Ììì Öûûò Ë Blockinöö Óóóòòòö Áòøøöòòøøóòòð

By Susan T. Dumais, John C. Platt, David Hecherman, Mehran Sahami. 1992

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

Natural Language Processing for EHR-Based Computational Phenotyping (2018, methodology): “explored other methods such as random forests [112], decision trees [100], [113], [131], [132], and the Longitudinal Gamma Poisson Shrinker [133], [134] for computational phe-”
SMOTE: Synthetic Minority Over-sampling Technique (2002, background): “The information retrieval (IR) domain (Dumais et al., 1998; Mladenić & Grobelnik, 1999; Lewis & Ringuette, 1994; Cohen, 1995a) also faces the problem of class imbalance in the dataset.”
Kernels for small molecules and the prediction of mutagenicity, toxicity and anti-cancer activity (2005, methodology): “An alternative to the TF–IDF scheme to preserve/enhance paths carrying the most relevant information for a given classification task is to consider a reduced set of paths selected according to the mutual information criterion (Yang and Pedersen, 1997; Dumais et al., 1998).”
Estimating the Generalization Performance of an SVM Efficiently (2000, methodology): “The following presents an approach to answering these questions in the context of text classi cation with Support Vector Machines (SVMs) (Joachims, 1998; Dumais et al., 1998).”
Feature Generation for Text Categorization Using World Knowledge (2005, methodology): “In particular, representations based on phrases [Dumaiset al., 1998; Fuernkranzet al., 2000], named entities [Kumaran and Allan, 2004 ], and term clustering[Lewis and Croft, 1990; Bekkerman, 2003 ] have been explored.”
Latent Semantic Kernels (2001, methodology): “Joachims (1998) and Dumais et al. (1998) have applied this technique to the basic vector space model for a classification task with impressive results.”
Sentiment Classification Using Word Sub-sequences and Dependency Sub-trees (2005, methodology): “They applied to document sentiment classiflcation a supervised machine learning method which had succeeded in other document classiflcation tasks (e.g., on the task classifying articles of Reuters to 10 categories, Dumais et al [ 8 ] achieved F-measure of 0.92 with SVMs.”
The MID4 dataset, 2002–2010: Procedures, coding rules and description (2015, methodology): “SVMs are commonly used algorithms for the classification of text documents (Abe, 2010; Dumais et al., 1998; Zhang and Oles, 2001).”
Learning on the border: active learning in imbalanced data classification (2007, background): “The standard machine learning algorithms yield better prediction performance with balanced datasets.”
Less is More: Active Learning with Support Vector Machines (2000, background): “Support vector machines have demonstrated excellent performance in many domains, particularly those involving text classification (Joachims, 1998b; Dumais et al., 1998).”
Graph kernels for chemical informatics (2005, background): “…also in the field of information retrieval, is to consider a reduced set of paths selected according to the mutual information criterion [Yang and Pedersen, 1997, Dumais et al., 1998] in order to preserve or enhance paths carrying the most relevant information for a given classification task.”
Discriminatively trained Markov model for sequence classification (2005, methodology): “However, because of the large size of the vocabulary, we performed feature selection based on mutual information [9] to reduce the vocabulary size to 300 words.”
Supervised term weighting for automated text categorization (2003, unknown): “5 An experimental comparison of subsets (a) and (c) is reported in [ 3 ].”