To determine the number of clusters in the clustering analysis that has a broad range of applied sciences, such as physics, chemistry, biology, engineering, economics etc., many methods have been proposed in the literature. The aim of this paper is to determine the number of clusters of a dataset in a model-based clustering by using an Analytic Hierarchy Process (AHP). In this study, the AHP model has been created by using the information criteria Akaike’s Information Criterion (AIC), Approximate Weight of Evidence (AWE), Bayesian Information Criterion (BIC), Classification Likelihood Criterion (CLC), and Kullback Information Criterion (KIC). The achievement of the proposed approach has been tested on common real and synthetic datasets. The proposed approach based on the corresponding information criteria has produced accurate results. The currently produced results have been seen to be more accurate than those corresponding to the information criteria.

Clustering analysis based on a mixture of multivariate normal distributions is commonly used in the clustering of multidimensional data sets. Model selection is one of the most important problems in mixture cluster analysis based on the mixture of multivariate normal distributions. Model selection involves the determination of the number of components (clusters) and the selection of an appropriate covariance structure in the mixture cluster analysis. In this study, the efficiency of information criteria that are commonly used in model selection is examined. The effectiveness of information criteria has been determined according to the success in the selection of the number of components and in the selection of an appropriate covariance matrix.

Nowadays, data dimensions have increased depending on the developments in the information and measurement technologies. Due to the high dimensionality, it is necessary to use pre-analysis data reduction methods for many analyzes such as classification and regression analysis. In the solution of high dimensionality, filter feature selection methods based on statistical criteria are widely used in terms of simplicity and efficiency. One of the important problems with filter feature selection methods is to select multiple features carrying the same information unnecessarily when strong correlations exist between features. In this study, a novel approach is proposed to solve this problem of the filter feature selection methods. In addition, with the proposed new approach, the question of how many appropriate features will be is also solved. The performance of the proposed approach is demonstrated on high-dimensional reflectance data having high correlations between features. The obtained results revealed that the proposed approach improves the classification performance of filter feature selection methods in mixture discriminant analysis in terms of classification accuracy and entropy criteria.

Background: Obesity in women of reproductive age often causes hormonal imbalances and fertility issues. Bariatric surgery effectively promotes weight loss and metabolic improvement, but its impact on reproductive hormones warrants further investigation. This study examines bariatric surgery’s effects on hormonal and clinical changes in women preoperatively and at 3 and 6 months postoperatively. Methods: This prospective study included 67 women undergoing bariatric surgery. Anthropometric measurements, hormonal profiles, ovarian morphology(via ultrasonography), and the presence of hirsutism and dysmenorrhoea were evaluated preoperatively and at 3- and 6-month follow-ups. Results: Bariatric surgery resulted in significant weight loss and hormonal changes. Body weight (p < 0.001) and BMI (p < 0.001) decreased substantially, with the most rapid reduction occurring in the first 3 months, followed by a slower decline. Androstenedione levels decreased significantly (p < 0.001). Total testosterone levels showed a significant reduction at 6 months (p < 0.001), while SHBG (p < 0.001; p = 0.014), DHEA-S, and AMH (p < 0.001; p < 0.001; p = 0.014) levels increased. No significant changes were observed in FSH and LH levels (p > 0.05). The severity of dysmenorrhea decreased significantly at 6 months (p < 0.001). Additionally, the prevalence of hirsutism (p < 0.001) and polycystic ovary (PCO) morphology (p < 0.001) decreased significantly at 6 months. Conclusions: Bariatric surgery significantly decreases testosterone and androstenedione, while increasing AMH, DHEA-S, and SHBG. Improvements in PCO morphology, dysmenorrhea, and hirsutism highlight the effects of bariatric surgery on hormonal balance and reproductive health.

PISA 2015 mathematical literacy score of Turkey is 420 while the average score of all countries is 461. It is understood that; Turkish students’ PISA 2015 mathematical literacy score was lower than the average. The basic reasons for the below average score need to be truly examined and developmental activities should be revealed. The aim of this study is to classify students according to the factors affecting their mathematical literacy score and to reveal the effects of these factors in classification.The data of the study is obtained from 5895 students who participated in PISA 2015. In this study, we used Random Forest, Naïve Bayes Classifier, Logistic Regression, Decision Tree Algorithm and Discriminant Analysis as classifiers. According to the results, Random Forest method produced more accurate scores than other methods with 76.32% accuracy. We also calculated the correct classification rate and determined the factors that positively and negatively affect the classification with discriminant analysis. According to the discriminant analysis home possessions, information and computer technology resources at home and students' expected occupational status were the most positive effective variables on mathematical literacy score. On the other hand, family wealth possessions, student-related factors affecting school climate and anxiety have negative effect on mathematical literacy score.

Borsa İstanbul 100 (BIST-100) endeksi, diğer büyük dünya endeksleri ile birlikte finans piyasalarının küreselleşme değişiminin bir parçası olmuştur. Endeksler arasındaki ilişkinin analizi yatırımcılara büyük avantajlar sağlayacaktır. Bu durumdan yola çıkarak çeşitli makine öğrenmesi algoritmaları ile büyük dünya endeksleri ve bazı makroekonomik göstergeler kullanılarak BIST-100 endeksinin değişim yönünün (artış-azalış) sınıflandırılması amaçlanmıştır. Bu amaç doğrultusunda BIST-100 endeksinin değişim yönünün sınıflandırmasında etkin rol oynayan değişkenler belirlenmiş ve belirlenen bu değişkenler yardımıyla sınıflandırma başarılarında değişim olup olmadığı incelenmiştir. Tüm değişkenler ile yapılan sınıflandırmada lojistik regresyonun %70,6; öznitelik seçimi ile yapılan sınıflandırmada da Destek Vektör Makinesi PUK çekirdeği algoritmasının %71,9 daha doğru sınıflandırma başarısı gösterdiği belirlenmiştir. Böylelikle daha az sayıda değişken ile daha yüksek sınıflandırma başarısı elde edilmiştir.