An Incremental Clustering Algorithm Based on Compact Sets with Radius α

Pons-Porrata, Aurora; Díaz, Germán Ponce; Cortés, Manuel Lazo; Ramírez, Leydis Alfonso

doi:10.1007/11578079_54

Cited by 3 publications

(3 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the aforementioned work, we pointed out that the β 0 -connected components algorithm shows a tendency of finding excessively large, uncohesive clusters, whereas the β 0 -compact sets algorithm and the extended stars algorithm show a tendency of finding cohesive but excessively small clusters. Here, by using the radius-α-β 0 -compact sets algorithm, we obtain a tunable combination of both behaviors, which has been experimentally proven to outperform each individual behavior when appropriate values are given to the α parameter [16]. Two extreme cases may be pointed out: if α = 0 the algorithm behaves like the classic β 0 -compact sets algorithm; whereas for α ≥ β maxβ 0 , where β max is the maximum similarity value in the collection, the algorithm behaves like the classic β 0 -connected components algorithm.…”

Section: Methodsmentioning

confidence: 99%

Introducing Biases in Document Clustering

Ramírez-Cruz¹

2014

CyS

View full text Add to dashboard Cite

In this paper, we present three criteria for introducing biases in document clustering algorithms, when information characterizing the document collections is available. We focus on collections known to be the result of a document categorization or sample-based document filtering process. Our proposals rely on profiles, i.e., document samples known to have been used for obtaining the collection, to extract statistics which determine the biases to introduce. We conduct an experimental evaluation over a number of collections extracted from the widely used corpus RCV1, which allows us to confirm the validity of our proposals and determine a number of situations where biased clusterings, according to different criteria, outperform their unbiased counterparts. Introducción de sesgos en el agrupamiento de documentosResumen. En este artículo se presentan tres criterios para la introducción de sesgos en algoritmos de agrupamiento de documentos, cuando se dispone de información que caracteriza las colecciones de documentos. Nos concentramos en colecciones de las que se conoce que son el resultado de un proceso de categorización o filtrado de documentos basado en muestras. Nuestras propuestas utilizan perfiles, es decir muestras de documentos de las que se conoce que han sido utilizadas para obtener la colección, para extraer estadísticos que determinan los sesgos a introducir. Llevamos a cabo una evaluación experimental sobre un conjunto de colecciones extraídas del corpus ampliamente utilizado RCV1, que nos permiten confirmar la validez de nuestras propuestas y determinar un número de situaciones donde los agrupamientos sesgados según diferentes criterios superan a sus contrapartes no sesgadas.Palabras clave. Agrupamiento de documentos, introducción de sesgos.

show abstract

Section: Methodsmentioning

confidence: 99%

Introducing Biases in Document Clustering

Ramírez-Cruz¹

2014

CyS

View full text Add to dashboard Cite

show abstract

“…A parallel effort has been devoted to designing and developing incremental clustering algorithms which are able to process new data as they are added to the collection. In particular, incremental clustering algorithms (e.g., [6], [5], [7]) are able to update the clustering structure after insertion and/or deletion of new objects. Among the previous approaches, the aim of the work presented in [7] is to identify clusters of objects characterized only by categorical (i.e., not numerical) features.…”

Section: Related Workmentioning

confidence: 99%

“…In particular, incremental clustering algorithms (e.g., [6], [5], [7]) are able to update the clustering structure after insertion and/or deletion of new objects. Among the previous approaches, the aim of the work presented in [7] is to identify clusters of objects characterized only by categorical (i.e., not numerical) features. Thus, this approach cannot be exploited to analyze traffic network data.…”

Section: Related Workmentioning

confidence: 99%

NetCluster: A clustering-based framework to analyze internet passive measurements data

et al. 2013

View full text Add to dashboard Cite

Internet measured data collected via passive measurement are analyzed to obtain localization information on nodes by clustering (i.e., grouping together) nodes that exhibit similar network path properties. Since traditional clustering algorithms fail to correctly identify clusters of homogeneous nodes, we propose the NetCluster novel framework, suited to analyze Internet measurement datasets. We show that the proposed framework correctly analyzes synthetically generated traces. Finally, we apply it to real traces collected at the access link of Politecnico di Torino campus LAN and discuss the network characteristics as seen at the vantage point.

show abstract

Algoritmo de votación incremental INC-ALVOT para clasificación supervisada

Escobar-Franco,

Sánchez-Díaz

2013

Rev.Fac.Ing.Univ.Antioquia

View full text Add to dashboard Cite

En este trabajo, se presenta un algoritmo incremental para clasificación supervisada llamado INC-ALVOT (algoritmo de votación incremental). Este algoritmo permite manejar conjuntos de datos mezclados, los cuales no se almacenan en la memoria principal. Además, el algoritmo permite incorporar nuevos objetos en el conjunto de datos inicial, realizando un número mínimo de operaciones para la clasificación de nuevos objetos con el conjunto de datos expandido. Se presentan los resultados obtenidos al aplicar el algoritmo propuesto en diversos conjuntos de datos reales comparado con el algoritmo clásico de votación ALVOT.

show abstract

An Incremental Clustering Algorithm Based on Compact Sets with Radius α

Cited by 3 publications

References 3 publications

Introducing Biases in Document Clustering

Introducing Biases in Document Clustering

NetCluster: A clustering-based framework to analyze internet passive measurements data

Algoritmo de votación incremental INC-ALVOT para clasificación supervisada

Contact Info

Product

Resources

About