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 5
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
- EP ID EP6997
- DOI -
- Views 310
- Downloads 18
Effect of KM initiatives on productivity Payme NoorHamedan university personnel
In the development literature, knowledge-based development is of great importance for experts to complete such a term is known. Knowledge-based development, the engine of the economy, science and the production of knowle...
Determine Of the Exercise Intensity That Elicits Maximal Fat Oxidation In Untrained Male Students
Fat and carbohydrate are the main sources of energy for consumption during rest, exercise and training activities, and they are related to factors like duration and intensity of exercise. The present study was designed t...
Quality Of +10 Police Offices' Services in Iran
This study aimed to analyze the compatibility of +10 police offices' activities with EFQM model. The indices of organizational EFQM excellence model were used in this research. This study included one main hypothesis and...
Using correlation and some genetics methods to study of morphological traits in corn (Zea mays L.) yield and yield components under drought stress condition
In order to study the effect of drought stress on morphological characteristics, yield and yield components of 18 new hybrids of corn (Zea maize L.) resistant to drought stress and warm (which were bred and screened in K...
A Clinical Applicable Optimal Control of HIV Using Discretization Scheme
Generally, to obtain a realistic optimal treatment strategy for Human Immunodeficiency Virus (HIV) infection, the amount of virus particles should be decreased and the amount of uninfected CD4+ Tcells must be increased....