ARTICLE
TITLE

OPTIMIZATION OF INFORMATION PREPROCESSING IN CLUSTERING SYSTEMS OF HIGH DIMENSION DATA

SUMMARY

The methodic of choice of optimal normalization method for object cluster structure of creation, with high dimension of feature space, is shown. The Shannon entropy criterion and entropy relative change were used as main criterions of estimating the data preprocessing quality during the data transformation. Decreasing of feature space dimension of tested objects was realized by component analysis. Model of system clustering with the use of fuzzy C-means algorithm was constructed, which the help of whith the estimate of clustering quality was established by the use of different data preprocessing methods. It’s shown that the best normalization method for tested data is decimal-scaling method, by which the entropy of processed signal gets minimal significance, and relative change of entropy doesn’t exceed permissible norms during the process of data transformation by component analysis.

PAGES
COLLECTIONS
Computing
Technology

 Articles related

A. S. Dovbysh,?. V. Shelehov,D. V. Prylepa    

A method of the emotional and mental person’s state recognition using facial image is considered. The fragments with eye and nose areas of the image are prompted for additional information obtaining. A forming of the input mathematical description o... see more


V.I. Levin    

Contex. In recent decades, in the civil and military spheres new information technologies are increasingly encountered based on newapproaches to describing various types of uncertainty. These technologies are widely used in engineering, economics, social... see more


A. S. Dovbysh,D. V. Velykodnyi,O. B. Protsenko,V. I. Zimovets    

Relevance. The actual task of increasing the functional efficiency of machine learning of the system of functional diagnosis of theelectric drive of a hoisting mine machine is solved.The specific objective of this study was to develop a method for the in... see more


O. V Bisikalo,T. V. Grischuk,V. V. Kovtun    

Context. The questions of adapting the convolution neural network classifier use in automatic speaker recognition system of critical use(ASRSCU) are considered. The research object is the individual features of the human speech process.Objective. Develop... see more


(1) Alwatben Batoul Rashed (Department of Information Technology, Qassim University, Saudi Arabia) (2) Hazlina Hamdan (Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Malaysia) (3) Nurfadhlina Mohd Sharef (Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Malaysia) (4) Md Nasir Sulaiman (Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Malaysia) (5) Razali Yaakob (Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Malaysia) (6) Mansir Abubakar (Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Malaysia)    

Clustering, an unsupervised method of grouping sets of data, is used as a solution technique in various ?elds to divide and restructure data to become more signi?cant and transform them into more useful information. Generally, clustering is dif?cult and ... see more