Çavur, Mahmut

Çavur, Mahmut

Name Variants

Çavur, Mahmut
M.,Çavur
M. Çavur
Mahmut, Çavur
Cavur, Mahmut
M.,Cavur
M. Cavur
Mahmut, Cavur
cavur, mahmut
Çavur, Mehmet
Cavur, M.

Email Address

Mahmut.cavur@khas.edu.tr

Main Affiliation

Management Information Systems

Status

Former Staff

Full item page

Scholarly Output

17

Articles

12

Citation Count

0

Supervised Theses

3 Scholarly Output Search Results

Now showing 1 - 10 of 17

Citation - WoS: 26
Citation - Scopus: 31
The Geothermal Artificial Intelligence for Geothermal Exploration
(Pergamon-Elsevier Science Ltd, 2022) Moraga, J.; Çavur, Mahmut; Duzgun, H. S.; Cavur, M.; Soydan, H.
Exploration of geothermal resources involves analysis and management of a large number of uncertainties, which makes investment and operations decisions challenging. Remote Sensing (RS), Machine Learning (ML) and Artificial Intelligence (AI) have potential in managing the challenges of geothermal exploration. In this paper, we present a methodology that integrates RS, ML and AI to create an initial assessment of geothermal potential, by resorting to known indicators of geothermal areas namely mineral markers, surface temperature, faults and deformation. We demonstrated the implementation of the method in two sites (Brady and Desert Peak geothermal sites) that are close to each other but have different characteristics (Brady having clear surface manifestations and Desert Peak being a blind site). We processed various satellite images and geospatial data for mineral markers, temperature, faults and deformation and then implemented ML methods to obtain pattern of surface manifestation of geothermal sites. We developed an AI that uses patterns from surface manifestations to predict geothermal potential of each pixel. We tested the Geothermal AI using independent data sets obtaining accuracy of 92-95%; also tested the Geothermal AI trained on one site by executing it for the other site to predict the geothermal/non-geothermal delineation, the Geothermal AI performed quite well in prediction with 72-76% accuracy.(c) 2022 Elsevier Ltd. All rights reserved.
Citation - WoS: 1
Yereraltı Maden İşçilerini Gerçek Zamanlı Takip Etmek İçin Rfıd Teknolojisine Dayalı Özgün Bir Entegrasyon Metodolojisi
(Gazi Universitesi, 2018) Çavur, Mahmut; Çavur, Mahmut
In recent years many companies want to keep track of their employees sources and working machines due to various reasons like security coordination performance monitoring. The purpose and requirements are the main factors that determine the methodology of tracking. The real-time tracking can be determined with high precision in open areas with the global positioning system (GPS). However previous research and developments for indoor tracking have mostly focused on infrared wireless LAN and ultrasonic. In this study a Radio-Frequency Identification (RFID) protocol and interface are integrated into an open source Information Systems (IS) software. A tight coupling methodology is developed for integration of RFID into an open source software. The use of open source software as a common interface also provides better spatial display and analysis capabilities. The tracking algorithm is completely unique original and it is encoded in the Java programming language. In the algorithm the accuracy of locating the proximity direction of miners and whether the RFID tag is on the right and left of the last point of RFID receiver is determined with 20 m accuracy. The system was tested in an underground salt mine. The developed methodology and system are now being commercialized in Turkey.
Tarafsız 3d Mineral Harita Tahminleri Elde Etmek için Random Forest Tree Sınıflandırması Kullanılarak Epoksi Bloklardaki Dikey Kesitlerin Değerlendirilmesi
(2021) Camalan, Mahmut; Çavur, Mahmut; Çavur, Mahmut
