Page Clustering Using a Distance-Based Algorithm

Text Mining is a field that is considered as an extension of data mining. In the context of text mining, document clustering is used to set apart likewise documents of a collection into the identical category, called cluster, and divergent documents to distinctive groups. Since every dataset has its own characteristics, finding an appropriate clustering algorithm that can manage all kinds of clusters, is a big challenge. Clustering algorithms has theirs unique approaches for computing the number of clusters, imposing a structure on the data, and attesting the out coming clusters. The idea of combining different clustering is an effort to overwhelm the faults of single algorithms and further enhance their executions. On the other hand, inspired by the gravitational law, different clustering algorithms have been introduced that each one attempted to cluster complex datasets. Gravitational Ensemble Clustering (GEC) is an ensemble method that employs both the concepts of gravitational clustering and ensemble clustering to reach a better clustering result. This paper represents an application of GEC to the problem of document clustering. The proposed method uses a modification of the original GEC algorithm. This modification tries to produce a more varied clustering ensemble using new parameter setting. The GEC algorithm is assessed using document datasets. Promising results of the presented method were obtained in comparison with competing algorithms.

show abstract

“…Mojica et al [17] used a modification of method, proposed by Gomez [14] for web page clustering in dynamic environments.…”

Section: B Gravitational Clusteringmentioning

confidence: 99%

Document clustering using gravitational ensemble clustering

Sadeghian

Nezamabadi‐pour

2015

2015 the International Symposium on Artificial Intelligence and Signal Processing (AISP)

View full text Add to dashboard Cite

show abstract

“…Cluster content Clustering algorithm Methodology Cadez et al (2002) and Pallis et al (2007) Web users Partitional hard Model based Perkowitz and Etzioni (1998) Web pages Partitional hard Similarity based Petridou et al (2008) Web users Partitional hard Similarity based Shokry et al (2006) and Suryavanshi et al (2005) Web pages Partitional fuzzy Similarity based Mojica et al (2005) Web pages Hierarchical hard Similarity based Yang and Padmanabhan (2005) Web users Hierarchical hard Similarity based Lazzerini et al (2003) Web users Hierarchical fuzzy Similarity based Castellano et al (2007) Web users Partitional fuzzy Similarity based He et al (2002) and Huang et al (2006) Web pages Hard Spectral clustering Mobasher (1999) and Mobasher et al (2002) Web pages Fuzzy Similarity based Zeng et al (2002) Web users/pages Hard Model based Liu et al (2005) Web users/pages Fuzzy Similarity based…”

Section: Approachmentioning

confidence: 99%

A fuzzy bi-clustering approach to correlate web users and pages

Koutsonikola¹,

Vakali²

2009

IJKWI

View full text Add to dashboard Cite

With the rapid development of information technology, the significance of clustering in the process of delivering information to users is becoming more eminent. Especially in the web information space, clustering analysis can prove particularly beneficial for a variety of applications such as web personalisation and profiling, caching and prefetching and content delivery networks. In this paper, we propose a bi-clustering approach, which identifies groups of related web users and pages. The proposed approach is a three-step process that relies on the principles of spectral clustering analysis and provides a fuzzy relation scheme for the revealed users' and pages' clusters. Experiments have been conducted on both synthetic and real datasets to prove the proposed method's efficiency and reveal hidden knowledge.

show abstract

“…Cluster content Clustering algorithm Methodology Cadez et al (2002) Web users Partitional hard Model based Petridou et al (2008) Web users Partitional hard Similarity based Shokry et al (2006) Web pages Partitional fuzzy Similarity based Mojica et al (2005) Web pages Hierarchical hard Similarity based Yang and Padmanabhan (2005) Web users Hierarchical hard Similarity based Lazzerini et al (2003) Web users Hierarchical fuzzy Similarity based Castellano et al (2007) Web users Partitional fuzzy Similarity based Zeng et al (2002) Web users/pages Hard Model based Liu et al (2005) Web users/pages Fuzzy Similarity based Table I. An overview of clustering approaches IJWIS 5,3 measures such as Euclidean or Manhattan distances (hence "divergence" rather than "distance"), it is expected to be robust against noise.…”

Section: Approachmentioning

confidence: 99%

A new approach to web users clustering and validation: a divergence‐based scheme

Koutsonikola

Petridou

Vakali

et al. 2009

International Journal of Web Information Systems

View full text Add to dashboard Cite

Purpose-Web users' clustering is an important mining task since it contributes in identifying usage patterns, a beneficial task for a wide range of applications that rely on the web. The purpose of this paper is to examine the usage of Kullback-Leibler (KL) divergence, an information theoretic distance, as an alternative option for measuring distances in web users clustering. Design/methodology/approach-KL-divergence is compared with other well-known distance measures and clustering results are evaluated using a criterion function, validity indices, and graphical representations. Furthermore, the impact of noise (i.e. occasional or mistaken page visits) is evaluated, since it is imperative to assess whether a clustering process exhibits tolerance in noisy environments such as the web. Findings-The proposed KL clustering approach is of similar performance when compared with other distance measures under both synthetic and real data workloads. Moreover, imposing extra noise on real data, the approach shows minimum deterioration among most of the other conventional distance measures. Practical implications-The experimental results show that a probabilistic measure such as KL-divergence has proven to be quite efficient in noisy environments and thus constitute a good alternative, the web users clustering problem. Originality/value-This work is inspired by the usage of divergence in clustering of biological data and it is introduced by the authors in the area of web clustering. According to the experimental results presented in this paper, KL-divergence can be considered as a good alternative for measuring distances in noisy environments such as the web.

show abstract

Page Clustering Using a Distance-Based Algorithm

Cited by 4 publications

References 14 publications

Document clustering using gravitational ensemble clustering

Document clustering using gravitational ensemble clustering

A fuzzy bi-clustering approach to correlate web users and pages

A new approach to web users clustering and validation: a divergence‐based scheme

Contact Info

Product

Resources

About