Semi-supervised information-maximization clustering

DOI10.1016/J.NEUNET.2014.05.016zbMATH Open1325.68207DBLPjournals/nn/CalandrielloNS14arXiv1304.8020OpenAlexW2170667765WikidataQ51072383 ScholiaQ51072383MaRDI QIDQ889291FDOQ889291

Authors: Daniele Calandriello, Gang Niu, Masashi Sugiyama 

Publication date: 6 November 2015

Published in: Neural Networks (Search for Journal in Brave)

Abstract: Semi-supervised clustering aims to introduce prior knowledge in the decision process of a clustering algorithm. In this paper, we propose a novel semi-supervised clustering algorithm based on the information-maximization principle. The proposed method is an extension of a previous unsupervised information-maximization clustering algorithm based on squared-loss mutual information to effectively incorporate must-links and cannot-links. The proposed method is computationally efficient because the clustering solution can be obtained analytically via eigendecomposition. Furthermore, the proposed method allows systematic optimization of tuning parameters such as the kernel width, given the degree of belief in the must-links and cannot-links. The usefulness of the proposed method is demonstrated through experiments.

Full work available at URL: https://arxiv.org/abs/1304.8020

Recommendations

• Semi-supervised clustering with discriminative random fields
• Semi-supervised graph clustering: a kernel approach
• Semi-supervised clustering via two-level random walk
• A semi-supervised fuzzy clustering algorithm applied to gene expression data
• Information-maximization clustering based on squared-loss mutual information

zbMATH Keywords

clusteringsemi-supervisedsquared-loss mutual informationinformation maximization

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Learning and adaptive systems in artificial intelligence (68T05) Pattern recognition, speech recognition (68T10)

Cites Work

• On Information and Sufficiency
• A Mathematical Theory of Communication
• Title not available (Why is that?)
• Title not available (Why is that?)
• Feature discovery in non-metric pairwise data
• Robust and efficient estimation by minimising a density power divergence
• Title not available (Why is that?)
• Density ratio estimation in machine learning. Foreword by Thomas G. Dietterich
• Sufficient Dimension Reduction via Squared-Loss Mutual Information Estimation
• Machine learning with squared-loss mutual information
• Information-Maximization Clustering Based on Squared-Loss Mutual Information

Cited In (3)

• Semi-supervised cross-entropy clustering with information bottleneck constraint
• A semi-supervised fuzzy clustering algorithm applied to gene expression data
• Triply stochastic gradient method for large-scale nonlinear similar unlabeled classification

This page was built for publication: Semi-supervised information-maximization clustering

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q889291)