Middle East Technical University Research Information System
Not a member yet
104897 research outputs found
Sort by
People and Management Debt in ML-Integrated Software Projects: Structuring Industry Insights
No full textThe accelerated development of Machine Learning (ML) tools, combined with broader access to frameworks and infrastructures, has driven the rapid adoption of ML-based solutions in industry. However, their integration into software systems introduces unique challenges, particularly for managing technical debt (TD). Traditional TD research focuses primarily on technical issues, but in ML systems, people and management factors, referred to as nontechnical debt (NTD), play a critical role in TD accumulation and persistence. In this study, we investigate the underexplored dimension of NTD in ML-integrated software systems, focusing on people- and management-related factors. Using Design Science Research (DSR) methodology, we developed an artifact in an iterative incremental manner that categorizes NTD issues in ML systems. As part of this process, we conducted semi-structured interviews with 18 professionals from 15 companies, examining 22 ML projects. Through thematic analysis, we identified 15 NTD categories, 10 of which relate to people debt, and the remaining 5 to management debt. Each category is associated with underlying causes, short-term fixes, and potential solutions. Our findings show that NTD in ML projects frequently arise from inadequate decision-making practices, particularly those related to technology adoption, knowledge management, and human resource planning. Additional sources of NTD include challenges in team dynamics, such as insufficient collaboration, poor skill integration, and ineffective team structuring, as well as communication barriers rooted in organizational culture and team interactions. These factors collectively and substantially impact project outcomes. While band-aid solutions may provide short-term relief, they frequently contribute to accumulation over time. To support practitioners and researchers, our study complements the proposed artifact with actionable recommendations informed by expert perspectives and literature
A deep-learning framework for concealed and unconcealed face analysis in sub millimeter wave imaging
No full text
A Local Eddy Viscosity Parameterization for Estuarine Exchange Flow
No full textStructure and intensity of estuarine exchange flow depend significantly on the eddy viscosity Av profile which isdynamically linked to various forces (e.g., gravitational, tidal, wind-driven). The impact of winds on the exchange flowis complex due to its direct (local and remote changes in shear and density stratification) and indirect (modifications toAv profiles) contributions (see Basdurak et al., 2021). This study aims (i) to include wind entrainment effects in the tidallyaveraged Av parameterization (Fig. 1); (ii) to develop an analytical one-dimensional model for the wind driven exchangeflow by using this novel parameterization and assess the tidally averaged dynamics over a relevant physical parameter-space, subdomains of which have not yet been explored numerically. This one-dimensional model is based on a balancebetween frictional forces and pressure gradient, calibrated with a tidally-resolving one-dimensional water-column modelwith second-moment closure. Structure and intensity of the resulting exchange flow profiles are analyzed with respect tothree dimensionless parameters (the unsteadiness of boundary layer mixing Un, scaled-directional wind stress W, andhorizontal stratification Si). While down-estuarine winds enhance the gravitational circulation, up-estuarine winds resultin either a two-layer inverted circulation opposing the gravitational circulation, or a three-layer flow (favored by relativelystrong Si, weak W, and moderate Un) that is up-estuarine at the surface with classical two-layer circulation underneath.Relative thicknesses of surface and bottom boundary layers affect both the intensity and the inflection depth of theexchange flow layers. Up-estuarine winds with W ≳0.5 yield unstable stratification and reduce the exchange flow intensitywith increasing W (see Basdurak et al., 2023).</p
CONSTRUCTION MATERIALS AND TECHNOLOGIES IN THE FIRST NATIONAL ARCHITECTURE MOVEMENT: BRIDGING TRADITION AND MODERNITY BİRİNCİ ULUSAL MİMARLIK AKIMINDA YAPI MALZEMELERİ VE YAPIM TEKNOLOJİLERİ: GELENEĞİ VE MODERNLİĞİ BULUŞTURMAK
