Mühendislik ve Doğa Bilimleri Fakültesi

Permanent URI for this community

http://hdl.handle.net/20.500.14701/60327

Browse

Now showing 1 - 20 of 35

A framework for applying the principles of depth perception to information visualization
(ACM, 2010) Çipiloğlu Yıldız, Zeynep; Bülbül, Abdullah; Capin, Tolga; Çipiloğlu Yıldız, Zeynep; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
During the visualization of 3D content, using the depth cues selectively to support the design goals and enabling a user to perceive the spatial relationships between the objects are important concerns. In this novel solution, we automate this process by proposing a framework that determines important depth cues for the input scene and the rendering methods to provide these cues. While determining the importance of the cues, we consider the user’s tasks and the scene’s spatial layout. The importance of each depth cue is calculated using a fuzzy logic–based decision system. Then, suitable rendering methods that provide the important cues are selected by performing a cost-proﬁt analysis on the rendering costs of the methods and their contribution to depth perception. Possible cue conﬂicts are considered and handled in the system. We also provide formal experimental studies designed for several visualization tasks. A statistical analysis of the experiments veriﬁes the success of our framework.
The extension-based inference algorithm for pD
(Elsevier, 2012-02) Öztürk, Övünç; Özacar Öztürk, Tuğba; Ünalır, Murat Osman; Öztürk, Övünç; Özacar Öztürk, Tuğba; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
In this work, we present a scalable rule-based reasoning algorithm for the OWL pD* language. This algorithm uses partial materialization and a syntactic ontology transformation (the extension-based knowledge model) to provide a fast inference. Because the materialized part of the ontology does not contain assertional data, the time consumed by the process, and the number of inferred triples, remain fixed with varying amounts of assertional data. The algorithm uses database reasoning and a query rewriting technique to handle the remaining inference. The extension-based knowledge model and the database reasoning prevent the expected decreases in query performances, which are the natural result of online reasoning during query time. This work also evaluates the efficiency of the proposed method by conducting experiments using LUBM and UOBM benchmarks.
II. et ürünleri çalıştayı
(Celal Bayar Üniversitesi, 2012-12) Kayaardı, Semra; Çağındı, Özlem; Dağbağlı, Seval; Kayaardı, Semra; Çağındı, Özlem; Dağbağlı, Seval; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Gıda Mühendisliği Bölümü; Kayaardı, Semra; Çağındı, Özlem; Dağbağlı, Seval
Önemli ölçüde protein açığı bulunan ülkemizde özellikle hayvansal protein açığının kapatılması için mevcut hammaddelerin verimli bir şekilde kullanımına özen gösterilmesi önemli bir konudur. Bu durum hem ülke ve hem de dünya ekonomisi açısından önemlidir. Bu anlamda mekanik olarak kemiklerinden ayrılmış et (MAE)’lerin, et teknolojisine katkısı dikkat çekicidir. Mekanik olarak kemiklerinden ayrılmış et (MAE); gövde/karkas üzerindeki etin normal yollarla ayrılmasından sonra kemik üzerinde kalan etlerle, et kemik ayrımı pahalıya mal olan yumurta verimini tamamlamış anaç tavuk etleri ve balık etlerinin mekanik olarak ayrılması işlemidir. Günümüzde binlerce ton kırmızı ve beyaz et, mekanik olarak kemiklerinden ayrılarak ileri derecede işlenmiş et ürünleri üretiminde kullanılmaktadır.Mekanik olarak kemiklerinden ayrılmış piliç eti, yoğunluğu ve düşük maliyetinden dolayı emülsifiye et ürünlerinin üretiminde sıkça kullanılmaktadır.Son 20 yılda Türkiye’de de piliç eti tüketiminde bir artış gözlenmekte, özellikle piliç but, göğüs ve kanat gibi parçalara olan talep sürekli artmaktadır. Artan talep nedeniyle piliç karkasının göğüs kafesi, sırt ve boyunu içeren yaklaşık % 40'lara varan kısmı geride kalır. Bu kalan parçalar genellikle mekanik olarak kemiklerinden ayrılarak islenir ve teknolojiye kazandırılır.Yüzeylerinde yenilebilecek birçok et mevcut olan bu bölgelerin etleri mekanik olarak ayrılarak ekonomiye dahil edilebilmektedir. Mekanik olarak ayrılmış etler yüksek emülsiyon stabiliteleri, emülsiyon kapasiteleri ve su tutma özellikleri yanında çok ucuz olmaları ile hayvansal protein kaynağı olarak et teknolojisinde salam, sosis gibi emülsifiye et ürünlerinin üretimi yanında köfte, burger tipi ürünlerin üretiminde kullanımı hızla yaygınlaşmıştır. Lizin, löysin ve histidin gibi esansiyel aminoasitleri içermesi gibi tüm bu olumlu etkilerinin yanında başka bir takım olumsuz özelliklerinden dolayı, ileri işlenmiş ürün üretiminde kullanımı kısıtlanmaktadır. Mekanik olarak ayrılmış etlerin ürün içindeki fonksiyonel özelliklerini geliştirmek için çeşitli modifikasyonlar yapılması ve elde edilen ürünlerin kalsiyum, demir, flor ve fosfor gibi mineral maddeler ile protein, yağ ve rutubet gibi kimyasal bileşimi ve mikrobiyolojik özellikleri yönünden standart kriterler koyularak bu özelliklerin uygunluğunun denetlenmesi önemlidir. Böylelikle mekanik olarak ayrılmış etler, işlenmiş ürünlerde daha güvenli ve daha denetlenebilir bir kullanım alanı bulacaktır.
