Yazar "Arslan, Enis" seçeneğine göre listele

Listeleniyor 1 - 7 / 7
  • Küçük Resim
    Öğe
    A Comparison of Graph Centrality Algorithms For Semantic Distance
    (Çanakkale Onsekiz Mart Üniversitesi, 2020-12-31) Arslan, Enis; Turan, Erhan; Tülü, Çağatay; Orhan, Umut
    Semantic networks are kind of datasets used for natural language processing (NLP). Distance measurement for semantic networks, which are generally based on a graph structure, is a vital requirement for semantic analysis on concepts. Centrality measures can be used for calculating the semantic distance between concepts in a semantic network. In this paper, we evaluated graph centrality algorithms including PageRank, Hyperlink-induced Topic Search (HITS), and Betweenness Centrality on a semantic network, which was created from a Turkish dictionary dataset. Centrality measures special to these algorithms are used to calculate the semantic distance between synonym pairs in the semantic network. Also, we have used a simple centrality method beside the other three popular centrality algorithms to find out the most accurate and cost-effective method on our semantic network. Working on a bipartite model of the network which increases the complexity of implementation for centrality algorithms and performing calculations on a semantic network, that can be expanded with new nodes and edges, are two major challenges to overcome. Considering all these conditions, results from each algorithm are compared to pick out an optimal method for the semantic network.
  • [ X ]
    Öğe
    Ağırlıklı Çizgeler Ile Canlı Türkçe Sözlük Ağı Tasarımı
    (2019) Orhan, Umut; Tahiroğlu, Bekir Tahir; Arslan, Enis; Turan, Erhan
    Dil bilimi ve yapay zekanın birleşmiş bir konusu olarak bilinen doğal dil işleme (DDİ) alanındaki neredeyse tüm çalışmalarda morfolojik ve anlamsal analiz önemli konulardır. Özellikle Türkçe gibi karmaşık morfosentaktik özelliklere sahip sondan eklemeli bir dilin mofolojik analizi diğer tüm DDİ işlemlerini etkilemektedir. Anlamsal sözcük ağları da DDİ alanındaki diğer bir önemli başlıktır. Bilinen sistematik ilk çalışma olması sebebiyle WordNet genel olarak anlamsal ağlara rol model olmuştur. Başta sadece İngilizce için hazırlanan WordNet zamanla diğer dünya dillerine uyarlanmıştır. Fakat uzmana dayalı sürdürülen WordNet çalışmaları, yeterli insan gücünün teminindeki zorluklar yüzünden sekteye uğramış, anlamsal ilişkilerin bilgisayar destekli tespiti başlığı önemli araştırma alanlarından birisi olmuştur. Türkçe üzerine yeterli çalışmanın yapılmadığını belirleyerek öncelikle morfolojik analizin geliştirilmesi, daha sonra da sözlük tanımlarının analiziyle bir Türkçe WordNet?in bilgisayarlı yöntemlerle hazırlanabilmesi üzerine odaklandık. Morfolojik analizde, çeşitli araştırmacılar tarafından önerilen Türkçenin morfosentaktik yapısını tanımlayan sonlu durum makineleri kullanılmıştır. Buna paralel olarak Viterbi tabanlı belirsizlik giderme ve tekrar sayılarına dayalı yeni gövde keşifleri gerçekleştirilmiştir. Ayrıca TDK Güncel Sözlük verisi üzerine yapılan analiz ile bazı tanım kalıpları belirlenmiş ve sözcükler arası birçok ilişki tespit edilmiştir. Bulunan bu ilişkiler MentionSense mimarisiyle oluşturulan anlamsal ağ üzerinde kontrol edilerek sadece madde başları arasında tanımlanan ilişkiler olmaktan öteye geçerek anlamdan anlama ilişkiler haline getirilmiştir. Bunun için ağdaki tüm düğümler ağırlıklandırılmış ve bulunan her ilişkideki iki sözcüğün çift yönlü olarak birbirlerine ulaşabilirliği gözetilmiştir. Bu anlamsal ağ sayesinde eş anlamlılık, ast/üst, karşıt anlamlılık gibi temel anlamsal ilişkilerin tespit edilmesi ve sayısının arttırılması hedeflenmiştir. Son olarak derlemden keşfedilen yeni madde başlarından türemiş veya bileşik yapıda olanlar için tanım önerilerinin nasıl hazırlanabileceği anlatılmıştır. Tartışma kısmında, çalışmada elde edilen tüm sonuçların değerlendirmesi yapılmıştır.
