395,851 research outputs found
Analysis of preprocessing methods on classification of Turkish texts
No full textÇakırman, Erhan (Dogus Author) -- Ganiz, Murat C. (Dogus Author) -- Akyokuş, Selim (Dogus Author) -- Gürbüz, Mustafa Z. (Dogus Author) -- Conference full title: 2011 International Symposium on Innovations in Intelligent Systems and Applications (INISTA 2011) Istanbul, Turkey, 15 - 18 June 2011Preprocessing is an important task and critical step in information retrieval and text mining. The objective of this study is to analyze the effect of preprocessing methods in text classification on Turkish texts. We compiled two large datasets from Turkish newspapers using a crawler. On these compiled data sets and using two additional datasets, we perform a detailed analysis of preprocessing methods such as stemming, stopword filtering and word weighting for Turkish text classification on several different Turkish datasets. We report the results of extensive experiments.TUBITAK, IEE
A novel classifier based on meaning for text classification
No full textGaniz, Murat Can (Dogus Author) -- Akyokuş, Selim (Dogus Author) -- Conference full title: International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2015; Madrid; Spain; 2 September 2015 through 4 September 2015.Text classification is one of the key methods used in text mining. Generally, traditional classification algorithms from machine learning field are used in text classification. These algorithms are primarily designed for structured data. In this paper, we propose a new classifier for textual data, called Supervised Meaning Classifier (SMC). The new SMC classifier uses meaning measure, which is based on Helmholtz principle from Gestalt Theory. In SMC, meaningfulness of terms in the context of classes are calculated and used for classification of a document. Experiment results show that new SMC classifier outperforms traditional classifiers of Multinomial Naïve Bayes (MNB) and Support Vector Machine (SVM) especially when the training data limited
Intelligent focused crawler: Learning which links to crawl
No full textAkyokuş, Selim (Dogus Author) -- Ganiz, Murat C. (Dogus Author) -- Conference full title: 2011 International Symposium on Innovations in Intelligent Systems and Applications (INISTA 2011) Istanbul, Turkey, 15 - 18 June 2011A web crawler is defined as an automated program that methodically scans through Internet pages and downloads any page that can be reached via links. With the exponential growth of the Web, fetching information about a special-topic is gaining importance. A focused crawler is a web crawler that attempts to download only web pages that are relevant to a predefined topic or set of topics. In order to determine a web page is about a particular topic, focused crawlers use classification techniques. In this study we focus on the classification of links instead of downloaded web pages to determine relevancy. We combine a Naïve Bayes classifier for classification of URLs with a simple URL scoring optimization to improve the system performance. Our results demonstrate that proposed approach performs better.TUBITAK, IEEE
A novel semantic smoothing kernel for text classification with class-based weighting
No full textAltınel, Berna (Dogus Author), Diri, Banu (Dogus Author), Ganiz, Murat Can (Dogus Author) -- #articleinpress#Altınel, Berna (Dogus Author), Diri, Banu (Dogus Author), Ganiz, Murat Can (Dogus Author)In this study, we propose a novel methodology to build a semantic smoothing kernel to use with Support Vector Machines (SVM) for text classification. The suggested approach is based on two key concepts; class-based term weighting and changing the orthogonality of vector space. A class-based term weighting methodology is used for transformation of documents from the original space to the feature space. This class-based weighting basically groups terms based on their importance for each class and consequently smooths the representation of documents. This is accomplished by changing the orthogonality of the Vector Space Model (VSM) with introducing class-based dependencies between terms. As a result, on the extreme case, two documents can be seen as similar even if they do not share any terms but their terms are similarly weighted for a particular class. The resulting semantic kernel can directly make use of class information in extracting semantic information between terms, therefore it can be considered as a supervised kernel. For our experimental evaluation, we analyze the performance of the suggested kernel with a large number of experiments on benchmark textual datasets and present results with respect to varying experimental conditions. To the best of our knowledge, this is the first study to use class-based term weighting in order to build a supervised semantic kernel for SVM. We compare our results with kernels that are commonly used in SVM such as linear kernel, polynomial kernel, Radial Basis Function (RBF) kernel and with several corpus-based semantic kernels. According to our experimental results the proposed method favorably improves classification accuracy over linear kernel and several corpus-based semantic kernels in terms of both accuracy and speed
Pulse width modulation using a recently developed CMOS core circuit
