In the medical field, successful classification of microarray gene expression data is of major importance for cancer diagnosis. However, due to the profusion of genes number, the performance of classifying DNA microarray gene expression data using statistical algorithms is often limited. Recently, there has been an important increase in the studies on the utilization of artificial intelligence methods, for the purpose of classifying large-scale data. In this context, a hybrid approach based on the adaptive neuro-fuzzy inference system (ANFIS), the fuzzy c-means clustering (FCM), and the simulated annealing (SA) algorithm is proposed in this study. The proposed method is applied to classify five different cancer datasets (i.e., lung cancer, central nervous system cancer, brain cancer, endometrial cancer, and prostate cancer). The backpropagation algorithm, hybrid algorithm, genetic algorithm, and the other statistical methods such as Bayesian network, support vector machine, and J48 decision tree are used to compare the proposed approach's performance to other algorithms. The results show that the performance of training FCM-based ANFIS using SA algorithm for classifying all the cancer datasets becomes more successful with the average accuracy rate of 96.28% and the results of the other methods are also satisfactory. The proposed method gives more effective results than the others for classifying DNA microarray cancer gene expression data.

Prostate cancer is one of the most common types of cancer among males as well as causing the most deaths. Early diagnosis of prostate cancer plays an important role in the treatment of the disease. Therefore, microarray technology is widely used in the diagnosis of inherited diseases such as prostate cancer. With this technology, it is possible to obtain more knowledge about cancer by analyzing thousands of gene expressions. However, it is quite difficult to analyze complex relationships among thousands of genes in microarray data. For this reason, high performance artificial intelligence-based classification methods are needed in recent years. In this study, a hybrid method has been proposed for optimizing the parameters of Adaptive Neuro Fuzzy Inference System (ANFIS) with Genetic Algorithm (GA) in order to classify prostate cancer gene expression profiles. The performance of the proposed method is compared with those of ANFIS models trained by different learning algorithms. According to obtained results, the proposed method is more successful than the other methods, with the accuracy of 90.32%.

In this study, the electrical activities in the brain were classified during mental mathematical tasks and silent text reading. EEG recordings are collected from 18 healthy male university/college students, ages ranging from 18 to 25. During the study, a total of 60 slides including verbal text reading and arithmetical operations were presented to the subjects. EEG signals were collected from 26 channels in the course of slide show. Features were extracted by employing Hilbert Huang Transform (HHT). Then, subjectdependent and subject-independent classifications were performed using k-Nearest Neighbor (k-NN) algorithm with parameters k=1, 3, 5 and 10. Subject-dependent classifications resulted in accuracy rates between 95.8% and 99%, whereas the accuracy rates were between 92.2% and 97% for subject independent classification. The results show that EEG data recorded during mathematical and silent reading tasks can be classified with high accuracy results for both subject-dependent and subject-independent analysis.

In this study, the effects of feature selection on classification of the electrical signals generated in the brain during numerical and verbal operations are investigated. 18 healthy university/college students were chosen for the experimental study. EEG signals were recorded during silent reading and mental arithmetic operations without using any pen and paper. A total of 60 slides, 30 of which contained reading passages and the rest contained arithmetic operations, were presented in the experiment. EEG signals recorded from 26 channels during the slide show. The recorded EEG signals were analyzed by Hilbert Huang Transform (HHT), and then features were extracted. 312 features were classified by Bayesian Network algorithm without applying feature selection with 92.60% average accuracy. Consistency measures and Correlation based Feature Selection methods were, then, used for feature selection and the numbers of selected features are 8 and 39 on average, respectively. Classification accuracies by using these feature selection algorithms were obtained as 93.98% and 95.58%, respectively. The results showed that feature selection algorithms contribute positively to the classification performance.

ÖZSınıflandırma, verilerin analiz edilmesi için önemli bir veri madenciliği tekniği olup tıp, genetik ve biyomedikal mühendisliği başta olmak üzere birçok alanda kullanılmaktadır. Özellikle tıp alanında DNA mikrodizi gen ekspresyon verilerini sınıflandırmaya yönelik yapılan çalışmalarda artış görülmektedir. Ancak, mikrodizi gen ekspresyon (ifade) verilerinde bulunan gen sayılarının çokluğu ve bu veriler arasında doğrusal olmayan bağıntılar bulunması gibi problemlerden dolayı geleneksel sınıflandırma algoritmalarının başarımları sınırlı kalabilmektedir. Bu sebeplerden dolayı son yıllarda sınıflandırma probleminin çözümü için yapay zekâ tekniklerine dayalı sınıflandırma yöntemlerine olan ilgi giderek artmaya başlamıştır. Bu çalışmada, karaciğer mikrodizi kanser veri setinin sınıflandırılması için Uyarlamalı Ağ Tabanlı Bulanık Mantık Çıkarım Sistemi (ANFIS) ve Genetik Algoritmaya (GA) dayalı hibrid bir yaklaşım önerilmiştir. Simülasyon sonuçları, diğer bazı yöntemlere ait sonuçlarla karşılaştırılmıştır. Elde edilen sonuçlardan, önerilen yöntemin diğer yöntemlere göre daha başarılı olduğu görülmüştür.Anahtar Kelimeler: Neuro-fuzzy, ANFIS, genetik algoritma, sınıflandırma, mikrodizi gen ifade Training ANFIS structure using genetic algorithm for liver cancer classification based on microarray gene expression data ABSTRACT Classification is an important data mining technique, which is used in many fields mostly exemplified as medicine, genetics and biomedical engineering. The number of studies about classification of the datum on DNA microarray gene expression is specifically increased in recent years. However, because of the reasons as the abundance of gene numbers in the datum as microarray gene expressions and the nonlinear relations mostly across those datum, the success of conventional classification algorithms can be limited. Because of these reasons, the interest on classification methods which are based on artificial intelligence to solve the problem on classification has been gradually increased in recent times. In this study, a hybrid approach which is based on Adaptive Neuro-Fuzzy Inference System (ANFIS) and Genetic Algorithm (GA) are suggested in order to classify liver microarray cancer data set. Simulation results are compared with the results of other methods. According to the results obtained, it is seen that the recommended method is better than the other methods.