Browsing by Subject "Morphological structures"

Now showing 1 - 4 of 4
  • No Thumbnail Available
    Item
    Investigation of erosive wear behavior of granulated blast furnace slag on hard coated and uncoated steels; [Yüksek fırın granüle cürufunun sert kaplama uygulanmış ve uygulanmamış çeliklere karşı erozif aşındırma davranışının incelenmesi]
    (Gazi Universitesi Muhendislik-Mimarlik, 2019) Demirsöz R.; Polat R.; Türk A.; Erdoğan G.
    One of the biggest problems in industrial plants is wear that directly affects equipment life. In integrated plants such as blast furnaces, many types of wear are encountered, one of which is erosive wear. Slurry erosion is one of the effective types of erosive wear. During the transfer of the slag into the liquid medium, the transport equipment is exposed to slurry erosion. The high amount of erosion seriously shortens the life of the equipment where the transportation is carried out and increases the maintenance-repair costs. In this study, blast furnace slag was used in the slurry erosive wear test system and the abrasivity of the St 37-2 reference standard pipe material, Hardox 400 and W2C-NiCrBSi coating materials was investigated. The slag was sieved to an average particle size of 505 μm and slurry containing 10%, 20% and 30% by weight of slag was prepared. The experiments were carried out at 2 m/s and 4 m/s peripheral speed values at the partical normal impact angle (90°). It has been determined by these experimental studies that the abrasivity of the slag prepared at a certain grain size is directly related to the concentration and the rotation speed (material conveyance speed) and that the wear is also increased by increasing the concentration and rotation speed. The wear values of the materials used as specimens are determined from high to low respectively as St 37-2, Hardox 400 and W2C-NiCrBSi coating material. In addition, the morphological structures of the worn surfaces were examined using SEM(Scanning Electron Microscope) in order to understand the wear mechanisms. © 2019 Gazi Universitesi Muhendislik-Mimarlik. All rights reserved.
  • No Thumbnail Available
    Item
    A hybrid sentiment analysis method for Turkish
    (Turkiye Klinikleri, 2019) Erşahin B.; Aktaş Ö.; Kilinç D.; Erşahin M.
    This paper presents a hybrid methodology for Turkish sentiment analysis, which combines the lexicon-based and machine learning (ML)-based approaches. On the lexicon-based side, we use a sentiment dictionary that is extended with a synonyms lexicon. Besides this, we tackle the classification problem with three supervised classifiers, naive Bayes, support vector machines, and J48, on the ML side. Our hybrid methodology combines these two approaches by generating a new lexicon-based value according to our feature generation algorithm and feeds it as one of the features to machine learning classifiers. Despite the linguistic challenges caused by the morphological structure of Turkish, the experimental results show that it improves the accuracy by 7% on average. © TÜBİTAK
  • No Thumbnail Available
    Item
    A study on the mechanical and morphological behavior of polyurethane-encapsulated cholesteric liquid crystal composite films
    (SAGE Publications Ltd, 2020) Kemiklioglu E.; Kemiklioğlu U.; Sayman O.
    This study explores the preparation and mechanical behavior of free-standing polymer-dispersed liquid crystal (PDLC) film membranes. Polyurethane (PU) was used as a thermoplastic polymer matrix to form these free-standing film membranes. Cholesteryl oleyl carbonate (COC) and cholesteryl pelargonate were used as liquid crystals (LCs) with different molecular weights. PDLC membranes were produced by casting method after LCs and polymer were mixed in the tetrahydrofuran solvent at room temperature. These membranes were formed at different concentration ratios of polymer and LCs. The relationship among the phase separation, LCs and polymer contents as well as the LCs molecular weights was investigated. The morphological structures of these membranes were studied using scanning electron microscopy (SEM). SEM images exhibited that the shapes of LC droplets embedded in PU matrix were more uniform and smaller than those of the membranes which include LC with lower molecular weight. The mechanical properties of the PDLC membranes were determined by carrying out the tensile tests. It was found that the membranes which include COC LC were more flexible. © The Author(s) 2020.
  • No Thumbnail Available
    Item
    Turkish medical text classification using BERT; [BERT modeli ile Türkçe medikal metin siniflandirma]
    (Institute of Electrical and Electronics Engineers Inc., 2021) Celikten A.; Bulut H.
    Medical text classification is mostly carried out on English data sets. The limited number of studies in Turkish is due to the compelling morphological structure of Turkish for natural language processing and the limited number of data sets in the medical domain. In addition, the use of domain specific words and abbreviations makes natural language processing studies more challenging. In this study, a classification model is implemented to assign article abstracts to appropriate disease categories using multilingual BERT and BERTurk models on a data set consisting of Turkish medical article abstracts. As a result of the experimental study, 0.82 and 0.93 F-score are obtained for multilingual BERT and BERTurk, respectively. The results show that the BERTurk is more successful than other compared models for Turkish medical text classification. © 2021 IEEE.