Middle East Technical University Research Information System
Not a member yet
104897 research outputs found
Sort by
Detection of Potato Fields Using Sentinel-2 and Landsat 8 Data-Based Machine Learning Models in Semi-arid Region of Central Anatolia, Türkiye
No full textAccurate mapping of crop types is essential for agricultural monitoring, resource management, and food security. This study evaluates the performance of two ensemble machine learning algorithms—random forest (RF) and gradient tree boosting (GTB)—for classifying potato fields using multispectral satellite imagery from Landsat 8 and Sentinel-2 in the Konya Plain, Türkiye. The methodology involved generating median composite images from the 2020 growing season (June–August), followed by feature extraction from training samples collected via ground truth, satellite, and synthetically generated data. Model performances were assessed using overall accuracy, kappa coefficient, F-score, and user’s and producer’s accuracy metrics. Five-fold cross-validation was employed to evaluate model generalizability. A sensitivity analysis was conducted on the number of trees, and 100 was selected for model training. Results indicate that the Sentinel-2 random forest model achieved the highest average overall accuracy (0.94) and kappa coefficient (0.86) across folds, demonstrating robust and consistent classification. Feature importance analysis showed that red-edge and SWIR bands were the most effective for model performance. Sentinel-2 imagery combined with random forest provided a reliable and efficient approach for potato classification in arid agricultural regions. These findings support the integration of remote sensing and machine learning for operational agricultural monitoring and suggest avenues for future research involving temporal analysis and multi-sensor data fusion. This approach can be extended to other crop types and regions, enhancing the use of satellite-based crop mapping in precision agriculture. It enables policymakers and agricultural agencies to implement timely, data-driven strategies for agricultural planning and food security
Geochemical signature, detrital zircon U–Pb, and mica Ar–Ar age systematics from the Karadere Basement Unit of the İstanbul-Zonguldak Terrane (NW, Türkiye): evidence for an Avalonian-Cadomian active continental margin
No full textThe İstanbul-Zonguldak Terrane is located in northwest Anatolia and includes several outcrops with pre-Cambrian basement units. The Karadere Basement Unit represents the easternmost basement of this terrane and is composed of paragneiss, quartzite, and orthogneiss. The paragneiss and quartzite are cut by orthogneisses and disconformably overlain by Lower Ordovician to Middle Devonian cover units. Its pre-Cambrian to Palaeozoic tectonic evolution remains poorly understood, and therefore, this study reports whole-rock trace element signatures, detrital zircon U–Pb, and muscovite Ar–Ar age data for samples of paragneiss, quartzite, amphibolite, and orthogneiss. The whole-rock geochemical data indicate that the protoliths of the paragneiss and quartzite samples investigated here could have been derived from arc-related felsic magmatic units located in an active continental margin. The protoliths of amphibolites and orthogneiss samples are gabbro and granite, respectively, and their geochemical signatures are akin to those from continental arc igneous suites. The detrital zircon U–Pb data are characterized by predominantly Mesoproterozoic (84%) and lesser amount of Palaeoproterozoic and Neoproterozoic ages and suggest an early Tonian (youngest peak age of 915 ± 18 Ma) maximum depositional age. Moreover, zircon U–Pb dating analyses from an orthogneiss sample yielded an age of 603 ± 2 Ma, indicating the occurrence of mid-Ediacaran arc magmatism. This represents one of the oldest Ediacaran ages obtained from the metamagmatic basement rocks of the İstanbul-Zonguldak Terrane. Ar–Ar dating of muscovite from ortho- and paragneiss samples yielded ages of 533 ± 25 Ma and 530 ± 9 Ma, respectively, which are interpreted as early Cambrian cooling ages. Overall, the combined results reported here reveal that the Karadere Basement Unit is markedly different from NE African basement units but displays similarities to the basement rocks of both Avalonian and Amazonian terranes. It may have been formed and metamorphosed in an Avalonian–Cadomian active continental margin during the mid- to late Neoproterozoic and early Cambrian
(Un)Silencing Academia in Times of Epistemic Conflicts Navigating Online Violence
No full textIn today’s academic landscape, scholars are increasingly dependenton digital platforms to sustain their careers, enhance visibility, and meetinstitutional demands for public engagement. While the digitalization ofacademic work offers significant opportunities for dissemination andcollaboration, it also generates new forms of precarity and exposure. Higher Education’s 2020 study has shown that digitalharassment targeting those who critique dominant structures, such as white,middle-class, or male supremacy, has become routine, revealing the gendered,racialized, and classed dimensions of this phenomenon. In this chapter, we aimto employ an intersectional approach to understand and explore online academicharassment, drawing on a critical literature review of research concerningonline academic harassment within precarious and highly digitalized academia.Our examination posits that online academic harassment is not simplyinterpersonal but is also a socio-technical mechanism of governance that exploitsdigital affordances and neoliberal academic logic to enforce power hierarchiesand marginalize dissenting or vulnerable voices.</div
Ultrafast carrier dynamics and electronic properties of PtSe2/MoSe2and WSe22D TMDC layered structures on mica: combined THz spectroscopy and DFT study
No full textA detailed investigation of structure, electronic and optical properties of two transition metal dichalcogenide (TMDC) structures is presented in this study. Sample 1 consists of epitaxially grown bilayer of PtSe2 (2 monolayers) on MoSe2 (1 monolayer) deposited on mica substrate – reported here for the first time. Sample 2 comprises a trilayer of WSe2 grown on mica. The photoconductivities of both samples were characterized using optical pump-terahertz probe spectroscopy under above- and near-bandgap excitations at 400 nm and 800 nm. Both structures exhibit rapid carrier generation and relaxation dynamics, with notable variations depending on excitation wavelength and structures. Complementary density functional theory (DFT) calculations are performed to evaluate the electronic and optical properties of free-standing single layers of MoSe2, PtSe2 and WSe2 and their combined structures corresponding to Sample 1 and Sample 2. The experimental results show strong agreement with calculated band structures. This consistency between experiment and theory underscores the potential of these TMDC structures for future applications in terahertz and high-frequency electronic devices
