Önder Demir

Computer-aided detection of lung nodules using outer surface features

In this study, a computer-aided detection (CAD) system was developed for the detection of lung nodules in computed tomography images. The CAD system consists of four phases, including two-dimensional and three-dimensional preprocessing phases. In the feature extraction phase, four different groups of features are extracted from volume of interests: morphological features, statistical and histogram features, statistical and histogram features of outer surface, and texture features of outer surface. The support vector machine algorithm is optimized using particle swarm optimization for classification. The CAD system provides 97.37% sensitivity, 86.38% selectivity, 88.97% accuracy and 2.7 false positive per scan using three groups of classification features. After the inclusion of outer surface texture features, classification results of the CAD system reaches 98.03% sensitivity, 87.71% selectivity, 90.12% accuracy and 2.45 false positive per scan. Experimental results demonstrate that outer surface texture features of nodule candidates are useful to increase sensitivity and decrease the number of false positives in the detection of lung nodules in computed tomography images.

Machine learning based phishing detection from URLs

Abstract Due to the rapid growth of the Internet, users change their preference from traditional shopping to the electronic commerce. Instead of bank/shop robbery, nowadays, criminals try to find their victims in the cyberspace with some specific tricks. By using the anonymous structure of the Internet, attackers set out new techniques, such as phishing, to deceive victims with the use of false websites to collect their sensitive information such as account IDs, usernames, passwords, etc. Understanding whether a web page is legitimate or phishing is a very challenging problem, due to its semantics-based attack structure, which mainly exploits the computer users’ vulnerabilities. Although software companies launch new anti-phishing products, which use blacklists, heuristics, visual and machine learning-based approaches, these products cannot prevent all of the phishing attacks. In this paper, a real-time anti-phishing system, which uses seven different classification algorithms and natural language processing (NLP) based features, is proposed. The system has the following distinguishing properties from other studies in the literature: language independence, use of a huge size of phishing and legitimate data, real-time execution, detection of new websites, independence from third-party services and use of feature-rich classifiers. For measuring the performance of the system, a new dataset is constructed, and the experimental results are tested on it. According to the experimental and comparative results from the implemented classification algorithms, Random Forest algorithm with only NLP based features gives the best performance with the 97.98% accuracy rate for detection of phishing URLs.

Retina Fundus Floresan Anjiyografi Görüntülerinde Drüsen Alanlarının Otomatik Tespiti ve Büyüklüklerinin Hesaplanması

Bilgisayar destekli tespit (BDT) sistemleri biyomedikal goruntulerin analizinde siklikla kullanilmaktadir. Bu calismada retinal fundus anjiyografi goruntuleri uzerinde yasa bagli makula dejenerasyonu (YBMD) hastaliginin tespiti icin bir BDT sistemi gerceklestirilmis ve patojenik drusen alanlarinin buyuklugunun hesaplanmasi saglanmistir. Calismanin amaci YBMD hastaliginin goruldugu alanlarin tespitinin ve buyuklugunu hesaplamanin yaninda hastaliga karsi uygulanan tedavinin sonucunun takibini de saglamaktir. Gelistirlen sistemin yardimiyla optalmoloji uzmanlari isaretlenen alanlari kisa surede tespit edebilirler ve hastaligin tedaviye verdigi cevabi basit bir sekilde gozlemleyebileceklerdir. Gelistirilen BDT sistemi 4 asamadan olusmaktadir, a) onisleme asamasi, b) bolutleme asamasi, c) ilgi alani tespiti ve d) oznitelik cikarma ve tespit asamasi. Gelistirlen BDT sistemi 75 goruntuden olusan bir verisetiyle test edilmistir. BST sisteminin elde ettigi sonuclar bir optalmoloji uzmaniyla karsilastirilarak sonuc bolumunde sunulmustur. Gelistirilen BDT sistemi 66 dogru, 9 hatali tespit yaparak %88 dogruluk orani saglamistir.

Applying social networks to engineering education

Social networking sites (SNSs) are a popular Internet‐based means for users to communicate and interact with each other. Although they have caught the attention of many researchers and are already being used as educational tools, very few studies have investigated the effects of using an SNS in engineering education. This study, therefore, aims to analyze the effects of using the Edmodo platform as a teaching and learning support tool on students’ academic and practical performance in the Introduction to Information Technology and Algorithms course, as well as in the Computer Programming course they took in the following semester. It also considers the students’ opinions about the Edmodo system. For this study, a total of 62 students studying in the Electrical and Electronics Engineering Department during the 2016–2017 fall semester were divided into two equally sized groups. The control group underwent a traditional face‐to‐face education, whereas the experimental group augmented this using the Edmodo system. A mixed‐methods approach with a post‐test‐only control group design was used: quantitative data were obtained from student tests, together with qualitative data from follow‐up interviews. The students’ grades were analyzed using Students t‐test and correlation analysis, showing that the experimental group performed better in their academic and laboratory assessments and that there was a moderately positive relationship between the post‐test results and performance in the subsequent Computer Programming course.

Feature selections for the machine learning based detection of phishing websites

Phishing websites are malicious sites which impersonate as legitimate web pages and they aim to reveal users important information such as user id, password, and credit card information. Detection of these phishing sites is a very challenging problem because phishing is mainly a semantics-based attack, which especially abuses human vulnerabilities, however not network or system vulnerabilities. As a software detection scheme, two main approaches are widely used: blacklists/whitelists and machine learning approaches. Machine learning solutions are able to detect zero-hour phishing attacks and they have superior adaption for new types of phishing attacks, therefore they are mainly preferred. To use this type of solution features of input must be selected carefully. The whole performance of the solution depends on these features. Therefore, in this paper, it is aimed to list and identify the important features for machine learning-based detection of phishing websites.

Lung nodule detection using template matching and similarity measurement

In this study, we developed a computer aided detection system (CAD) to detect lung nodules on computed tomography images. Template matching and similarity measurement methods used for examine whether region of interest which extracted using image pre-processing techniques are nodule candidate. Intensity thresholding, distance thresholding, neighbourhood analysis are preprocessing techniques of the developed CAD system. Pearsons correlation coefficient, simple matching coefficient, Jaccards coefficient, Euclidean Distance and Sokal & Sneath similarity coeeficient calculated to measure similarity between nodule candidate and the template. Sensitivity of the CAD and number of false positives per slice are given in conclusion.