Similarity Measures for Categorical Data: A Comparative Evaluation

By Shyam Boriah, Varun Chandola, Vipin Kumar. SDM, 2008

662 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

View Paper: https://www.semanticscholar.org/paper/c6544a12fec2097bddc49adbb159426d9dc15d2c

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 Computational Theory and Mathematics

Citations in the corpus

Learning to Detect Patterns of Crime (2013, methodology): “This is a simplified version of Goodall’s measure [22].”
The k-means Algorithm: A Comprehensive Survey and Performance Evaluation (2020, background): “However, after a thorough experimental analysis in [71] on a number of datasets, the authors came to the conclusion that “No single measure is always superior or inferior.”
Coupled nominal similarity in unsupervised learning (2011, methodology): “Two attribute values are similar if they present analogous frequency distributions for one attribute [2]; this reflects the intra-coupled similarity within a feature.”
Distance based similarity models for content based multimedia retrieval (2013, methodology): “As we will see in the remainder of this chapter, some distance functions inherently utilize the opposing concept of a similarity measure [Santini and Jain, 1999, Boriah et al., 2008, Jones and Furnas, 1987].”
A Comparison Study on Similarity and Dissimilarity Measures in Clustering Continuous Data (2015, background): “Mahalanobis distance is a data-driven measure in contrast to Euclidean and Manhattan distances that are independent of the related dataset to which two data points belong [20,33].”
Unsupervised Machine Learning on a Hybrid Quantum Computer (2017, background): “A familiar choice for a distance measure is the Euclidean distance, but specific applications may naturally lead to very different choices for the metric and the distance [22, 23].”
Network Anomaly Detection: Methods, Systems and Tools (2014, background): “[50], categorical [51] and mixed type [52] data are listed in Table V.”
A Random Forest-Assisted Evolutionary Algorithm for Data-Driven Constrained Multiobjective Combinatorial Optimization of Trauma Systems (2020, methodology): “RF 100 CARTs that generated from the bootstrap samples of 2 ⌊√ n ⌋ randomly selected decision variables CART Splitting tolerance is set to 1e− 4 ∗σ2 (σ2 is the variance for the entire data) RBF n hidden nodes using the overlap measure [74] as the kernel function FS Top q decision variables with accu”
Nearest neighbor imputation algorithms: a critical evaluation (2016, methodology): “In the present paper we only focused on imputation problems with continuous or dichotomous variables, hence there was no need to consider other similarity measures for categorical or ordinal data as described in [16].”
Anomaly detection: A survey (2009, methodology): “For categorical attributes, simple matching coefficient is often used but more complex distance measures can be used [Boriah et al. 2008; Chandola et al. 2008].”
Outlier Analysis (2013, methodology): “A number of such measures are proposed in [42, 303, 411], and summarized very well in [75].”
Coupled Matrix Factorization Within Non-IID Context (2014, methodology): “ of the existing methods assume that the attributes are independently. However, in reality, the attributes are more or less interacted and coupled via explicit or implicit relationships [25] [26] [6] [5] [7]. In this paper, dierent from these methods trying to in uence the latent factors, we turn to”