  • Küçük Resim
    Öğe
    Application of spectral analysis to determine geothermal anomalies in the Tuzla region, NW Turkey
    (Springer Heidelberg, 2019) Erenoğlu, Ramazan Cüneyt; Arslan, Niyazi; Erenoğlu, Oya; Arslan, Enis
    We used remote sensing data from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) satellite to identify the mineral properties and geothermal anomalies related to hot springs in the Tuzla area, including the fault system with NW-SE trend, which is located southwest of Canakkale, NW Turkey. In the study area, the lithological units of the Tuzla geothermal field and the surrounding area consist of Miocene volcanic (trachyandesite, trachyte, and ignimbrites) and Pliocene sedimentary (conglomerate, sandstone, and mudstone) rocks with siliceous, argillaceous, and ferrous alteration linked to the geothermal fluid. ASTER visible/near-infrared (VNIR), short-wave infrared (SWIR), and TIR bands were analyzed by different approaches in order to highlight hot springs in the study area. From these approaches, band ratios were constructed from ASTER VNIR, SWIR, and TIR bands for obtaining geological properties of the region. The geothermal areas were defined by the minimum noise fraction (MNF) and principal component analysis (PCA) methods that was extracted from 5 thermal infrared (TIR) bands as well. Land surface temperatures (LST) support the results from MNF and PCA that were estimated for 5 TIR bands using the inversion of Planck function method. Four days of data including daytime and nighttime satellite images from ASTER were used for the analysis. The used procedure displayed a good match with the ground reality based on field observations in the Tuzla Region.
  • Küçük Resim
    Öğe
    Assessment of hotspots using sparse autoencoder in industrial zones
    (Springer, 2019) Arslan, Enis; Erenoğlu, Ramazan Cüneyt
    Remote sensing satellite systems can be used to detect industrial zones by means of thermal infrared bands. There are several satellite systems loaded with thermal infrared sensors such as Landsat and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). In this study, ASTER thermal infrared data were converted to land surface temperature (LST) in order to determine hotspots caused by industrial zones. High LST values surrounded by low LST values are called hotspots here. These hotspots can be determined by applying different methodologies. One of these methods of sparse autoencoder can be used to indicate hotspots using different sizes of hidden layers. The principle of sparse autoencoder depends on unlabeled data in unsupervised learning. It does not need any information about labeled data as in supervised learning. The autoencoder reproduces its output with the same dimensions as the input image by managing the size of the hidden layer. The reconstruction of the image depends on the minimization of a cost function. The size of the hidden layer sets the fitting degree of the function for the reproduced image. A low-order reproduced image is the main target for hotspot detection. In this study, the difference between the original image and the reproduced image was analyzed for hotspot detection. Sparse autoencoder was successfully applied to ASTER thermal band 10 for hotspot detection in 7 pre-defined sites of a region known for steel industry for the two different days.
