On comparing partitions

By M. Cugmas, A. Ferligoj. 2015

5,287 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Object-Centric Learning with Slot Attention (2020, methodology): “For a detailed description of the ARI score, see the aforementioned repository.”
Weakly Supervised Action Labeling in Videos under Ordering Constraints (2014, methodology): “Yet another way to evaluate is to use a loss between partitions [14] as in [3].”
Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions (2018, background): “Moreover, in order to validate that each component captures only a single ball (and thus compositionality is achieved), we report the Adjusted Rand Index (ARI; Hubert & Arabie (1985)) score between the soft-assignments γ and the ground-truth assignment of pixels to objects.”
Continuous Scene Representations for Embodied AI (2022, methodology): “For each video, we measure the Adjusted Rand Index (ARI) [27], a common clustering accuracy metric, between the CSR and GT assignments.”
An extensive comparative study of cluster validity indices (2013, methodology): “In order to avoid the possible bias introduced by the selection of a particular partition similarity measure, we replicated all the experiments using three partition similarity measures: Adjusted Rand [31], Jaccard [41] and Variation of Information [42] .”
Statistical clustering of temporal networks through a dynamic stochastic block model (2015, methodology): “We rely on the adjusted Rand index (ARI Hubert and Arabie, 1985) to evaluate the agreement between the estimated and the true latent structure. This index is smaller than 1, two identical latent structures (up to label switching) having an ARI equal to 1. Note that it can take negative values and is”
A Survey of Clustering Algorithms for Big Data: Taxonomy and Empirical Analysis (2014, background): “In general, finding an optimal number of clusters is an ill-posed problem of crucial relevance in clustering analysis [20].”
Community detection in networks: A user guide (2016, methodology): “The Omega index [89] is an extension of the Adjusted Rand index [76].”
How to find an appropriate clustering for mixed‐type variables with application to socio‐economic stratification (2013, methodology): “In order to do this, the clustering methods were applied to all datasets with one variable at a time left out, and the adjusted Rand index (ARI; Hubert and Arabie, 1985) was computed between the resulting partition and the one obtained on the full dataset.”
A Nonparametric Approach for Multiple Change Point Analysis of Multivariate Data (2013, methodology): “Each simulation applies these methods to a set of 1,000 independent sequences with two change points, and computes the average Rand index (Fowlkes and Mallows 1983; Hubert and Arabie 1985), defined below, and approximate standard errors.”
Comprehensive survey on hierarchical clustering algorithms and the recent developments (2022, methodology): “Some internal evaluation indices commonly used are shown in Table 4 (Hubert and Arabie 1985).”
Kubric: A scalable dataset generator (2022, background): “Object discovery – Object segmentation performance, measured in terms of foreground ARI [36, 42] (FG-ARI↑) in %.”
K-Profiles: A Nonlinear Clustering Method for Pattern Detection in High Dimensional Data (2015, methodology): “In Figure 2, the average ARI values were plotted against the noise level.”
Data-Driven Tree Transforms and Metrics (2017, methodology): “We compare the clustering of the samples relative to the reﬁned subtype labels [42] using three measures: the Rand index (RI) [47], the adjusted Rand index (ARI) [48], and the variation of information (VI) [49].”
Density-Based Clustering Validation (2014, methodology): “External clustering validity approaches such as the Adjusted Rand Index [19] compare clustering results with a pre-existing clustering (or class) structure, i.e., a ground truth solution.”
Analysis of Similarity Measures in Times Series Clustering for the Discovery of Building Energy Patterns (2013, background): “…of how the solution matches a pre-deﬁned structure based on a previous intuition concerning the data nature (e.g., the adjusted Rand index [24]); internal criteria , which evaluate the solution only considering the quantities and relationships of the vectors of the data set (e.g.,…”
A mixture of generalized hyperbolic distributions (2013, methodology): “Because we know the true classes, we can assess the performance of these mixture models in terms of classification accuracy, which we measure using the adjusted Rand index (ARI; Rand, 1971; Hubert and Arabie, 1985).”