Browsing by Author "Arslan, Guvenc"

Now showing 1 - 4 of 4

Citation - WoS: 5
Development of Building Damage Functions for Big Earthquakes in Turkey
(Elsevier Science Bv, 2015) Fawzy, Diaa E.; Arslan, Guvenc
The current work is an attempt to predict building reactions to big earthquakes using real data collected from surveys carried out after the occurrence of earthquakes. With the development of building damage functions for big earthquakes in Turkey one can predict the damage levels as a function of earthquakes' intensity and the building parameters. Our model is based on neural networks techniques which allow for the non-linear correlations to be taken into account. We analyse data collected for damaged buildings after the following three big earthquakes: Afyon (2002; Mw - 6.0), Bingol (2003; Mw - 6.4) and Duzce (1999; Mw - 7.2). The current model includes some of the main important factors affecting the health of any structure, namely, age, number of stories, floor areas, and the column areas. Our method of damage prediction is based on several earthquakes and buildings with different damage levels. The obtained results show that there is a strong correlation between the strength of the earthquake, the basic building parameters and the damage level. The obtained building damage function is essential for future plans and regulations for new constructions and can be considered as an essential module for hazards mitigation systems. (C) 2015 The Authors. Published by Elsevier Ltd.
Citation - WoS: 5
Citation - Scopus: 4
A Fuzzy Bayesian Classifier With Learned Mahalanobis Distance
(Wiley, 2014) Kayaalp, Necla; Arslan, Guvenc
Recent developments show that naive Bayesian classifier (NBC) performs significantly better in applications, although it is based on the assumption that all attributes are independent of each other. However, in the NBC each variable has a finite number of values, which means that in large data sets NBC may not be so effective in classifications. For example, variables may take continuous values. To overcome this issue, many researchers used fuzzy naive Bayesian classification for partitioning the continuous values. On the other hand, the choice of the distance function is an important subject that should be taken into consideration in fuzzy partitioning or clustering. In this study, a new fuzzy Bayes classifier is proposed for numerical attributes without the independency assumption. To get high accuracy in classification, membership functions are constructed by using the fuzzy C-means clustering (FCM). The main objective of using FCM is to obtain membership functions directly from the data set instead of consulting to an expert. The proposed method is demonstrated on the basis of two well-known data sets from the literature, which consist of numerical attributes only. The results show that the proposed the fuzzy Bayes classification is at least comparable to other methods.
A Java Program for the Multivariate Z(p) and C-P Tests and Its Application
(Elsevier Science Bv, 2011) Arslan, Guvenc; Ozmen, Ilknur
The multivariate normality assumption is used in many multivariate statistical analyses. It is, therefore, important to assess the validity of this assumption. The main aim of this study is to develop a JAVA program for applying the recently developed Z(p) and C-p test statistics. The application and results of the program are illustrated on two real data sets. (C) 2010 Elsevier B.V. All rights reserved.
Citation - WoS: 3
Citation - Scopus: 5
A New Characterization of the Power Distribution
(Elsevier Science Bv, 2014) Arslan, Guvenc
A new characterization for the power function distribution is obtained which is based on products of order statistics. This result may be considered as a generalization of some recent results for contractions. The result is obtained by applying a new variant of the Choquet-Deny theorem. We note that in this new result the product consists of order statistics from independent samples. This characterization result may also be interpreted in terms of some special scheme of ranked set sampling. (C) 2013 Elsevier B.V. All rights reserved.