A Divisive Information-Theoretic Feature Clustering Algorithm for Text Classification

By Inderjit S. Dhillon, Subramanyam Mallela, Rahul Kumar. Journal of machine learning research, 2003

650 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/cc520354a3cd105b7db4cb8c82225374f8886ba8

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

A generalized maximum entropy approach to bregman co-clustering and matrix approximation (2004, unknown): “Dhillon et al. (2003b) showed that I(X; Y ) − I(X̂, Ŷ ) = KL(p(X, Y )||q(X, Y )) , (1.”
Information-theoretic co-clustering (2003, methodology): “We bootstrap at k = 2 by choosing initial word cluster distributions to be “maximally” far apart from each other [1, 6].”
A Fast Clustering-Based Feature Subset Selection Algorithm for High-Dimensional Data (2013, methodology): “[18] proposed a new information-theoretic divisive algorithm for word clustering and applied it to text classification.”
Divergence Estimation for Multidimensional Densities Via $k$-Nearest-Neighbor Distances (2009, methodology): “…divergence has been used for detecting changes or testing stationarity in Internet measurements [6], [21], and for classification purposes in speech recognition [30], image registration [15], [29], text/multimedia clustering [11], and in a general classification framework called KLBoosting [26].”
Estimation of Information Theoretic Measures for Continuous Random Variables (2008, background): “There are other relevant applications in different research areas in which divergence estimation is used to measure the difference between two density functions, such as multimedia [19] and text [13] classification, among others.”
Using an Ensemble of One-Class SVM Classifiers to Harden Payload-based Anomaly Detection Systems (2006, methodology): “Accuracy and hardness of evasion are obtained by constructing our anomaly-based IDS using an ensemble of one-class SVM classifiers that work on different feature spaces.”
Wikipedia-based Semantic Interpretation for Natural Language Processing (2009, methodology): “Several studies performed feature construction using the WordNet electronic dictionary (Fellbaum, 1998) and other domain-specific dictionaries (Scott, 1998; Scott & Matwin, 1999; Urena-Lopez, Buenaga, & Gomez, 2001; Wang, McKay, Abbass, & Barlow, 2003; Bloehdorn & Hotho, 2004). Scott and Matwin (199”
Clustering-Based Hyperspectral Band Selection Using Information Measures (2007, methodology): “• Exploits band correlation through a clustering-based algorithm.”
McPAD: A multiple classifier system for accurate payload-based anomaly detection (2009, methodology): “…classify as normal only those payloads that have a very low anomaly score (as computed by PAYL itself), and for which we are therefore very conﬁdent that they do not carry an attack (as we can see from Figure 5, 7, 9, and 11, PAYL has a high detection rate at false positive rates higher than 1%).”
Semantic Kernel Forests from Multiple Taxonomies (2012, background): “, text and images, or speech and video; basic approaches include recovering the underlying shared latent space for both views [22, 20], bootstrapping classifiers formed independently per feature space [23, 24], or accounting for the view dependencies during clustering [25, 26].”
Data clustering: 50 years beyond K-means (2008, unknown): “Co-clustering is most popular in the field of bioinformatics, especially in gene clustering, and has also been successfully applied to document clustering (Slonim and Tishby, 2000; Dhillon et al., 2003).”
Feature selection in machine learning: A new perspective (2018, methodology): “Dhillon [89] proposed a global criterion for feature/word clustering, and presented a fast, divisive method for text classification.”