One-Class SVMs for Document Classification

By Larry Michael Manevitz, Malik Yousef. Journal of machine learning research, 2002

1,555 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/25c61beda55be5a1f665190384971980afef5386

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

Learning classifiers from only positive and unlabeled data (2008, background): “Unfortunately, the outcome of these methods is sensitive to the values chosen for tuning parameters, and no good way is known to set these values [13].”
A Survey of Recent Trends in One Class Classification (2009, methodology): “OSVMs have been successfully applied in a wide variety of application domains such as Handwritten Digit Recognition [1][2][19], Information Retrieval [20], Face Recognition Applications [28][37], Medical Analysis [38], Bioinformatics [39][40], Spam Detection [41], AnomalyDetection [42][43], Machine ”
Robust Support Vector Machine Training via Convex Outlier Ablation (2006, background): “Most work focuses on the unsupervised case where there is no designated class variable, but we focus on the supervised case here.”
A novel hybrid intrusion detection method integrating anomaly detection with misuse detection (2014, unknown): “A one-class support vector machine (1-class SVM) is a popular outlier detection algorithm in various applications such as document classiﬁcation (Manevitz & Yousef, 2002), machine fault detection (Shin, Eom, & Kim, 2005), and so on.”
Open Category Detection with PAC Guarantees (2018, background): “However, they have been found to perform poorly for open category detection due to poor generalization (Zhou & Huang, 2003), which has been partly addressed by later work (Manevitz & Yousef, 2002; Wu & Ye, 2009; Jin et al., 2004; Cevikalp & Triggs, 2012).”
A Survey of App Store Analysis for Software Engineering (2017, methodology): “machine (SVM) [149] on each cluster to detect outliers in terms of API usage.”
Checking app behavior against app descriptions (2014, methodology): “) OC-SVM has been successfully applied in various contexts that span from document classification [17] to automatic detection of anomalous Windows registry accesses [10].”
On the Class Imbalance Problem (2008, background): “Mainly, two classes of learners were previously studied in the context of the recognition-based one-class approach: SVMs [35][37] and auto-encoders [34][37], and they were found to be competitive [37].”
Text Classification Algorithms: A Survey (2019, methodology): “In general, multi-class SVM is an optimization problem of the following form: min w1,w2,..,wk ,ζ 1 2 ∑k wTk wk + C ∑ (xi ,yi)∈D ζi (111) st. wTyi x− w T k x ≤ i− ζi, ∀(xi, yi) ∈ D, k ∈ {1, 2, ..., K}, k 6= yi (112) where (xi, yi) represent the training data points such that (xi, yi) ∈ D, C is the pe”
Detecting AI Trojans Using Meta Neural Analysis (2019, background): “class SVM [41] will train a hyper-plane which separates all the training data from the origin while maximizing the distance ρ from the origin to the hyper-plane.”
The application of machine learning to structural health monitoring (2007, unknown): “There are strategies for multi-class problems that overcome the need for several SVMs, some are summarized in Bennett (1999).”
A clinical text classification paradigm using weak supervision and deep representation (2019, methodology): “The tf-idf document representation is a common term weighting scheme in information retrieval, which has been also found effective for document classification [52 – 54].”