Unsupervised document classification using sequential information maximization

By Noam Slonim, Nir Friedman, Naftali Tishby. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2002

319 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/12bb048034a2842b8197580606ea51021f154962

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

In Statistical and Nonlinear Physics

Citations in the corpus

A Survey of Text Classification Algorithms (2012, methodology): “For example, the method in [122] performs purely unsupervised clustering (with no knowledge of class labels), and then as a final step assigns all documents in the cluster to the dominant class label of that cluster (as an evaluation step for the unsupervised clustering process in terms of its abili”
Disambiguating Web appearances of people in a social network (2005, methodology): “Following Slonim et al. [18], in order to get closer to the global maximum of our objective function, at each iteration we perform a number of random restarts of Steps 1-2 and then of Steps 3-4.”
The WEKA data mining software: an update (2009, unknown): “•Sequential information bottleneck clusterer[ 29 ]: a clusterer that was implemented primarily for document clustering.”
Multivariate Information Bottleneck (2001, methodology): “As in the symmetric IB, we first randomly initialize Tp and optimize it using the original sIB algorithm (Slonim et al., 2002), such that I (Tp;W) ismaximized.”
Relation Extraction Using Label Propagation Based Semi-Supervised Learning (2006, methodology): “JS divergence has ever been used as distance measure for document clustering, which outperforms cosine similarity based document clustering (Slonim et al., 2002).”
Improving the quality of predictions using textual information in online user reviews (2013, methodology): “This method has been successfully used in document classification [31], unsupervised image clustering [14] and many other applications.”
The Information Bottleneck : Theory and Applications (2006, methodology): “…Bottleneck: Theory and Applications Thesis submitted for the degree “Doctor of Philosophy” Noam Slonim Submitted to the Senate of the Hebrew University in the year 2002 This work was carried out under the supervision of Prof. Naftali Tishby This thesis is dedicated to the memory of my father”
Mining Text Data (2012, methodology): “For example, the method in [122] performs purely unsupervised clustering (with no knowledge of class labels), and then as a final step assigns all documents in the cluster to the dominant class label of that cluster (as an evaluation step for the unsupervised clustering process in terms of its abili”
PAC-Bayesian Analysis of Co-clustering and Beyond (2010, methodology): “Instead,G(Q,β) can be minimized (with respect toQ) using sequential minimization (Slonim et al., 2002; Dhillon et al., 2003).”
A Survey of Text Clustering Algorithms (2012, methodology): “A number of similar methods along this spirit are proposed in [4, 14, 35, 47, 89] with varying levels of supervision in the clustering process.”
Video search reranking via information bottleneck principle (2006, methodology): “Furthermore, their experiment was limited to image queries of simple ob jects such as bottles, cars, etc., instead of natural language queries as those in TRECVID.”