Abstract:The objective of this study is to compare the predictive ability of Bayesian regularization with Levenberg-Marquardt Artificial Neural Networks. To examine the best architecture of neural networks, the model was tested with one-, two-, three-, four-, and five-neuron architectures, respectively. MATLAB (2011a) was used for analyzing the Bayesian regularization and Levenberg-Marquardt learning algorithms. It is concluded that the Bayesian regularization training algorithm shows better performance than the Levenberg-Marquardt algorithm. The advantage of a Bayesian regularization artificial neural network is its ability to reveal potentially complex relationships, meaning it can be used in quantitative studies to provide a robust model.

ABSTRACT

Abstract. It has been noticed that internet addiction scale which can apply and measure dependency level of individual couldn't reach at saturation level in national literature. So, it was aimed to adapt an internet addiction scale. Adapted scale was developed by Nichols and Nicki in 2004 and this study was sounded as "Internet Addiction Scale". The reliability cronbach alfa of original scale was obtained as .95 by Nichols and Nicki. In adapting study, cronbach alfa value was obtained as .93. The sample has been constituted from 277 students who belong to Yuzuncu Yil University and who visits periodically Computer Sciences Researches and Application Centre for surfing internet application. In this study, exploratory factor analysis was used to determine the validity of scale and after, confirmatory factor analysis was handled to test the fitness of the model. The main aim of this study is to adapt Internet Addiction Scale. Besides, the dependency level of sample was examined. According to the results, 32 people are consisted of the addiction risk group. As a result, the validity and the reliability findings are in optimal values which literature has been declared.

No abstract

No abstract