Improvement of intramolecular charge transfer within a donor–acceptor blend by doping novel synthesized benzothiadiazole small molecules in solid state
(Optical Materials, 2014) Dinçalp, Haluk; Murat, Gözde; İçli, Sıddık; Dinçalp, Haluk; Murat, Gözde; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Kimya Bölümü
Three electron-deficient small molecules based on 2,1,3-benzothiadiazole (BTD) units namely, 4,7-bis (3-methoxyphenyl)-2,1,3-benzothiadiazole (BT1), (3-{7-[3-(dimethylamino)phenyl]-2,1,3-benzothiadi azole-4-yl}phenyl)dimethylamine (BT2) and 3,30 -(2,1,3-benzothiadiazole-4,7-dyl)dianiline (BT3) were synthesized and their photophysical properties were investigated systematically to understand their potential usage in ternary organic solar cells (OSCs) as additive material to enhance the cell efficiency. All these molecules show broad absorption bands in 350–750 nm on glass substrate and their optical band gaps were calculated to be around 2.50–2.80 eV. BTD fluorescence dynamics were measured in polymer:BT1:fullerene blends with varying emission wavelengths of active layer. Fluorescence emission and time resolved measurements indicated photoinduced energy shift from BT1 dye to fullerene and also from polymer to BT1 dye upon excitation of the active layer.
IRIS: A Prot´eg´e plug-in to extract and serialize product attribute name-value pairs
(2014) Özacar Öztürk, Tuğba; Özacar Öztürk, Tuğba; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
This article introduces IRIS wrapper, which is developed as a Prot´eg´e plug-in, to solve an increasingly important problem: extracting information from the product descriptions provided by online sources and structuring this information so that is sharable among business entities, software agents and search engines. Extracted product information is presented in a GoodRelations-compliant ontology. IRIS also automatically marks up your products using RDFa or Microdata. Creating GoodRelations snippets in RDFa or Microdata using the product information extracted from Web is a business value, especially when you consider most of the popular search engines recommend the use of these standards to provide rich site data for their index.
A new method for computer-assisted detection, definition and differentiation of the urinary calculi
(Taylor & Francis, 2014-09) Öztürk, Övünç; Yıldırım, Düzgün; Tutar, Onur; Nurili, Fuad; Bozkurt, İbrahim Halil; Kayadibi, Hüseyin; Karaarslan, Ercan; Öztürk, Övünç; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Purpose: Urinary stones are common and can be diagnosed with computed tomography (CT) easily. In this study, we aimed to specify the opacity characteristics of various types of calcified foci that develop through the urinary system by using an image analysis program. With this method, we try to differentiate the calculi from the non-calculous opacities and also we aimed to present how to identify the characteristic features of renal and ureteral calcules. Materials and methods: We obtained the CT studies of the subjects (n = 48, mean age = 41 years) by using a dual source CT imaging system. We grouped the calculi detected in the dual-energy CT sections as renal (n = 40) or ureteric (n = 45) based on their locations. Other radio-opaque structures that were identified outside but within close proximity of the urinary tract were recorded as calculi "mimickers". We used ImageJ program for morphological analysis. All the acquired data were analyzed statistically. Results: According to thorough morphological parameters, there were statistically significant differences in the angle and Feret angle values between calculi and mimickers (p < 0.001). Multivariate logistical regression analysis showed that Minor Axis and Feret angle parameters can be used to distinguish between ureteric (p = 0.003) and kidney (p = 0.001) stones. Conclusions: Computer-based morphologic parameters can be used simply to differentiate between calcular and noncalcular densities on CT and also between renal and ureteric stones.
