitive baseline used by most previous papers (see (Kazawa et al., 2004; Ueda and Saito, 2003) for instance).”
Lift: Multi-Label Learning with Label-Specific Features (2011, background): “Early researches on multi-label learning originate from the investigation of multi-label text categorization techniques [31], [40], [46].”
Learning shared subspace for multi-label dimensionality reduction via dependence maximization (2015, methodology): “Multi-label classification has recently gained significant attention in many applications such as multi-topic document categorization [1,2], protein function prediction [3,4] and automatic image annotation [5,6].”
Mining Multi-label Data (2010, methodology): “A similar word-based mixture model for multi-label text classification is presented in [17].”
A convex formulation for learning shared structures from multiple tasks (2009, methodology): “Experiments In this section, we evaluate the proposed cASO algorithm on the tasks of multi-topic web pages categorization using the Yahoo web pages data sets (Ueda & Saito, 2002).”
Multi-Label Learning with Global and Local Label Correlation (2017, background): “For example, a scene image can be annotated with several tags [1], a document may belong to multiple topics [2], and a piece of music may be associated with different genres [3].”
Transductive Multilabel Learning via Label Set Propagation (2013, methodology): “Extension of other traditional learning techniques have also been studied, such as probabilistic generative models [24], [31], decision trees [8], neural networks [34], maximal margin methods [15], [20], maximum entropy methods [14], [41], and ensemble methods [12].”
Exclusive Lasso for Multi-task Feature Selection (2010, methodology): “We use the Yahoo dataset (Ueda and Saito 2003) in our experiments.”
Scalable multi-label annotation (2014, background): “This technique is important in multiple domains, such as labeling actions in videos [11], news article topics [15], functional classes of genes [8], musical attributes or emotions in songs [12], semantic classes of scenes [2], product categories customers are likely to buy [21], and categories of we”
ML-KNN: A lazy learning approach to multi-label learning (2007, methodology): “Recently, Ueda and Saito [21] presented two types of probabilistic generative models called parametric mixture models (Pmm1, Pmm2) for multilabel text.”