  • Küçük Resim
    Öğe
    Flood Analysis and Mapping Using Sentinel Imagery: A Case Study from Tarsus Plain, Turkey
    (Çanakkale Onsekiz Mart Üniversitesi, 2021-06-30) Arslan, Enis; Erenoğlu, Ramazan Cüneyt
    Floods are natural disasters that corrupt vegetation, cause loss of lives, and harm economies. There are many cases floods originate, sometimes natural, sometimes man-made. The use of agricultural fields unconsciously, land cover modifications, incorrect city planning can be listed as unnatural reasons. Modeling and mapping the floods, real-time monitoring with satellite are cost-efficient ways of decreasing the causes of floods and helping the authorities to give the exact decisions during or after the event. Synthetic-aperture radar (SAR) satellite imagery helps in monitoring disasters like flooding. The allweather operating capability provides cloud-free day and night imagery, even in the worst weather conditions. In this paper, Sentinel-1 satellite imagery provided by European Space Agency (ESA) is used to investigate the flood event that happened in January 2020 in the Tarsus agricultural field (West Cukurova Region) of Mersin, Turkey. Sentinel-1 imagery for the nearest dates is collected, pre-processed, and thresholded with Otsu’s method and a flood map is obtained. Sentinel-2 satellite imagery for the same study area is used to verify the Sentinel-1 output composite. Spectral indices are applied on Sentinel-2 composite and classification is done with Random Forests, CART, Support Vector Machine (SVM) and Naive Bayes algorithms. Random Forest and SVM algorithms provided the best classification result. Finally, Sentinel-1 and Sentinel-2 products are overlaid as change management.
  • [ X ]
    Öğe
    Flood Inundation Mapping with Supervised Classifiers: 2021 Gediz Plain Flood
    (2023) Arslan, Enis; Kartal, Serkan
    Generation of flood inundation maps is beneficial in flood risk assessment and evaluation. Flood inundation mapping can be achieved by many remote sensing techniques like change detection (CD) with thresholding and machine learning-based (ML) methods. Optical and synthetic aperture radar (SAR) imagery are widely used, provided by different satellite systems. This study used Sentinel-1 SAR and Sentinel-2 MSI satellite data in Google Earth Engine (GEE) with supervised ML algorithms. Gediz Plain, Turkey was selected as the study area, which is an agricultural area covered mostly by croplands. A flood event that occurred on February 2, 2021, was examined and flood inundation map for the study area was composed. Support Vector Machines (SVM), Random Forest (RF) and K-Nearest Neighbor (KNN) ML algorithms were selected and models were trained with manually created labelled data in GEE. Also, CD was applied on after and before event SAR images in a traditional approach. RF classifier performs best in Sentinel-2 MSI imagery with 94% overall classification accuracy where KNN classifier gives 93.3% accuracy value for Sentinel-1 SAR dataset, indicating the robustness of SAR imagery for all-weather conditions.
  • [ X ]
    Öğe
    The Use of Unmanned Aerial Vehicles in the 3D Documentation of Historical and Cultural Heritage: The Case of Ceyhan Kurtkulağı Caravanserai
    (2024) Arslan, Enis; Şekertekin, Aliihsan
    Detailed documentation of historical and cultural heritage is a necessity for further analysis, interpretation, and physical reconstruction. Usage of The Unmanned Aerial Vehicles (UAVs) in this role have been increasing day by day. As a result of its important contributions to the production of three-dimensional (3D) terrain models, it has reached an important point in the discipline of surveying engineering. Especially in 3D modeling and documentation of historical and cultural heritage, UAVs are advantageous tools in terms of time and cost when compared to the classical methods. The aim of this study is to generate a 3D model of the Kurtkulağı Caravanserai, located in the Kurtkulağı town of Ceyhan District of Adana, using UAV (DJI Phantom 4 RTK) and to reveal the importance of UAV in the documentation of this historical structure. In this context, according to a planned flight on the UAV, following the capture of the images of the caravanserai in a multiview aspect, a metric 3D model was produced using photogrammetric methods. Root Mean Square Errors (RMSEs) in X, Y, Z coordinates were calculated based on the Ground Control Points (GCPs) and corresponding model coordinates. The RMSEs for X, Y, Z coordinates were calculated 0.019m, 0.025m, and 0.033m, respectively.