Yiltan Bitirim

• PhD
• Professor (Associate) at Eastern Mediterranean University

Emotion classification using physiological signals is still a challenging task even the sensor technology and machine learning algorithms evolved within the decades. In this study, the performance of KNN, DT, RF, LR, and XGB algorithms on emotion classification was evaluated in terms of accuracy on the CASE dataset. Three sub-datasets namely Downsa...

Emoji prediction is an important research task that focuses on finding the most appropriate emoji(s) quickly and effortlessly for a specific text. Now that Turkish is on the list of the top 20 most spoken languages in the world and there are a considerable number of Turkish-speaking social media users, studying emoji prediction in Turkish holds sig...

Background: Diabetes Mellitus (DM) is a significant risk, mostly causing blindness, kidney failure, heart attack, stroke, and lower limb amputation. A Clinical Decision Support System (CDSS) can assist healthcare practitioners in their daily effort and can improve the quality of healthcare provided to DM patients and save time. Objective: In thi...

Human recognition and authentication through biometrics generally rely on feature extraction from images of physiological traits. These images can be represented in different color models. This study represents human palm vein images in five different color spaces (RGB, XYZ, LAB, YUV, and HSV) and combines their decisions in CNN models for person r...

Hand dorsal biometric recognition system proposed in this study combines the strength of information in regions of dorsal vein biometric trait in a deep learning based Convolutional Neural Networks (CNN) model. The approach divides each dorsal image into five overlapping regions; consequently, five different training and test sets are obtained for...

In this study, the KNN performance was evaluated for emotion and valence-arousal classifications using CASE dataset. Two class labels (valence-arousal and emotion) were assigned for each data record. The data was downsampled based emotion labels in order to have a balanced dataset. The KNN classifier was employed using 5-fold cross validation with...

Face recognition is one of the most commonly used form of biometric authentication today especially with the proliferation of smartphones and similar devices that armed with this capability. However, attacks on this kind of technology have also become very common; these can be in form of reconstructed samples or self-manufactured synthetic samples....

Voice recognition offers a wide range of applications, ranging from human-machine interaction, video categorization and tagging, voice authentication, voice search, gender categorization of phone calls, among others. Gender classification by speech is generally very easy for human; for machines, however, it is not the case. This study presents Logi...

In this study, a content-based recommendation approach is proposed. It utilizes the preprocessed 245 top movie summaries of IMDB and the favorite movie genres of the user elicited with the questionnaire method and then, recommends potential products -from a product pool- that the user can “like”. For testing; a test dataset that consists of real pr...

This study investigates the reverse image search performance of Google, in terms of Average Precisions (APs) and Average Normalized Recalls (ANRs) at various cut-off points,on finding out similar images by using fresh Image Queries (IQs) from the five categories “Fashion,” “Computer,” “Home,” “Sports,” and “Toys.” The aim is to have an insight abou...

A palm vein recognition system is proposed in this paper. The efficiency of three convolutional neural network models (VGG16, VGG19 and AlexNet) in palm vein biometrics is compared and then this study proposes to fuse them with Decision-Level Fusion. These models employ the use of high number of filters during training which leads to very high comp...

pp. 34, Abstracts Book (ISBN: 978-605-69925-4-4)

pp. 5-6, Proceedings Book (ISBN:978-625-7720-49-6).

A wrist vein recognition system is proposed in this paper. The system combines three texture-based feature descriptors, namely, multiple filters of Binarized Statistical Image Features (M-BSIF), 2D Gabor filter, and Histogram of Gradient orientation by Decision-Level Fusion. The method was tested on two publicly available datasets obtained from FYO...

pp. 10, Proceedings Book.

In this paper, we propose a palm vein recognition system that combines two approaches using a decision-level fusion strategy. The first approach employs Binarized Statistical Image Features (BSIF) descriptor method on five overlapping sub-regions of palm vein images and the second approach uses a convolutional neural networks (CNN) model on each pa...

The purpose of this survey-based study is to make an analysis of search engine users' behaviors on the Search Engine Results Pages (SERPs) based on the three demographic characteristics gender, age, and program studying. In this study, a questionnaire was designed with 12 closed-ended questions. Remaining questions other than the demographic charac...

This study investigates reverse image search performance of Google, in terms of Average Precisions (APs) at various cut-off points, on finding out similar images by using fresh Image Queries (IQs) from the five categories “Fashion”, “Computer”, “Home”, “Sports”, and “Toys”, in order to have an insight about reverse image search performance of Googl...

In this article, a system Smart Emergency Notification System (SENS) is proposed for both emergency responders and the community. SENS detects single/multiple emergency case(s) (i.e. road accident, fire, and injury) automatically from images sent by a smartphone via the Internet by the proposed promising approach; afterward, it notifies the police,...

Multimodal biometric systems are preferred as a defense compared to unimodal systems. This study introduces an open access multimodal vein database named FYO with each letter dedicated to each author’s name. The database involves three biometric traits; palm vein, dorsal vein and wrist vein of the same individuals, to explore and enhance research i...

Diabetic Retinopathy (DR) is a medical condition, also known as diabetic eye disease, which is vision-threatening damage to the retina of the eye caused by diabetes. As the technology advances, researchers are becoming more interested in intelligent medical diagnosis systems to assist screening of DR in earlier stages. In this study, variety of sta...

In this paper, a novel smart system based on hydroponics is proposed. It is aimed to help educate children by contributing to their improvement on cognitive domain, affective domain and psychomotor domain. This hydroponics-based smart education system is task oriented, does not interfere the child’s daily needs such as studying and sleeping and inc...