No full textThis article explores the construction materials and technologies employed during the First National Architecture Movement, marked by rapid industrialization and socio-political change in the late 19th and early 20th centuries. The movement, rooted in Ottoman Westernization and the rise of Turkish nationalism, blended traditional architectural elements with modern innovations brought about by the Industrial Revolution. Therefore, buildings constructed during this period ought to be preserved in their original characteristics, and according to international conservation principles. Within this research, literature and archival investigation have been conducted to present a comprehensive overview of the period’s technologies and materials. In addition, limited site visits and observations were carried out on some prominent buildings in Ankara and Istanbul, representing different periods and designed by various architects. The study demonstrates that modern materials such as Portland cement, cement-based materials, concrete, and steel, as well as technologies like ferroconcrete and reinforced concrete frame, were used alongside traditional ones such as masonry and timber framing. Moreover, these new materials and technologies were integrated into established architectural practices, such as masonry structures with reinforced concrete floor slabs or steel beams. The research also highlights discrepancies and contradictions in the existing literature regarding the identification of building materials and structural systems, revealing the need for more meticulous and accurate representation in future studies. The article concludes that buildings from this period must be analyzed and evaluated through the lens of the architectural technologies of their time, ensuring their preservation in accordance with international conservation principles. Proper recognition and identification of these construction materials and technologies will facilitate the rightful conservation of these buildings, which form an essential part of Türkiye’s architectural heritage
A Comparative Study of Clustering Algorithms on Power Consumption Data: The Role of Cyclic Feature Extraction
No full textPower consumption data, as an increasingly important type of data in the digital era, plays a key role in power forecasting and consumption clustering analysis. In this study, we hypothesize that meaningful features can be extracted from the cyclic nature inherent in energy consumption data. The ability to interpret and classify this data from new perspectives provides valuable contributions to the field of data science. To test this hypothesis, we examine clustering performance using hybrid methodologies, focusing on extracting non-cyclic features from various data types and cyclic features from wavelet-transformed data. By conducting separate analyses for each, we assess the outputs of two well-known algorithms, namely, OPTICS and DBSCAN, both of which depend heavily on precise parameter selection and optimization for accurate results. After obtaining the results, we observe a clear and decisive outcome in OPTICS clustering. On the other hand, in the DBSCAN clustering algorithm, effective clustering is observed regardless of whether the data is non-cyclic or cyclic
Enhancing teamwork during mobile game development: Potentials for the design of collaboration tools
No full textFunctional Bioimaging Probes: Fluorescent Conjugated Polymer Nanoparticles Coupled with Pleural Fluid-Derived Peptides and Proteins
No full textThiol-functionalized conjugated polymers offer a versatile platform for designing fluorescent nanomaterials with biomedical relevance. In this study, a thiol modified conjugated polymer composed of benzoxadiazole (BO) and carbazole (POxC-SH) was synthesized, then converted into fluorescent nanoparticles (POxC-SH NPs) via a reprecipitation method. The nanoparticles exhibited strong photoluminescence, colloidal stability, and monodispersity in media. Surface thiol groups enabled conjugation with peptide and protein components isolated from the pleural fluid of lung adenocarcinoma patients using SMCC cross-linking. The resulting bioconjugated nanoprobe was characterized by spectroscopic methods, FTIR, XPS, and Mass spectrometry. Cellular studies in A549 and BEAS-2B cell lines demonstrated efficient internalization and low toxicity of both native and conjugated nanoparticles. This work presents a proof of concept for using thiol-modified conjugated polymer nanoparticles as intrinsically fluorescent, patient-adaptable imaging agents, bridging conjugated polymer chemistry with targeted biomedical applications
Enhancing culturally responsive teaching in Turkish EFL education: insights and policy recommendations
No full textThis multiple cross-case study examined the identity development of English Language Teaching (ELT) preservice teachers in three ELT programs as cases. The study based its theoretical frame on Bourdieu’stheory and thinking toolkits in examining ELT identity development. The participants were pre-serviceteachers (n = 30) and teacher educators (n = 24) affiliated with three Education Faculties. Data relied onindividual and focus group interviews and field visits to understand how the programs operate. Thick dataanalyzed yielded three major themes: *ELT as a Field, *Teacher Educators as Agents, and *Pre-serviceTeachers as Agents. The study claimed a one-size policy approach is not effective for teacher educationand identity development. Institutional habitus is a crucial factor in this process</p