Alansal mineral haritaları, epoksi reçinenin dibine çöken cevher tanelerinin yüzeylerini içeren parlak kesitlerinden yapılmaktadır.Fakat, ağır mineraller nispeten dibe çökebilmekte ve parlak yüzeyi ağır mineraller açısından zengin yapabilmektedir. Bu ise parlakkesitlerden hesaplanan alansal (2D) mineral haritalarının, hacimsel (3D) haritaların taraflı tahminleri haline gelmesine sebepolabilmektedir. Bu çalışma, parlak kesite dik olarak (parçacıkların çökelme yönü boyunca) alınan rastgele bir kesitin bir kromitcevheri numunesinin 3D mineral haritasının tarafsız bir tahmini olarak kullanılıp kullanılamayacağını test etmeyi amaçlamaktadır.Bu çalışmanın amacı için, dikey kesitlerin 2D haritaları, öncesi ve sonrası görüntü işleme araçlarıyla bütünleşmiş Random Forestsınıflandırmasıyla elde edilmiştir. Daha sonra, 2D haritalar, stereolojik hatalar olmadığı varsayılarak 3D mineral haritalarınadönüştürülmüştür. 3D haritalardan tahmin edilen modal mineraloji ve tane boyu dağılımları, sırasıyla XRD ve kuru elemeanalizlerinden tahmin edilen sonuçlarla karşılaştırılmıştır. Herhangi bir 2D harita gerçek analizlere yakın modal mineraloji ve taneboyu dağılımı veriyorsa, bu 2D harita cevher numunesinin 3D haritasının tarafsız bir tahmini olarak seçilmiştir. Bu çalışmanınsonuçları herhangi bir dikey kesitin, ağır minerallerin öncelikli olarak çöktüğü parlak kesitten farklı olarak gerçek 3D haritanıntarafsız bir tahmini olacağını desteklemektedir.
Gelı̇şmekte Olan Ülkelerde Matematı̇k Başarısını Etkı̇leyen Faktörlerı̇n Araştırılmasında Makı̇ne Öğrenme Teknı̇klerı̇nı̇n Kullanılması: Türkı̇ye, Meksı̇ka, Tayland ve Bulgarı̇stan Örneğı̇
(2023) Arpa, Tuba; Çavur, Mahmut; Çavur, Mahmut
Matematik tüm eğitim sistemlerinin vazgeçilmez bir parçasıdır. Çünkü matematik, hem günlük yaşamın önemli bir unsuru hem de pek çok meslek ve alan için olmazsa olmaz bir temeli teşkil etmektedir. Bu nedenle, matematik başarısını etkileyen unsurları belirlemenin, ülkelerin gelişimine katkı sağlayacağı söylenebilir. Bu doğrultuda, bu çalışmada PISA 2018 verileri kullanılarak, benzer eğitim sistemi ve ekonomik gelişmişliğe sahip dört ülke olan Türkiye, Bulgaristan, Meksika ve Tayland'ın matematik başarılarını etkileyen faktörleri makine öğrenmesi modelleri ile belirlemek, bu modellerin başarılarını karşılaştırmak amaçlanmıştır. İlgili alanyazında bu amaç için sıklıkla sınıflandırma algoritmaları tercih edildiği görülmektedir. Bu çalışmada hem sınıflandırma hem de regresyon modelleri kullanılmıştır. Çalışmada, regresyon algoritması olarak doğrusal regresyon, destek vektör regresyonu, karar ağacı regresyonu ve rastgele orman regresyonu; sınıflandırma algoritması olarak ise lojistik regresyon, destek vektör sınıflandırması, karar ağacı sınıflandırması ve rastgele orman sınıflandırması kullanılmıştır. Ayrıca, matematik başarısını tahmin etmek için en önemli faktörlerin belirlenmesinde XGradient Boosting algoritması kullanılmıştır. Son olarak, eksik verilerin doldurulmasında, K-Means metodu tercih edilmiştir. Çalışmanın sonuçlarına göre, dört ülke için de matematik başarına en büyük katkı sağlayan değişkenlerin öğrencinin ekonomik, sosyal ve kültürel statüsü, öğrencinin evde sahip olduğu çalışma materyali, öğrencinin sahiplik hissi ve ailenin refah düzeyi olduğu bulunmuştur. Model başarısı açısından hem regresyon hem de sınıflandırma açısından en yüksek başarıya sahip algoritmanın rastgele ormanlar olduğu bulunmuştur. Ayrıca, sınıflandırma algoritmaları ikili ve üçlü sınıflandırma üzerinden incelenmiş, ikili sınıflandırmanın daha yüksek başarıya sahip olduğu görülmüştür. Sonuç olarak, çalışmamızda elde edilen bulgular matematik başarısını tahmin etmede kullanılacak en uygun algoritmanın seçimi v konusunda önemli bir öngörü sunmaktadır. Ayrıca, çalışmanın bulguları, eğitim politikalarının geliştirilmesi ve öğrenci başarısını artırmak için uygulayıcı ve politika yapıcılara önemli iç görüler sağlamaktadır.
Citation - Scopus: 25
Land Use and Land Cover Classification of Sentinel 2-A: St Petersburg Case Study
(International Society for Photogrammetry and Remote Sensing, 2019) Çavur, Mahmut; Çavur, Mahmut; Düzgün, Hafize Şebnem; Kemeç, Serkan; Demirkan, Doğa Çağdaş