Google is the most popular search engine in the world. Another service of Google Inc. is Google Trends which is based on Google search data. The World Health Organization held an informal consultation in 2017 and announced the priority disease list. There was an urgent need for research and development for the priority diseases. “Lassa Fever”, “Mar...

Obesity is a public health problem that has become widespread worldwide. Obesity may increase the risk for many health problems such as early puberty, Type 2 diabetes, certain types of cancer, high blood pressure, heart diseases, and sleep apnea. The detection of obesity in children and adolescents at early stages is important and it is crucial to...

This study introduces the Remote Medical Tracking Mobile System (TrackARTI) which guides parents or health expert users in the treatment of Acute Respiratory Tract Infections (ASYE) diseases for children aged 0-6 years. The main purpose of this system is to gather actual data belonging to the medical cases of ASYE period in various environments ins...

Bu çalışmada, Google Play mağazası üzerinden kullanıcının erişebildiği akıllı park sistemi sınıfına giren Android tabanlı mobil uygulamalar araştırılmıştır. Öncelikle, Google Play mağazası üzerinde altı sorgu çalıştırılarak 66 tekil ilgili uygulama adayına erişilmiştir. Daha sonra, incelemeler yapılarak bu ilgili uygulama adayları içerisinden dokuz...

pp. 39, Abstracts Book (ISBN: 978-605-84722-8-0).

This study evaluates the popular search engines, Google, Yahoo, Bing, and Ask, on finding Turkish documents by comparing their current performances with their performances measured six years ago. Furthermore, the study reveals the current information retrieval effectiveness of the search engines. First of all, the Turkish queries were run on the se...

Bu bildiride, İnternet’in geniş kullanıcı kitlesine sahip olması avantajından faydalanıp, Kuzey Kıbrıs Türk Cumhuriyeti (KKTC)’nin temel ekonomik kaynaklarından biri olan yerel ticari kuruluşların müşterilerle iletişimini kolaylaştırmayı sağlayacak bir sistemin mimari yapısının öntasarımı sunulmaktadır. İnternet ortamında çalışması öngörülüp gerçek...

This paper investigates the semantic search performance of search engines. Initially, three keyword-based search engines (Google, Yahoo and Msn) and a semantic search engine (Hakia) were selected. Then, ten queries, from various topics, and four phrases, having different syntax but similar meanings, were determined. After each query was run on each...

Short Message Service (SMS) has been one of the most popular services in mobile communication for years, and its usage is still growing. However, with the usage of Turkish character(s) in SMS message text, the maximum length of a single message reduces from 160 to 70 characters due to the transmission regulations of SMS designated by Cellular Telec...

This paper investigates the retrieval performance and necessity of major meta-image search engines (MISEs). Our study is realized in two phases. In the first phase, major image search engines (ISEs), namely, Google, Yahoo and Ask, are selected. Then, fifteen queries are determined from various topics and classified as one-, two- and three-word quer...

This paper investigates the information retrieval effectiveness of major image search engines based on various query topics. Initially, major image search engines, namely, Google, Yahoo, Ask and MSN are selected. Then, seven appropriate topics are determined from the categories of the top search terms used on the web and five queries per topic are...

In this article, the impact of the number of query words on information retrieval effectiveness of image search engines was investigated. First of all, three popular search engines, namely, Google, Yahoo and MSN, were selected. Then, 40 queries were extracted from the list of Wordtracker and categorized in four groups as one-, two-, three- and four...

In this article, the impact of increasing number of query words on information retrieval effectiveness of image search engines is investigated. First of all, four popular search engines, namely Google, Yahoo, Msn, and Ask are selected. Then, forty queries are extracted from the list of Wordtracker and categorized in four groups as one-word, two-wor...

This article investigates the information retrieval performance of popular search engines on finding Turkish documents. First of all, five popular search engines (Google, Yahoo, Msn, AlltheWeb and Ask) and a list of Turkish queries are determined. Each query is run on each search engine one by one and first twenty documents on each retrieval output...

In this article, two popular database connection types of mobile devices, which are Windows Authentication over IP (internet protocol) and IIS (internet information services) [1], are used and IEEE 802.11 standards, namely IEEE 802.11b and IEEE 802.11g, and GPRS/EDGE technology are evaluated with respect to query retrieval time of different number...

pp. 1-49

In this article, two popular database connection types of mobile devices and database server computers in a client/server model are investigated by some performance evaluation experiments. This investigation is carried out with respect to query result retrieval times on different number of attributes and records. In the first concept, database conn...

In this paper, we evaluate the effectiveness of a new stemming algo- rithm, FINDSTEM, for use with Turkish documents and queries, and compare the use of this algorithm with the other two previously defined Turkish stem- mers, namely "A-F" and "L-M" algorithms. Of them, the FINDSTEM and A-F algorithms employ inflectional and derivational stemmers, w...

This is an investigation of information retrieval performance of Turk- ish search engines with respect to precision, normalized recall, coverage and novelty ratios. We defined seventeen query topics for Arabul, Arama, Netbul and Superonline. These queries were carefully selected to assess the capabil- ity of a search engine for handling broad or na...

Evaluation of Information Retrieval Performance of Turkish Search Engines This is an investigation on the information retrieval performances of search engines based on various measures. We searched 17 queries of differing types on four Turkish search engines, namely Arabul, Arama, Netbul and Superonline. We classified each document/Web site contain...