A tool for producing structured interoperable datafrom product features on the web
(Pergamnon, 2015) Özacar Öztürk, Tuğba; Özacar Öztürk, Tuğba; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
This paper introduces a tool that produces structured interoperable data from product features, i.e., attribute name-value pairs, on the web. The tool extracts the product features using a web site specific template created by the user. The value of the extracted data is maximized by using GoodRelations, which is the standard vocabulary for modeling product types and their features. The final output of the tool is GoodRelations snippets, which contain product features encoded in RDFa or Microdata. These snippets can be embedded into existing static and dynamic web pages in a way accessible to major search engines like Google and Yahoo, mobile applications, and browser extensions. This increases the visibility of your products and services in the latest generation of search engines, recommender systems, and other novel applications.
Word-context matrix based query expansion in information retrieval for Turkish Text
(BCS Learning and Development Ltd. Proceedings of the 6 th Symposium on Future Directions in Information Access 2015, 2015-09) Şatır, Emre; Alpkoçak, Adil; Kılınç, Deniz; Şatır, Emre; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
In this paper, we proposed a Query Expansion (QE) approach on a Turkish Text collection based on word-context matrix with a sliding fixed sized window and Singular Value Decomposition (SVD) method. Our query expansion approach uses semantic relationship of terms to improve the existing query expansion methods available in the literature, namely Bo1 (Bose-Einstein 1), Bo2 (Bose-Einstein 2) and KL (Kullback-Leibler). We evaluated our approach on Milliyet collection, which is a Turkish IR test bed containing more than 400K documents and 72 queries. The experimentation shows that our approach clearly improves the all three QE methods in terms of major Information Retrieval (IR) performance measures such as MAP, R-precision and P@10. tr
The behavior of evolutionary algorithms in combinatorial optimization
(2016-05-26) Abidin, Didem; Abidin, Didem; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Since combinatorial optimization is a branch of optimization, its domain consists of solution sets having different combinations of feasible answers to a problem. Some of the most famous combinatorial optimization problems can be listed as: Finding the shortest path (Travelling Salesman Problem in more specific), finding minimum spanning tree in a graph, black box optimization, solving a maze, optimizing airline network and so on. Evolutionary computation (EC) is one of the best computing techniques for solving combinatorial optimization problems. In EC, there are some algorithms inspiring from nature to find the best solution among the possible solutions' set. Genetic algorithm (GA) is a stochastic search technique which can be used to solve sequencing problems. Although it is being discussed for decades, Genetic Algorithms (GA) still finds very interesting application areas including most of the famous combinatorial optimization problems. All kinds of sequencing problems can be handled with GA. Its mechanisms (crossover, mutation and selection operators) simulate the behavior of reproduction process in nature and the result set is able to "evolve" to find better solutions. For this reason, they can easily be applied on problems which need to find the best solution among the solution candidates. GA is very convenient technique to observe performance with the changing parameters. For this reason, parameter tuning can be done on the same set of candidate solutions to find the best parameter combination of GA to solve the problem. When parameter tuning is applied, it has been observed that different parameter combinations can be more effective on different problems. In this study, the solution of a combinatorial optimization problem is searched by using GA. The operators used to simulate evolution mechanism play an important role on the behavior of the algorithm. The crossover rates for solving a sequencing problem with GA have been given as 0.7, 0.75, 0.8, 0.85, 0.9, 0.95 and 1.0. When the corresponding crossover rate values of the successful scenarios are examined, it is observed that the number of scenarios having crossover rates of 0.7 and 0.95 is more than the others. The same extraction can be done for the mutation rates. As mutation rates, 0.1, 0.15 and 0.2 were used. The number of successful scenarios with the mutation rate of 0.2 was more than the others. The population size interval is given as 100 – 200 (100, 120, 140, 160, 180 and 200) in the study. The system needs to test different sizes of populations because increasing the population size to a certain extent encourages the diversity of the population that the GA deals with. This means that as the population size increases, the possibility of having individuals with various values of fitness also increases. With populations of a small size, runtime and fitness values are more likely to be the best. As the population grows, finding “very reliable” solutions gets more difficult. Also the lengths of the sequences to be optimized directly affect the performance of GA. Two different sizes of data sets were selected. The smaller dataset has 17, the larger dataset