Land use and land cover (LULC) maps in many areas have been used by companies, government offices, municipalities, and ministries. Accurate classification for LULC using remotely sensed data requires State of Art classification methods. The SNAP free software and ArcGIS Desktop were used for analysis and report. In this study, the optical Sentinel-2 images were used. In order to analyze the data, an object-oriented method was applied: Supported Vector Machines (SVM). An accuracy assessment is also applied to the classified results based on the ground truth points or known reference pixels. The overall classification accuracy of 83,64% with the kappa value of 0.802 was achieved using SVM. The study indicated that of SVM algorithms, the proposed framework on Sentinel-2 imagery results is satisfactory for LULC maps.
Citation - WoS: 4
Citation - Scopus: 5
Development of a Supervised Classification Method To Construct 2d Mineral Maps on Backscattered Electron Images
(Tubitak, 2020) Camalan, Mahmut; Çavur, Mahmut; Çavur, Mahmut
The Mineral Liberation Analyzer (MLA) can be used to obtain mineral maps from backscattered electron (BSE) images of particles. This paper proposes an alternative methodology that includes random forest classification, a prospective machine learning algorithm, to develop mineral maps from BSE images. The results show that the overall accuracy and kappa statistic of the proposed method are 97% and 0.94, respectively, proving that random forest classification is accurate. The accuracy indicators also suggest that the proposed method may be applied to classify minerals with similar appearances under BSE imaging. Meanwhile, random forest predicts fewer middling particles with binary and ternary composition, but the MLA predicts more middling particles only with ternary composition. These discrepancies may arise because the MLA, unlike random forest, may also measure the elemental compositions of mineral surfaces below the polished section.
Citation - Scopus: 3
Sentinel-1 Sar Verileri Kullanilanarak Maden Kaymalarini ve Deformasyonlarini İzleme [monıtorıng Of Mıne Landslıde And Deformatıon Usıng Sentınel-1 Sar Data]
(Baski, 2019) Çavur, Mahmut; Çavur, Mahmut; Camalan, Mahmut; Ketizmen, Hakkı; Ağıtoğlu, Suud
In this study, an original DInSAR method was used to monitor landslides and deformation in a coal mine area. The open-pit mine operation belonging to the Ciner Group in Silopi, Sirnak was selected fort he case study. Between 21 November 2017 and December 31, 2017, 2-month Sentinel-1 data were analyzed every 12 days and interferometric results were obtained. It has been shown that the DInSAR method can be used effectively in order to monitor the mineral movements by using satellite images. The results of the analysis were reported in mm and accuracy analysis was performed on the field. SNAP, Cygwin, and ArcGIS 10.4 software are used for reporting and analysis purposes. The maximum subsidence was measured by radar as 45 mm. The mean subsidence rate of one class was found to be 45 mm as landslide and 46 mm as uplift where cracks most severely developed. The proposed method is an effective method for mining in order to determine the effects that may occur as a result of landslides, displacement, and uplift caused by underground and surface mining.
Citation - WoS: 8
Citation - Scopus: 12
Displacement Analysis of Geothermal Field Based on Psinsar and Som Clustering Algorithms a Case Study of Brady Field, Nevada-Usa
(MDPI, 2021-02) Çavur, Mahmut; Çavur, Mahmut; Moraga, Jaime; Düzgün, H. Şebnem; Soydan, Hilal; Jin, Ge
The availability of free and high temporal resolution satellite data and advanced SAR techniques allows us to analyze ground displacement cost-effectively. Our aim was to properly define subsidence and uplift areas to delineate a geothermal field and perform time-series analysis to identify temporal trends. A Persistent Scatterer Interferometry (PSI) algorithm was used to estimate vertical displacement in the Brady geothermal field located in Nevada by analyzing 70 Sentinel-1A Synthetic-Aperture Radar (SAR) images, between January 2017 and December 2019. To classify zones affected by displacement, an unsupervised