An acceleration of backpropagation algorithm with momentum (BPM) is introduced. At every stage of the learning process, local quadratic approximation of the error function is performed and the Hessian matrix of the quadratic function is approximated. Effective learning rate and momentum factor are determined by means of maximum and minimum eigenvalues of the approximated Hessian matrix at each step. BPM algorithm is modified so as to work automatically with these effective parameters. Performance of this new approach is demonstrated in comparison with well-known training algorithms on conventional problems by an experimental evaluation.

Public investments are directed to the social and economic areas. The state must carry public investments and services to every corner of the country, such as transportation, sewage, energy, education and health care by considering social stability and meeting human needs. Otherwise, unrest and disobedience may start to be experienced in the under developed regions and migration begins towards the developed areas in which the infrastructure services are completed. The regional insufficiency in public investments creates political, social and economic imbalances and instability. Differences in physical, social and geographical conditions between the regions of a country are generally based on the problems such as, the uneven distribution of the population, the differences in agricultural production and dissemination of industrial activities only in certain regions. The regions which are close to the national market are generally known to be more developed due to the use of market advantage. Transportation infrastructure and access to the national market affect the economic performance of the regions. In all provinces of Turkey, Anadolu University Open Education System has an important place by providing open and distance learning services. In this study, the relationship between the demands of open education system is investigated both with public investments and national market accessibility index respectively. For this purpose, different regression models were established and these models were evaluated statistically. In conclusion, the presence of linear positive correlation was determined.

In this study, it is aimed to determine the factors affecting the participation of women in the labor force in Turkey and to model the labor force participation rate with different statistical methods. For this purpose, 12 regions included in Statistical Region Units Classification (NUTS) Level 1 for the period of 2015-2020 were taken into consideration. Then, a one-way random effect panel data model was constructed by taking into account the variables of the labor force participation rate, employment rate, unemployment rate and education level, which are among the most important factors affecting women's labor force participation. Then, it is aimed to reveal the nonlinear effects of these determinants. For this purpose, considering the period of 1988-2019, additive and semiparametrik additive models were constructed by using the natural cubic spline for the effect of each determinant on the labor force participation rate with the roughness penalty approach. The constructed models were compared using adjusted Rsquare, deviation, generalized cross validity and Akaike information criterion, and the most suitable model was selected. These analyzes show that the unemployment rate, employment rate and population variables also have non-linear effects on the female labor force participation rate.

The aim of this study is to put forward how the demand for Anadolu University Open Education System is affected by socioeconomic factors in the provinces of Turkey. We tried to construct a structural equality model which is called as path analysis. First, factor analysis was applied to socioeconomic variables which were obtained in province basis and three important f actors were found. These are the levels of economic development, labor and education in the province. Then, multiple regression analysis was performed between these factors and the variables of Anadolu University Open Education System. As the result of the analysis, we found that economic development and education variables positively affect the demand for Anadolu University Open Education System while labor variables affect negatively. Thus, we have shown the positive contributions of economic development and education levels of the province to the demand for Open Education System. It is not surprising that individuals in the workforce do not benefit from open education. Because, individuals after increasing their skills through open education, generally le ave education and join the workforce, with the intention of finding a new job or new initiatives to be employed.