Semra Erbaş - Academia.edu (original) (raw)

Papers by Semra Erbaş

Kumeleme analizi hem gozetimsiz ogrenme olarak veri madenciliginde hem de veri seti altinda yatan... more Kumeleme analizi hem gozetimsiz ogrenme olarak veri madenciliginde hem de veri seti altinda yatan dogal gruplari ortaya cikaran cok degiskenli istatistiksel bir yontem olarak istatistikte yaygin olarak kullanilmaktadir. Ancak , ikili ortusmeden oturu dogal bir heterojenlik temsili ortaya cikaran sonlu karma modellerine iliskin homojen grup sayisini belirlemek zor bir islemdir. Bu calismada, grup homojenligi acisindan sonlu karma modellerden biri olan Gauss karma modelleri ele alinmistir. Bu amacla , Gauss karma bilesenlerin dogru siniflama oranlarini degerlendirmek ve homojen gruplari elde etmede bilesenlerin daha fazla bolunmesinin gerekli olup olmadigina karar vermek icin birlestirilmis kumeleme ve lineer diskriminant analizi ile birlestirilmis kumeleme ve karesel diskriminant analizi karsilastirilmistir.

Toplum ve sosyal hizmet, Oct 15, 2007

ABSTRACT: The aim in analyzing a 2k factorial design is to estimate the 2k main and interaction e... more ABSTRACT: The aim in analyzing a 2k factorial design is to estimate the 2k main and interaction effects. If some of these main and interaction effects are known to be zero or negligible, it is not necessary to estimate all the main and interaction effects in 2k factorial design. When S main and interaction effects are non-zero, all possible sets of S treatment combinations are not sufficient for estimating these main and interaction effects. For this reason, a method is introduced to obtain the smallest set of the S treatment combinations. In this study, two smallest sets are obtained for all possible scenarios of interest for 23 factorial design using this method given by Tsao and Wibowo. An illustration of this method is solved for 23 factorial design by using SPSS 13.0 package program.

The rate of incidence of suicide gradually increases especially in adolescence period. There are ... more The rate of incidence of suicide gradually increases especially in adolescence period. There are a lot of factors affecting on suicidal ideation. Social environment plays a significant role as potential risk factor along with biological disposition and psychiatric disorder. The aim of this study is to determine risk factors related with family, school and work affecting on suicidal ideation among adolescents in Turkey. In the study 6747 adolescents between the age of 12-18, who live in rural and urban area and are conducted face to face survey, were evaluated considering Research on Adolescent Profile in Turkey (TEPA 2013). Participants were selected through multistage, stratified, random sampling. (%50 female and %50 male). Pearson chi-square, odds ratio, correspondence analysis, independent samples t-test and binary logistic regression were used in the study. The relation between suicidal ideation and several factors such as cigarette and alcohol usel, drug addiction, being expose...

The aim in analyzing a 2k factorial design is to estimate the 2k main and interaction effects. If... more The aim in analyzing a 2k factorial design is to estimate the 2k main and interaction effects. If some of these main and interaction effects are known to be zero or negligible, it is not necessary to estimate all the main and interaction effects in 2k factorial design. When S main and interaction effects are non-zero, all possible sets of S treatment combinations are not sufficient for estimating these main and interaction effects. For this reason, a method is introduced to obtain the smallest set of the S treatment combinations. In this study, two smallest sets are obtained for all possible scenarios of interest for 23 factorial design using this method given by Tsao and Wibowo. An illustration of this method is solved for 23 factorial design by using SPSS 13.0 package program.

Bir 2k faktoriyel duzeni analiz etmekteki amac, 2k tane etki ve etkilesimi tahmin etmektir. Bu et... more Bir 2k faktoriyel duzeni analiz etmekteki amac, 2k tane etki ve etkilesimi tahmin etmektir. Bu etki ve etkilesimlerin bazilari sifir veya onemsiz olarak biliniyorsa, 2k faktoriyel duzende yer alan tum etki ve etkilesimi tahmin etmek gerekli degildir. Genelde, S tane etki ve etkilesim sifirdan farkli oldugu zaman, sadece S tane deneme kombinasyonu, bu etki ve etkilesimlerin tahmini icin gereklidir. S tane etki ve etkilesim sifirdan farkli oldugu zaman, S tane deneme kombinasyonun mumkun her kumesi, bu etki ve etkilesimlerin tahmini icin yeterli degildir. Bu nedenle, S tane deneme kombinasyonlarinin en kucuk kumesini elde etmek icin bir yontem tanitilmistir. Bu calismada, tanitilan bu yontem ile 23 faktoriyel duzenin tum mumkun durumlari icin iki tane en kucuk kume elde edilmistir. Ayrica, bu kumelerin elde edilmesinde yararlanilan dogrusal programlama modelinin cozumunde WINQSB paket programi kullanilmistir