Self-Organizing Map (SOM) algorithm was applied to classify points based on their behavior in time, and those clusters were used to determine subsidence, uplift, and stable regions automatically. Finally, time-series analysis was applied to the clustered data to understand the inflection dates. The maximum subsidence is -19 mm/yr with an average value of -6 mm/yr within the geothermal field. The maximum uplift is 14 mm/yr with an average value of 4 mm/yr within the geothermal field. The uplift occurred on the NE of the field, where the injection wells are located. On the other hand, subsidence is concentrated on the SW of the field where the production wells are located. The coupling of the PSInSAR and the SOM algorithms was shown to be effective in analyzing the direction and pattern of the displacements observed in the field.
Forecasting Employees' Promotion Based on the Personal Indicators by Using a Machine Learning Algorithm
(Kadir Has Üniversitesi, 2022) IBRIR, YASMINE AYA; Çavur, Mahmut; cavur, mahmut
Job promotion is considered one of the most important issues of importance in any organization, as it is vital for administrative development, and a means of motivating the worker for self-development and willingness to bear the burden and responsibility of work and the position attached to it, and thus it contributes to providing the necessary needs of the forces of mankind to occupy positions higher on the career ladder. Thus, this study aims to set up a sufficient framework to predict the promotion of an employee in an organization based on a variety of characteristics such as, but not limited to, the number of training, previous year rating, duration of service, awards earned, and average training score. Hence, this framework can be used and generalized to all prediction problems, not just our problem of predicting employee promotion. In this study, we used promotion data provided by Analytics Vidhya Data to test and prove the success of the framework. Our methodology is mainly composed of five phases: Input data, Data Pre processing, Data Manipulation, Data Modeling, and finally Data Evaluation. We constructed a new number of features in this study. Then we used several features including creating features and providing insights into the promotion and commitment of employees and using supervised learning techniques, namely XGBoost, Random Forest, Decision Tree, Logistic Regression, AdaBoost, and Gradient Boosting. Experimental results show that the XGBoost model has a higher accuracy of 94%, proving to be the most efficient. The result is accentuated by the high validation score similar to accuracy and efficiency. It is a very important and valuable study as it is the first study to predict employee promotion using the XGBoost classifier method.
Citation - WoS: 9
Citation - Scopus: 12
Rssi-Based Hybrid Algorithm for Real-Time Tracking in Underground Mining by Using Rfid Technology
(Elsevier, 2022) Cavur, Mahmut; Çavur, Mahmut; Demir, Ebubekir
Knowing the precise and real-time location of underground mining workers is essential for their health and safety in any emergency. However, the standard Global Positioning System (GPS) is insufficient for such indoor environments as it requires new infrastructure based on different technologies and algorithms. Instead, Radio Frequency Identification (RFID)-based real-time indoor localization systems and a hybrid algorithm are developed. The received-signal-strength (RSS) based positioning techniques are investigated and applied in indoor environments. A unique hybrid approach based on fingerprinting is proposed and developed to solve the disadvantages of the existing techniques. Consequently, the accuracy of this one-of-a-kind algorithm is found to be 2.52 m in an office and 3.13 m in an underground mine. We also compared the proposed hybrid algorithm to the Weighted K-Nearest Neighbor (WKNN). WKNN, on the other hand, has an accuracy of 4.01 m in the office and 4.33 m in underground mining environments. (C) 2022 Elsevier B.V. All rights reserved.