SCHOLARLY PUBLICATION

Comparing clusterings

Marina Meilă

📖 International Conference on Machine Learning → 📅 2005-01-01 🔗 DOI: 10.1145/1102351.1102424

View Journal Profile → View Official Paper ↗

📄 Abstract

This paper views clusterings as elements of a lattice. Distances between clusterings are analyzed in their relationship to the lattice. From this vantage point, we first give an axiomatic characterization of some criteria for comparing clusterings, including the variation of information and the unadjusted Rand index. Then we study other distances between partitions w.r.t these axioms and prove an impossibility result: there is no “sensible” criterion for comparing clusterings that is simultaneously (1) aligned with the lattice of partitions, (2) convexely additive, and (3) bounded.

📄 Abstract

📤 Share this page

🔬 Related Research Papers

Three new graphical models for statistical language modelling

Fast time series classification using numerosity reduction

Grassmann discriminant analysis

Pachinko allocation

Practical solutions to the problem of diagonal dominance in kernel document clustering

Learning structural SVMs with latent variables

Maximum margin planning

R 1 -PCA

Building high-level features using large scale unsupervised learning

Classification using discriminative restricted Boltzmann machines

A Co-training Approach for Multi-view Spectral Clustering

Listwise approach to learning to rank