BoosTexter: A Boosting-based System for Text Categorization

By Robert E. Schapire, Yoram Singer. Machine-mediated learning, 2000

2,430 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/e7a07c3aaef303850e5a1fcc81bb44f6d2db6696

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

Citations in the corpus

Efficient Object Category Recognition Using Classemes (2010, methodology): “To use classemes for multiclass classification, a number of strategies were implemented: multiclass SVMs [21], neural networks, decision forests [22], a nearestneighbour-like classifier, boosting [23].”
Multilabel Classification with Principal Label Space Transformation (2012, background): “, 2008) and text categorization (Schapire and Singer, 2000), multi-label classification is attracting more and more research attention.”
Multitask Learning for Spoken Language Understanding (2006, methodology): “We performed our tests using the Boostexter classification tool [11], an implementation of the Boosting family of classifiers.”
Combining instance-based learning and logistic regression for multilabel classification (2009, methodology): “…learning in recent years, and a number of methods has been developed, often motivated by speciﬁc types of applications such as text categorization [ Schapire and Singer, 2000; Ueda and Saito, 2003; Kazawa et al. , 2005; Zhang and Zhou, 2006], computer vision [Boutell et al. , 2004], and…”
Learning shared subspace for multi-label dimensionality reduction via dependence maximization (2015, background): “Typical examples include neural network [8,9], lazy learning [10–12], Adaboost MR [13,14], and rank SVM [15] .”
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”
Ensembles of Learning Machines (2002, methodology): “While in bagging the samples are drawn with replacement using a uniform probability distribution, in boosting methods the learning algorithm is called at each iteration using a different distribution or weighting over the training examples [111, 40, 112, 39, 115, 110, 32, 38, 33, 32, 16, 17, 42, 41]”
Two-Stage Label Embedding via Neural Factorization Machine for Multi-Label Classification (2019, methodology): “Many techniques (Schapire and Singer 2000; Zhang and Zhou 2007) have been proposed to deal with Multi-Label Classiﬁcation (MLC) problems.”
Sharing Visual Features for Multiclass and Multiview Object Detection (2007, methodology): “However, nearly all of this work focuses on feedforward neural networks, whereas we use quite a different kind of classifier based on boosted decision stumps [ 29 ].”
A review of associative classification mining (2007, methodology): “Moreover, the right method is often problematic and depends heavily on the features of the conducted problem (Schapire & Singer, 2000; Boutell et al., 2003).”
A Literature Survey on Algorithms for Multi-label Learning (2010, background): “Discriminative SVM Based Methods One important problem with tree based boosting [44] (discussed in section 2.2) is that, they are likely to overfit with relatively smaller ( 1000) training set.”
Discriminative Methods for Multi-labeled Classification (2004, neutral): “Text classification can be broadly categorized into discriminative techniques, typified by support vector machines [1] (SVMs), decision trees [2] and neural networks; and generative techniques, like Näıve Bayes (NB) and Expectation Maximization (EM) based methods.”
What is left to be understood in ATIS? (2010, background): “MH multiclass multilabel classification algorithm [12].”
Neural Approaches to Conversational AI (2018, positive): “et al., 2017d). Among them, the ﬁrst two tasks are often framed as a classiﬁcation problem, which infers the domain or intent (from a predeﬁned set of candidates) based on the current user utterance (Schapire and Singer, 2000; Yaman et al., 2008; Sarikaya et al., 2014). Neural approaches to multi-cl”
Large Margin Metric Learning for Multi-Label Prediction (2015, background): “For instance, a document can be associated with a range of topics, such as News, Finance and Sports (Schapire and Singer 2000); a gene belongs to the functions of protein synthesis, metabolism and transcription (Barutcuoglu, Schapire, and Troyanskaya 2006); an image may have both beach and urban tag”