has 30 chromosomes in each individual. This means that the genetic algorithm has to work in two different chromosome sizes. For all parameter combinations of genetic operators, three different crossover operators were applied in four different ways. That is; 1 – point order crossover (called GA1), 2 – point crossover (called GA2), partially mapped crossover – PMX (called GA3 in this project) and choosing one of the three crossover techniques randomly in each run (called GA4). Two different mutation operators were used randomly in each run. These are swap mutation and inversion mutation. The population used in GA is generated randomly. Since the different combinations of chromosomes indicate a different solution, obtaining random sequences of chromosomes is a crucial task for the genetic algorithm. This part can be considered as the most important and time consuming part of preparing data. Preparing the data in the proper format is also related with obtaining the possible solution set because sequencing chromosomes randomly is tightly dependent with the encoding mechanism of the genetic algorithm. There are several encoding mechanisms like binary encoding and real number encoding. For a combinatorial optimization problem having distinct integer values as the genes of a chromosome, permutation encoding is the correct encoding mechanism. In this kind of encoding, the order of neighboring genes has an importance. The importance of the order of the genes can be up to some prerequisite rules. For this reason, it is needed to have nonrecurring sequence of genes in a chromosome. For GA, the best solution has the reliability percentage of 98.53, which is a remarkable result emphasizing that GA mechanism is one of the best ways to implement solutions to combinatorial optimization problems. Differential Evolution (DE) is also a branch of evolutionary computation, in which optimization is done to find the best solution among all candidate solutions. As GA does, DE also creates new candidate solutions in each generation to obtain the individual having the best fitness value. In this optimization technique, unlike GA, mutation operator has a great importance to create a mutant individual. There are many algorithms to create the mutant individual, which is a candidate solution if it has better fitness value than the individual randomly chosen for mutation. The mutation mechanisms used in the study are DE/rand/1, DE/best/1, Simplex1 and Simplex2. For mutation, again unlike GA, DE has no mutation rates. This means that mutation is done in each generation. For crossover, there is a simple selection structure with a certain crossover rate. The crossover rates differ from the rates used in GA for this study because crossover is not the main operator for DE. The crossover rates used in this study are 0.5 and 0.8. Instead of the mutation rate, DE has another important parameter to execute. The F value is a constant used for the calculations of mutation operators of DE. The values used for F are 0.5 and 0.6. When four mutation operators are executed for two different crossover rates and two different F values, very satisfactory results are obtained. The results for Simplex1 and Simplex2 algorithms were better than the DE/rand/1 and DE/best/1 algorithms. The best solution has given the reliability percentage of 96.84, which puts forward that DE is a good tool to make optimization with. The best result was obtained for the parameter combination of 0.5 crossover rate and 0.5 F value. The idea behind using evolutionary algorithms like genetic algorithms or differential evolution algorithms for combinatorial problems is that the structure of these algorithms very much suit to the structure of combinatorial optimization problems. As future work, a hybrid version of genetic algorithms and differential evolution algorithms will be applied to a specific domain to observe the performance changes in finding the best solution.
A perceptual quality metric for dynamic triangle meshes
(Springer International Publishing, 2017) Çipiloğlu Yıldız, Zeynep; Capin, Tolga; Çipiloğlu Yıldız, Zeynep; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
A measure for assessing the quality of a 3D mesh is necessary in order to determine whether an operation on the mesh, such as watermarking or compression, affects the perceived quality. The studies on this field are limited when compared to the studies for 2D. In this work, we aim a full-reference perceptual quality metric for animated meshes to predict the visibility of local distortions on the mesh surface. The proposed visual quality metric is independent of connectivity and material attributes. Thus, it is not associated to a specific application and can be used for evaluating the effect of an arbitrary mesh processing method. We use a bottom-up approach incorporating both the spatial and temporal sensitivity of the human visual system. In this approach, the mesh sequences go through a pipeline which models the contrast sensitivity and channel decomposition mechanisms of the HVS. As the output of the method, a 3D probability map representing the visibility of distortions is generated. We have validated our method by a formal user experiment and obtained a promising correlation between the user responses and the proposed metric. Finally, we provide a dataset consisting of subjective user evaluation of the quality of public animation datasets.
