Clustering categorical data using a swarm-based method

Izakian, Hesam; Abraham, Ajith; Snášel, Václav

doi:10.1109/nabic.2009.5393623

Cited by 7 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Distance is either entropy [41] or Hamming distance [43]. Other algorithms have been proposed to improve optimization of the k-modes objective function without repeated reinitialization, including a tabu search [50], a genetic algorithm [51], a partical swarm optimizer [52], an artificial bee colony optimizer [39] and a Cuckoo search [53]. Most global optimizers and some initializers employ the traditional k-modes algorithms within their approach, and hence could benefit from use of the OTQT or OT algorithms.…”

Section: B Resultsmentioning

confidence: 99%

An Efficient $k$-modes Algorithm for Clustering Categorical Datasets

Dorman,

Maitra

2020

Preprint

View full text Add to dashboard Cite

Mining clusters from datasets is an important endeavor in many applications. The k-means algorithm is a popular and efficient distribution-free approach for clustering numerical-valued data but can not be applied to categorical-valued observations. The k-modes algorithm addresses this lacuna by taking the k-means objective function, replacing the dissimilarity measure and using modes instead of means in the modified objective function. Unlike many other clustering algorithms, both k-modes and kmeans are scalable, because they do not require calculation of all pairwise dissimilarities. We provide a fast and computationally efficient implementation of k-modes, OTQT, and prove that it can find superior clusterings to existing algorithms. We also examine five initialization methods and three types of K-selection methods, many of them novel, and all appropriate for k-modes. By examining the performance on real and simulated datasets, we show that simple random initialization is the best intializer, a novel K-selection method is more accurate than two methods adapted from k-means, and that the new OTQT algorithm is more accurate and almost always faster than existing algorithms.

show abstract

Section: B Resultsmentioning

confidence: 99%

An Efficient $k$-modes Algorithm for Clustering Categorical Datasets

Dorman,

Maitra

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…, z ld } for 1 ≤ l ≤ k where k is the number of clusters. Assume X and Y are two categorical data objects in D, the simple matching distance measure between X and Y is defined as [31] :…”

Section: K-modes Algorithm(k-ma)mentioning

confidence: 99%

“…This algorithm identifies the modes using a frequency-based technique by solving the following mathematical problem [31].…”

Section: K-modes Algorithm(k-ma)mentioning

confidence: 99%

Using Clustering Methods to Detect the Revealed Preferences of Moroccans towards the Electric Vehicles: Latent Class Analysis (LCA) and K-Modes Algorithm (K-MA)

Harrouti¹,

Azhari²,

Deqqaq³

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Particle Swarm Optimization (PSO) is combined with kmodes clustering algorithm is proposed in [12]. PSO is one of the popular metaheuristic and swarm intelligence based optimization algorithms.…”

Section: Related Workmentioning

confidence: 99%

CLUSTERING CATEGORICAL DATA USING k-MODES BASED ON CUCKOO SEARCH OPTIMIZATION ALGORITHM

Lakshmi¹,

Visalakshi²,

Shanthi³

et al. 2017

IJSC

View full text Add to dashboard Cite

Cluster analysis is the unsupervised learning technique that finds the interesting patterns in the data objects without knowing class labels. Most of the real world dataset consists of categorical data. For example, social media analysis may have the categorical data like the gender as male or female. The k-modes clustering algorithm is the most widely used to group the categorical data, because it is easy to implement and efficient to handle the large amount of data. However, due to its random selection of initial centroids, it provides the local optimum solution. There are number of optimization algorithms are developed to obtain global optimum solution. Cuckoo Search algorithm is the population based metaheuristic optimization algorithms to provide the global optimum solution. Methods: In this paper, k-modes clustering algorithm is combined with Cuckoo Search algorithm to obtain the global optimum solution. Results: Experiments are conducted with benchmark datasets and the results are compared with k-modes and Particle Swarm Optimization with k-modes to prove the efficiency of the proposed algorithm.

show abstract

Clustering categorical data using a swarm-based method

Cited by 7 publications

References 7 publications

An Efficient $k$-modes Algorithm for Clustering Categorical Datasets

An Efficient $k$-modes Algorithm for Clustering Categorical Datasets

Using Clustering Methods to Detect the Revealed Preferences of Moroccans towards the Electric Vehicles: Latent Class Analysis (LCA) and K-Modes Algorithm (K-MA)

CLUSTERING CATEGORICAL DATA USING k-MODES BASED ON CUCKOO SEARCH OPTIMIZATION ALGORITHM

Contact Info

Product

Resources

About