A Color-Based Data Poisoning Backdoor Approach for Misleading Adversarial Privacy Prediction
No full textThe big data era has created a plethora of platforms providing access to large amounts of image data on the Internet, which may contain private information. Private images are a hot target for attackers, who train deep learning models to automatically predict which images among the sea of data on the Internet contain privacy-sensitive information. One effective method for dealing with these privacy prediction attacks is misleading the deep learning models through data poisoning at training time to cause the model to make mistakes during inference. In this paper, we propose a novel color-based data poisoning backdoor approach for misleading adversarial privacy prediction models, which causes insignificant visual difference to human sight. We have performed experiments with the publicly available Privacy Alert dataset with classic image classification models including AlexNet, VGG16, Resnet18, and GoogleNet to evaluate the effectiveness of the method. Experiment results show that our algorithm can preserve the functionality of the model on clean data and sets triggers into images successfully. By setting the labels of affected data items to the opposite one, the average privacy prediction accuracy drops from 75.9% to 64.2% when the affected data ratio reaches 0.3 on the test set, demonstrating the effectiveness of the proposed approach in misleading adversarial privacy prediction
Detection of herbicides in food samples using hybrid carbon dot sensors and test kit fabrication via paper-based analytical devices
No full textIn this study, hybrid carbon dots were prepared via surface modification of carbazole (Car@CD) and anthracene (Ant@CD) and applied for fluorometric determination of benfluralin, aclonifen, and pendimethalin herbicides in food samples. The characterizations of hybrid carbon dots were performed using different spectroscopic, thermal, and microscopic techniques. The optimal detection conditions and analytical parameters (LOD, LOQ, RSD%, etc.) were determined for the presented fluorometric methods. According to the results, the surface fluorophore groups increased the number of interaction points with analytes, and fluorescence responses were observed owing to PET processes between herbicides and nanosensors. The detection limits were calculated to be 6.00–9.40 nM, with a wide linear working range. The presented fluorometric techniques were applied to quantify benfluralin, aclonifen, and pendimethalin in food samples, and the novel methods were validated using GC–MS analyses and spike/recovery tests. Finally, the fluorescence-quenching responses of Car@CD and Ant@CD were used for RGB analysis with a smartphone via fabricated paper-based test kits. The fluorometric responses of Car@CD and Ant@CD demonstrated accurate and reliable detection due to their high selectivity and sensitivity to herbicides. Additionally, the colorimetric readout, accessible via a smartphone, enabled a convenient and portable assessment. This study introduces carbazole- and anthracene-modified carbon dots that integrate π-π stacking and hydrogen-bonding functionalities, thereby enhancing herbicide recognition. Considering that fluorescence-based detection of benfluralin and aclonifen is still almost absent and studies on pendimethalin remain scarce, the presented approach offers one of the first comprehensive strategies that combines high sensitivity, real-sample validation, and a portable smartphone-assisted paper kit for practical food safety monitoring
An effective Legendre wavelet technique for the time-fractional Fisher equation
No full textThis study modifies the time-fractional Fisher equation by adding a source term and generalizing the non-linear power to an arbitrary order. A numerical technique is proposed for the modified time-fractional Fisher equation using Legendre wavelets and the quasilinearization technique. The non-linear term is iteratively linearized using the quasilinearization technique. The convergence analysis and error estimates of the proposed method are studied. Several test problems are solved using the proposed technique, and numerical outcomes are contrasted with those obtained using some other approaches existing in the literature
Justifying violence and hostility through discourse: A critical discursive psychology analysis of anti-refugee hostility on social media during disasters
No full textDisasters create a fertile context for the scapegoating of minority groups, yet the discursive strategies used to legitimize this hostility remain understudied. This study addresses this gap by analysing how anti-refugee sentiment is discursively justified on social media during a crisis. We investigated how language on social media is used to legitimize hostility against refugees during disasters. Using a critical discursive psychology (CDP) approach, we analysed 345 posts on X that targeted Afghan refugees during the devastating July 2021 wildfires in Türkiye. Our analysis identifies three key discursive strategies that function to justify exclusion while avoiding charges of anti-refugee hostility and racism: (1) constructing refugees as a catastrophic threat akin to disasters; (2) circulating conspiracy theories that blame refugees for causing the crisis; and (3) delegitimizing refugees through categorization practices that question their moral worth and right to belong. Rather than relying on overtly racist language, these strategies draw on rational-seeming arguments about security, resource competition and cultural difference to build a warrant for exclusion. Our research expands the literature on the interplay between discourse and racism by demonstrating how racist verbal strategies are leveraged during disasters to legitimize hostility against refugees, thereby reinforcing social hierarchies and naturalizing exclusionary policies