A semantic web case study: representing the ephesus museum collection using erlangen CRM ontology
(Springer Verlag, 2017-11) Özacar Öztürk, Tuğba; Öztürk, Övünç; Yüksel, Fulya; Lobaba, Salloutah; Abdülbaki, Baraa; Bilici, Elif; Özacar Öztürk, Tuğba; Öztürk, Övünç; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Cultural heritage has recently become an important application area for Semantic Web technologies. Semantic Web technologies and ontologies provide a solution for intelligent integration of heterogeneous data about the cultural heritage. The objective of this paper is the construction of an ontology for the cultural heritage related to Selçuk region in Western Turkey. We use a subset of the Erlangen CRM as our ontology schema, then we populate the ontology with 814 objects in the Ephesus Museum. One of the objectives of this work is to integrate the ontology with other projects which use Erlangen CRM as ontology schema. Therefore, we present an integration case study that aggregates content from Ephesus Museum and British Museum.
Klasik Türk müziğinde makam tanıma için veri madenciliği kullanımı
(Gazi Üniversitesi, 2017-12) Abidin, Didem; Özacar, Tuğba; Öztürk, Övünç; Abidin, Didem; Özacar, Tuğba; Öztürk, Övünç; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Müzik eserlerinin sayısal ortama aktarılmasıyla birlikte, bilgisayar bilimleri müzikoloji çalışmalarının içerisinde kendine yer bulmaya başlamıştır. Müzik eserleri bilimsel araştırmalarda veri olarak kullanılmakta ve hesaplamalı müzik alanı bu alanda yapılan çalışmalar ile hızla gelişmektedir. Her ne kadar yapılan çalışmaların büyük bir bölümü sembolik olarak ifade edilmesi daha kolay olan Batı Müziği eserleri üzerine olsa da, Türk Müziği eserleri de artık çeşitli çalışmaların konusu olmaktadır. Türk Müziğinin temeli olan makam sistemi, bilgisayar bilimleri ile uğraşan araştırmacıların dikkatini çekmiş ve Türk Müziği eserler veri madenciliği, makine öğrenmesi, sınıflandırma gibi çalışmaların konusu olmuştur. Bu çalışmada, bir ses dosyasından nota tanıma ile elde edildiği varsayılan ve 1261 Türk Müziği eserine ait sadece nota dizilerini içeren veri dosyası üzerinde makine öğrenmesi metodu ile makam tahmini yapmaya yönelik deneysel bir çalışma gerçekleştirilmiştir. Music XML biçimindeki eserlerin makine öğrenmesi uygulamasında kullanılabilmesi amacıyla bir yazılım geliştirilmiş, bu yazılımla makine öğrenmesinde başarımı arttırmak için özgün veri kümesine dört farklı türetilmiş veri sahası eklenmiştir. Sonuç olarak, ‘Rastgele Orman’algoritması ile makam tanımada %89,7 oranında başarım gözlemlenmiştir.
Optoelectronic performance comparison of new thiophene linked benzimidazole conjugates with diverse substitution patterns
(Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2018) Dinçalp, Haluk; Murat Saltan, Gözde; Kırmacı, Eser; Kıran, Merve; Zafer, Ceylan; Dinçalp, Haluk; Murat Saltan, Gözde; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Kimya Bölümü
In an approach to develop efficient organic optoelectronic devices to be used in light-driven systems, a series of three thiophene linked benzimidazole conjugates were synthesized and characterized. The combination of two thiophene rings to a benzimidazole core decorated with different functional groups (such as \\OCH3, \\N(CH3)2,\\CF3) resulted in donor-acceptor type molecular scaffold. The effect of the electronic behavior of the substituents on the optical, electrochemical, morphological and electron/hole transporting properties of the dyes were systematically investigated. DTBI2 dye exhibited distinct absorption properties among the other studied dyes because N,N-dimethylamino group initiated intramolecular charge transfer (ICT) process in the studied solvents. In solid state, the dyes exhibit peaks extending up to 600 nm. Depending on the solvent polar ities, dyes show significant wavelength changes on their fluorescence emission spectra in the excited states. Mor phological parameters of the thin films spin-coated from CHCl3 solution were investigated by using AFM instrument; furthermore photovoltaic responses are reported, even though photovoltaic performances of thefabricated solar cells with different configurations are quite low.
