Categorical data clustering: What similarity measure to recommend?

Santos, Thauana Mericcy Mota; Zárate, Luis Enrique

doi:10.1016/j.eswa.2014.09.012

Cited by 75 publications

(37 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Further more, pairwise vertex similarity is a fundamental index for many network functions and physical systems [53,54]. That is to say, the proposed similarity index can find applications in solving many network-related problems, such as link predication [55,56], community detection [57,58,59], spreading activation [60], network evolution [61,62], web searching [63,64], data clustering [65,66], and gene ranking [67]. By contrast, it would be hard for the hybrid transition matrix to be applied to solve these problems.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

A vertex similarity index for better personalized recommendation

Chen

Zhang

Lü

et al. 2017

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

Recommender systems benefit us in tackling the problem of information overload by predicting our potential choices among diverse niche objects. So far, a variety of personalized recommendation algorithms have been proposed and most of them are based on similarities, such as collaborative filtering and mass diffusion. Here, we propose a novel vertex similarity index named CosRA, which combines advantages of both the cosine index and the resource-allocation (RA) index. By applying the CosRA index to real recommender systems including MovieLens, Netflix and RYM, we show that the CosRA-based method has better performance in accuracy, diversity and novelty than some benchmark methods. Moreover, the CosRA index is free of parameters, which is a significant advantage in real applications. Further experiments show that the introduction of two turnable parameters cannot remarkably improve the overall performance of the CosRA index.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

A vertex similarity index for better personalized recommendation

Chen

Zhang

Lü

et al. 2017

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

show abstract

“…Parameters that could be specified for the program were: photo file, number of clusters, maximum iterations number, distance measure and random number generator seed. The available distance measures were Euclidean, Manhattan metrics and two versions of Gower coefficient [3] -regular (7,9) and modified (8,9). The modified version that takes specifics of RGB vector data into account was prepared for experiments in this paper.…”

Section: Results Overviewmentioning

confidence: 99%

Applying fuzzy clustering method to color image segmentation

Sakarya¹

2015

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-The goal of this paper was to apply fuzzy clustering algorithm known as Fuzzy C-Means to color image segmentation, which is an important problem in pattern recognition and computer vision. For computational experiments, serial and parallel versions were implemented. Both were tested using various parameters and random number generator seeds. Various distance measures were used: Euclidean, Manhattan metrics and two versions of Gower coefficient similarity measure. The F and Q segmentation evaluation measures and output images were used to assess the result of color segmentation. Serial and parallel run times were compared.

show abstract

“…(see e.g. [1,14,17,18,19,22]). A general problem is as follows: the set A should be partitioned into 1…”

Section: Data Clusteringmentioning

confidence: 99%

Research project grouping and ranking by using adaptive Mahalanobis clustering

Turkalj

Markulak

Singer

et al. 2016

Croat. Oper. Res. Rev.

View full text Add to dashboard Cite

Abstract. The paper discusses the problem of grouping and ranking of research projects submitted for a call. The projects are grouped into clusters based on the assessment obtained in the review procedure and by using the adaptive Mahalanobis clustering method as a special case of the Expectation Maximization algorithm. The cluster of projects assessed as best is specially analyzed and ranked. The paper outlines several possibilities for the use of data obtained in the review procedure, and the proposed method is illustrated with the example of internal research projects at the University of Osijek.

show abstract

Categorical data clustering: What similarity measure to recommend?

Cited by 75 publications

References 22 publications

A vertex similarity index for better personalized recommendation

A vertex similarity index for better personalized recommendation

Applying fuzzy clustering method to color image segmentation

Research project grouping and ranking by using adaptive Mahalanobis clustering

Contact Info

Product

Resources

About