Text Categorization with Support Vector Machines: Learning with Many Relevant Features

By Thorsten Joachims. 1999

How researchers see this paper

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

In Natural Language Processing

Citations in the corpus

Revisiting LSTM Networks for Semi-Supervised Text Classification via Mixed Objective Function (2019, methodology): “Early machine learning approaches for text classiﬁcation were based on the extraction of bag-of-words features followed by a supervised classiﬁer such as na¨ıve Bayes (McCallum and Nigam 1998) or a linear SVM (Joachims 1998).”
Character-level Convolutional Networks for Text Classification (2015, background): “To date, almost all techniques of text classification are based on words, in which simple statistics of some ordered word combinations (such as n-grams) usually perform the best [12].”
FastText.zip: Compressing text classification models (2016, background): “Linear classiﬁers often obtain state-of-the-art performance while being scalable (Agarwal et al., 2014; Joachims, 1998; Joulin et al., 2016; McCallum & Nigam, 1998).”
Bag of Tricks for Efficient Text Classification (2016, background): “A simple and efﬁcient baseline for sentence classiﬁcation is to represent sentences as bag of words (BoW) and train a linear classiﬁer, e.g., a logistic regression or an SVM (Joachims, 1998; Fan et al., 2008).”
Optimization Methods for Large-Scale Machine Learning (2016, background): “In summary, both theoretical arguments [38] and experimental evidence [81] indicate that a carefully selected family of prediction functions and such high-dimensional representations of documents lead to good performance while avoiding overfitting— recall (2.”
Hierarchical Attention Networks for Document Classification (2016, methodology): “Traditional approaches of text classiﬁcation represent documents with sparse lexical features, such as n -grams, and then use a linear model or kernel methods on this representation (Wang and Manning, 2012; Joachims, 1998).”
Machine Learning Models that Remember Too Much (2017, methodology): “gapmeasureshowoverfittedthemodelistoitstrainingdataset. Linear models. Support Vector Machines (SVM) [17] and logistic regression (LR) are popular for classification tasks such as text categorization [35] and other natural language processing problems [8]. We assume feature space X= Rd for some dime”
Hacking smart machines with smarter ones: How to extract meaningful data from machine learning classifiers (2013, background): “on ([39, 34, 22]) or as part of complex systems for similar tasks ([48, 6]). Other authors propose SVM-based systems for privacy-critical tasks, such as cancer diagnostic [30, 7], text categorization [36], or face recognition [4]. SVM for network traﬃc classiﬁcation As shown by the extensive literat”
Large-Scale Hierarchical Text Classification with Recursively Regularized Deep Graph-CNN (2018, background): “Topical text classification is a fundamental text mining problem for many applications, such as news classification [18], question”
Text Categorization as a Graph Classification Problem (2015, methodology): “The ﬁrst three rows correspond to the baselines: unsupervised n-gram feature extraction and then supervised learning using kNN, NB (Multinomial but Bernoulli yields similar results) and linear SVM.”
Training a Multilingual Sportscaster: Using Perceptual Context to Learn Language (2014, background): “SVMs are state-ofthe-art machine learning methods that learn maximum-margin separators to prevent over-fitting in very high-dimensional data such as natural language text (Joachims, 1998).”
Kernel Methods for Relation Extraction (2002, methodology): “The normalization corresponds to the standard unit norm normalization of examples in the feature space corresponding to the kernel space (Cristianini and Shawe-Taylor, 2000): p For both SVM Light and Voted Perceptron, we used their standard conﬁgurations (e.g., we did not optimize the value of C tha”
Supervised and Semi-Supervised Text Categorization using LSTM for Region Embeddings (2016, methodology): “The state-of-the art methods for text categorization had long been linear predictors (e.g., SVM with a linear kernel) with either bag-of-word or bag-of-n -gram vectors (hereafter bow ) as input, e.g., (Joachims, 1998; Lewis et al., 2004).”