Electrical analogue of arterial blood pressure signals
(Gümüşhane Üniversitesi, 2018) Emek, Sevcan; Evren, Vedat; Bora, Şebnem; Emek, Sevcan; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
In this study, we propose an electrical circuit model that will be useful for understanding of the mechanisms and dynamics of the human cardiovascular system, which is considered as a complex system in the field of physiology. The electrical circuit model, defined as the Windkessel model, plays an important role in the observation of the characteristic effect of the blood pressure on the arterial system. An electrical circuit model, which we have connected to the input terminals of the Windkessel model, ensures that the mean arterial blood pressure signals are observed within the expected range of values. The Windkessel circuit model that we have tried to develop in this study was constructed in a laboratory environment and the results were observed. It is thought that this study will contribute to the literature in terms of the development of the Windkessel model by increasing the number of parameters involved in the models of heart and arterial system.
A machine learning framework for full-reference 3D shape quality assessment
(Springer Berlin Heidelberg, 2018) Çipiloğlu Yıldız, Zeynep; Öztireli, A. Cengiz; Capin, Tolga; Çipiloğlu Yıldız, Zeynep; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
To decide whether the perceived quality of a mesh is inﬂuenced by a certain modiﬁcation such as compressionors impliﬁcation, a metric for estimating the visual quality of 3D meshes is required. Today, machine learning and deep learning techniques are getting increasingly popular since they present efﬁcient solutions to many complex problems. However, these techniques are not much utilized in the ﬁeld of 3D shape perception. We propose a novel machine learning-based approach for evaluating the visual quality of 3D static meshes. The novelty of our study lies in incorporating crowdsourcing in a machine learning framework for visual quality evaluation. We deliberate that this is an elegant way since modeling human visual system processes is a tedious task and requires tuning many parameters. We employ crowdsourcing methodology for collecting data of quality evaluations and metric learning for drawing the best parameters that well correlate with the human perception. Experimental validation of the proposed metric reveals a promising correlation between the metric output and human perception. Results of our crowdsourcing experiments are publicly available for the community.
WALRUS: A retro communication gadget based on internet of things technologies
(2018) Öztürk, Övünç; Küçük, Yunus Emre; Yalnız, Ahmet; Öztürk, Övünç; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
In this work, a new system depending on a device that can encode and decode push-button signals, modulated using Morse code conventions, were developed to build a low-cost communication medium based on Internet of Things (IoT). The proposed system consists of two parts: a base station and handheld terminals. The base station is a single board computer with a web application based on Node.js. Handheld terminals are small battery powered devices, developed using MCU's, that can communicate with the base station over the wireless network. They can encode and decode Morse code, and convert to text or speech depending on the configuration of the terminal, which can be extended by using different add-ons, such as an OLED screen or a text to speech module. Communication between terminals is orchestrated by the base station using IoT Technologies like MQTT. The handheld terminals can be used by disabled people as a mean for private conversation, or a gadget for entertainment purposes. The system is an uncomplicated and low-cost communication medium and implemented to find alternative use cases for the IoT technologies.