No full textGöknar, İzzet Cem (Dogus Author) -- Minaei, Shahram (Dogus Author) -- Yıldız, Merih (Dogus Author) -- Akçakaya, Ergül (Dogus Author)In this paper a new approach for a Pulse Width Modulation (PWM) circuit operating in current-mode using a CMOS classifier core circuit, and its application to level crossing are presented. The proposed architecture is much simpler than existing PWM methods and the generated PWM signal can be controlled electronically through the control currents of a core circuit. Measurements performed with DU-TCC 1209, an IC designed and manufactured using 0.35 μm AMS technology parameters, show a perfect match with theoretical results
Doğuş himota insansız deniz taşıtı
No full textTüzel, Nuri (Dogus Author) -- Akıncı, Gürcan Şahin (Dogus Author) -- Tükel, Dilek (Dogus Author) -- Conference full title: Elektrik Elektronik Mühendisliği Kongresi, EEMKON 2017: İstanbul, Türkiye, 16-18 Kasım 2017.İnsansız deniz araçları, insanlı araçlarla yapılamayacak askeri operasyonlarda veya insan hayatını riske eden görevlerde yer alabilmektedir. Projemizin hedefi Raspberry Pi 3 kullanarak insansız deniz aracı yapmaktır. Aracımız operatör tarafından manüel olarak veya önceden belirlenmiş yörüngeleri otomatik olarak takip edebilme yeteneğine sahiptir.Unmanned sea vehicles have the potential, and in some cases the demonstrated ability, to reduce risk to manned forces to accomplish military missions, perform tasks. Our project mission is to develop an unmanned sea vehicle that have onboard PC-based Raspberry Pi 3 controller. Dogus Himota USV can be operated manually or it can follow autonomously predetermined trajectories
Modeling and simulation strategies in wireless propagation and coverage planning
No full textUluışık, Çağatay (Dogus Author) -- Sevgi, Levent (Dogus Author)This paper presents two-dimensional (2D) user-friendly, tailored with attractive and useful graphical user interfaces (GUI), Matlab propagation packages based on the classical knife-edge model and the parabolic equation method that can handle realistic propagation scenarios with non-flat, arbitrarily-designed and user-selected terrain profiles. Any 2D propagation scenario may be built by the user, and field vs. range and/or height profiles can be simulated
Evaluation of classification models for language processing
No full textKilimci, Zeynep Hilal (Dogus Author) -- Ganiz, Murat Can (Dogus Author) -- Conference full title: International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2015; Madrid; Spain; 2 August 2015 through 4 August 2015.Naïve Bayes is a commonly used algorithm in text categorization because of its easy implementation and low complexity. Naïve Bayes has mainly two event models used for text categorization which are multivariate Bernoulli and multinomial models. A very large number of studies choose multinomial model and Laplace smoothing just based on the assumption that it performs better than multivariate model under almost any conditions. This study aims to shed some light into this widely adopted assumption by analyzing Naïve Bayes event models and smoothing methods from a different perspective. To clarify the difference between events models of Naïve Bayes, their classification performance are compared on different languages - English and Turkish - datasets. Results of our extensive experiments demonstrate that superior performance of multinomial model does not observed all the time. On the other hand, multivariate Bernoulli model can perform well when combined with an appropriate smoothing method under different training data size conditions
Application of the SpecHybrid algorithm to text document clustering problem
No full textUykan, Zekeriya (Dogus Author) -- Ganiz, Murat C. (Dogus Author) -- Conference full title: 2011 International Symposium on Innovations in Intelligent Systems and Applications (INISTA 2011) Istanbul, Turkey, 15 - 18 June 2011In this paper, we present a relaxed version of the SpecHybrid Algorithm originally proposed for wireless cellular systems, and apply it to text document clustering problem. We conduct several experiments on two different datasets; a widely used benchmark dataset in English, and a Turkish textual dataset commonly used in text classification. Our results show that the proposed algorithm gives superior performance in text document clustering as compared to the standard k-means algorithm for any number of clusters while giving a comparable or better performance as compared to the standard EM algorithm for relatively large number of clusters depending on the similarity matrices used.TUBITAK, IEEE
Low volume store planning for workload balancing and truckload
No full textOkutkan, Caner (Dogus Author) -- Çelepçıkay, Ömer (Dogus Author) -- Çimen, Egemen Berki (Dogus Author)In this study, it is aimed to decrease the number of days planned to be shipped in stores with low sales volume, to create a balanced work load and to increase vehicle occupancy capacity . Particularly, Less than Truckload problem is analyzed and model in this study. The techniques used in this study showed that adaptability can be achieved with shipment and store planning as well as green supply chain environment to minimize Less than Truckload (LTL) as well as gas consumption. A mathematical model was created in the direction of multi objective problem and solved in the GAMS environment. Moreover, scenario analyzes were made and the study was tested on five specially identified regions on Turkey
- …