gazi university journal of science, 2017

Our aim is to evaluate parameter estimation of two parameters item response theory (2-PL IRT) mod... more Our aim is to evaluate parameter estimation of two parameters item response theory (2-PL IRT) model using Joint Maximum Likelihood (JML). A simulation study is approached in terms of different sample sizes, number of items and levels of ability parameters for different level of discirimination parameter. As a result of this simulation study examinee ability parameter, item discrimination and difficulty parameters are obtained as well as Test Information Function and Point-biserial Correlation. One of the highlighted results shows that the level of discrimination parameter plays an important role in parameter estimation for 2-PL IRT models.

gazi university journal of science, 2012

This article presents tests for interaction in a two way table with one observation per cell. The... more This article presents tests for interaction in a two way table with one observation per cell. The conventional linear model theory cannot be used to check for interaction when there is exactly one observation per cell (no replication) in two way ANOVA models. For this, two different approaches are presented in the literature. In the first of these approaches is to assume a specific functional form for the interaction terms. In the second approach is not to assume a specific functional form for the interaction terms. In this article, we present four additivity tests by proposed Tusell (1990), Boik (1993a), Piepho (1994), Kharrati-Kopaei and S adooghi-Alvandi (2007) for second approach. We compared their performance by means of simulation studies with respect to the power. This article presents tests for the interaction in a two way table with one observation per cell. The conventional linear model theory cannot be used to check for the interaction when there is exactly one observatio...

gazi university journal of science, 2014

Normal 0 21 false false false TR X-NONE X-NONE Zero partial association models correspond to the ... more Normal 0 21 false false false TR X-NONE X-NONE Zero partial association models correspond to the conditional independence relation of a variable pair, given the rest of variables in the model. For determining the model that best fits the data from the contingency tables measured at a nominal level, this study has used the C (CATANOVA) test statistic (with response as one of its variables, and factor as another one of its variables) instead of the chi-square statistic, which is used as the test statistic in Wermuth’s backward elimination method with zero partial associations. Numerical analyses were performed on two samples, and the associations between these statistics were evaluated. Interpretations were provided for the obtained results.

General Pharmacology: The Vascular System, 1993

Taurine is found as a free amino acid in plasma and many tissues such as heart, muscle, brain and... more Taurine is found as a free amino acid in plasma and many tissues such as heart, muscle, brain and blood. Its exact role is not fully defined but it appears to have an important effect on the function of the heart. 2. The purpose of this study was to investigate the effects of taurine (10-3 M and 10-2 M) on malondialdehyde levels in perfused heart muscle. 3. In this study we found that administrated taurine before anoxia and repeffusion increased the malondialdehyde levels but administrated taurine after anoxia decreased the malondialdehyde levels in perfused heart muscle compared to the taurine administrated group before anoxia. 4. Anoxia did not change the perfused heart muscle malondialdehyde levels.

PROMET - Traffic&Transportation, 2012

Log-linear modelling is advanced as a procedure to identify factors that underlie the relative fr... more Log-linear modelling is advanced as a procedure to identify factors that underlie the relative frequency of occurrence of various characteristics. The purpose of this study is to present a modelling effort using log-linear models to estimate the relationships between driver’s fault and carelessness and the traffic variables such as gender, accident severity, and accident time. The study was conducted in four different districts in Ankara, the capital of Turkey. There were 1,325 people selected for the study; and they were asked whether they had been in an accident. Four hundred and forty-eight of them answered that they had been involved in an accident. As drivers, 276 out of 448 people, namely 61.6%, had traffic accidents. The data on the variables, namely gender, driver’s fault and carelessness, accident severity and accident time, were collected through a questionnaire survey. Detailed information has been created based on this information. The analysis showed that the best-fit m...

Communications in Statistics - Simulation and Computation

Acta Odontologica Turcica, 1986

Turkiye Klinikleri Journal of Biostatistics, 2012

Communications in Statistics - Theory and Methods