TSMONT ontolojisi: Türk Sanat Müziği makamlarının anlamsal olarak modellenmesi
(Dicle Üniversitesi, 2018) Öztürk, Övünç; Abidin, Didem; Özacar Öztürk, Tuğba; Öztürk, Övünç; Abidin, Didem; Özacar Öztürk, Tuğba; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Anlamsal web teknolojileri, yalnızca insanlar tarafından değil makineler tarafında da anlaşılabilir bir veri ağı oluşturulmasını hedefler. Bu veri ağında birbirleriyle ilişkilendirilmiş ontolojiler aracılığıyla veri anlamı ile bağlanır. Ontoloji belirli bir alandaki (örneğin bu çalışmada müzik alanı) ilgili olan bir dizi kavram ve ilişkinin biçimsel belirtimidir. Bu makalede, Türk Sanat Müziği teorisinde yer alan kavramların anlamsal olarak modellenmesine yönelik TSMONT ontolojisi geliştirilmiştir. Bu ontoloji ile Türk Sanat Müziği’nde kullanılan makamlar, diziler, çeşniler, perdeler; bu kavramlar arasındaki ilişkiler ve kavram ve ilişkilere ait örnekler modellenmektedir. Böylelikle müzik alanındaki heterojen veri kaynaklarındaki bilgiden dağıtık ve birlikte çalışabilir bir bilgi tabanı oluşturulabilir.Ayrıca, TSMONT ontolojisi, en çok kullanılan ontoloji değerlendirme araçlarından olan OntoQA metrikleri çerçevesinde değerlendirilmiştir. Yapılan değerlendirme sonucunda TSMONT ontolojisi, derin bir taksonomiye sahip, zengin ilişkilerin olduğu, kavramların kapsamlı bir biçimde modellendiği, yeterli sınıf örneğine sahip ve kavramların birbirine yakın öneme sahip olduğu bir ontoloji olarak değerlendirilmiştir.
Buz pateni öğrenenler için hareket algılama tabanlı bir sistem tasarımı
(Institute of Electrical and Electronics Engineers Inc., 2018-05) Öztürk, Övünç; Kahramanlı, Murat Mecit; Öztürk, Övünç; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Bu bildiride buz patenini yeni öğrenmeye başlayanlar için hareket algılama tabanlı, ergonomik, kurulumu basit ve taşınabilir bir sistem tasarımı sunulmaktadır. Sistem kullanıcıların temel kayma tekniklerini doğru bir şekilde uygulamasına destek olacaktır. Sistem tasarlanırken, geliştirmeye açık, esnek ve modüler olmasına dikkat edilmiştir. Böylece, daha ileri seviye tekniklerin kontrolünün sağlanması veya sistemin vücut koordinasyonu gerektiren diğer sporlar için adaptasyonu daha kolay bir şekilde gerçekleştirilebilecektir. Sistemin temel destek elemanları ataletsel ölçüm birimleridir. Sistemdeki ataletsel ölçüm birimi bile¸senleri bedenin her bir uzvunun yön vektörünü belirlemektedir. Bu sensörlerden gelen veriler aracılığıyla belirlenmiş olan yönler, sistemde kullanıcıyı doğru bir şekilde yönlendirmek için altyapı oluşturmaktadır. Kullanıcı kulaklıkları ve sensör modüllerini taktıktan sonra sensörlerden gelen veriler kullanılarak kullanıcıya sesli olarak açıkça anlaşılır, motive edici ve kısa komutlar verilecektir.
Movie rating prediction with machine learning algorithms on IMDB data set
(2018-05-11) Abidin, Didem; Bostancı, Caner; Site, Atakan; Abidin, Didem; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Predicting movie success with machine learning algorithms has become a very popular research area. There are many algorithms which can be applied on a data set to make movie success prediction if the data set is prepared and represented properly. In this study, we explained how IMDB movie data was used for movie rating prediction. The data set extracted from IMDB was formatted and prepared for datamining algorithms. These algorithms were executed on WEKA application environment and the performances in movie ratings and confusion matrices were obtained. The seven machine learning algorithms used have performed well on the data set with varying performance ratings of 73.5% to 92.7%. Random Forest algorithm had the best performance of 92.7%. This is the highest score obtained among similar studies.
Using classification algorithms for Turkish Music Makam recognition
(Selçuk Üniversitesi, 2018-09) Abidin, Didem; Özacar Öztürk, Tuğba; Öztürk, Övünç; Abidin, Didem; Özacar Öztürk, Tuğba; Öztürk, Övünç; Fakülteler > Mühendislik Ve Doğa Bilimleri Fakültesi > Bilgisayar Mühendisliği Bölümü
Turkish Music pieces are used in various studies including makam recognition in computational music domain. Turkish Music pieces offer a rich content to the researchers because of their different makam properties. SymbTr is one of the most referred Turkish Music data sets in this area. In this study, the pieces from SymbTr data set belonging to 13 makams are used to execute 10 different machine learning algorithms for makam recognition and the performances of these algorithms are evaluated. These algorithms were executed on WEKA application environment and the performances in makam recognition were obtained with F-measure and recall metrics. The machine learning algorithms performed between 82% and 88%.

Browse

Browsing by Access Type "açık erişim (open access)"

Results Per Page

Sort Options