Evaluation of the Compound clustering algorithms based on a subset of the preliminary results

Journal Title: International Research Journal of Applied and Basic Sciences - Year 2015, Vol 9, Issue 6

Abstract

Clustering is the process of partitioning a set of data objects into subsets. Each subset is a cluster, such that objects in a cluster are similar to one another, yet dissimilar to objects in other clusters. Because a cluster is a collection of data objects that are similar to one another within the cluster and dissimilar to objects in other clusters, a cluster of data objects can be treated as an implicit class. Clustering is known as unsupervised learning because the class label information is not present. Different clustering methods may generate different clusters on the same data set. Compound clustering usually gives better results than simple algorithms because in this method different clustering algorithms with different number of runs are used. It has shown that compound methods have better results than simple clustering algorithms, but sometimes it cannot be achieved accurately. In this method, each of the clustering algorithms emphasis different aspects of data in various ways. So we want to develop methods to combine the strength of each algorithm. In fact, the main reason for the compound (Hybrid) clustering is to search the best clusters by using the combination of other algorithm results .This paper, reviews recent research in the area of compound clustering based on a subset of the preliminary results.

Authors and Affiliations

Ali Heydarzadegan| Department of Computer Engineering,Yasooj Branch, Islamic Azad University, Yasooj, Iran, Seyed Ali Hosseini Ghohroodi*| Department of Computer Engineering,Yasooj Branch, Islamic Azad University, Yasooj, Iran, email: ali.hosseini.gh@gmail.com, Farhad Raad| Department of Computer Engineering,Yasooj Branch, Islamic Azad University, Yasooj, Iran

Keywords

Related Articles

The relations between Job Satisfaction dimensions with Organizational Commitment of personnel

The general goal of this plan is finding any relations among different dimensions of job satisfaction and organizational commitment of personnel at saman cement company. Specific goals of this project are follows: Recogn...

Effects of thiamethoxam in sublethal concentrations, on life expectancy (ex) and some other biological characteristics of Hippodamia variegata (Goeze) (Coleoptera: Coccinellidae)

Hippodamia variegata Goeze (Coleoptera: Coccinellidae), is an important aphidophagous predator that is noticeable in IPM as a suitable biological control agent. In this study, influence of sublethal concentrations (LC10...

On Linear Regression and Correlation Coefficient for Breast Cancer in Al-Sader Medical City in 2005-2009

The main propose of this study is to evaluate the Breast Cancer in Al-Sader Medical City in AL-Najaf in Iraq for the period 2005-2009 for two variables age and Tumors. The data is analyzed using two methods regression an...

Influence of relational capabilities on Business performance Case of: Kermanshah industrial city SMEs

In today's dynamic business environment, due to the emergence of three influential trends including severe international competition, discrete markets and changing technology, the notion of firm performance is discussed...

The Relationship between Forward Shoulder Posture and Self Esteem in Non-athlete Female Students of Tehran University

The objective of this study was to investigate the relationship between forward shoulder and self esteem in female students. In order to do so, researchers used Coopersmith self esteem inventory (questionnaire) as well a...

Download PDF file
  • EP ID EP7020
  • DOI -
  • Views 343
  • Downloads 15

How To Cite

Ali Heydarzadegan, Seyed Ali Hosseini Ghohroodi*, Farhad Raad (2015). Evaluation of the Compound clustering algorithms based on a subset of the preliminary results. International Research Journal of Applied and Basic Sciences, 9(6), 909-912. https://europub.co.uk/articles/-A-7020