Middle East Technical University Research Information System
Not a member yet
104897 research outputs found
Sort by
Extending Persistent Homology Through Zigzag Modules
No full text Zigzag persistence is a variant of the ordinary persistence in topological data analysis, which can be used to extract topological features from datasets. It was introduced by Carlsson and Silva [1] as a generalization of the original persistence algorithm for sequence of topological spaces, that are not related by only an increasing filtration, but also have inclusions and exclusions. The aim of this thesis is to present a survey of zigzag persistence by presenting its mathematical foundations and elaborating the key tools and algorithms used for its computation. Subject Keywords Topological Data Analysis, Zigzag Module, Zigzag Persistence, The Diamond Principle Zigzag persistence is a variant of the ordinary persistence in topological data analysis, which can be used to extract topological features from datasets. It was introduced by Carlsson and Silva [1] as a generalization of the original persistence algorithm for sequence of topological spaces, that are not related by only an increasing filtration, but also have inclusions and exclusions. The aim of this thesis is to present a survey of zigzag persistence by presenting its mathematical foundations and elaborating the key tools and algorithms used for its computation. Subject Keywords Topological Data Analysis, Zigzag Module, Zigzag Persistence, The Diamond Principle </p
An exploration of practice-based professional development programs and the role of the facilitator: Changes in primary school teachers' perceived knowledge and attitudes towards the teaching of fundamental motor skills
No full textThis study investigated changes in the perceived knowledge and attitudes of primary school teachers towards the teaching of fundamental movement skills (FMS) and to understand how a facilitator affects the knowledge and attitudes of teachers through a practice-based professional development program (PBPD). Teachers participated in a 10-week PBPD program. Data were collected by the facilitator through a series of semi-structured interviews and analyzed using a reflexive thematic approach. Teachers indicated that they gained knowledge about FMS, Physical Education (PE) curriculum literacy, and material literacy. In conclusion, the PBPD program itself and the facilitator had a positive effect on teachers' perceived knowledge and attitudes for teaching FMS
Klasik ve Makine Öğrenmesi Modellerinin OASIS-2 Verisine Uygulanması
No full textAbstractBackgroundAlzheimer’s disease (AD) is a significant global health issue that affects both individuals and society for older adults. Early detection is essential in Alzheimer’s disease, as there is no definite treatment, and prevention is possible through early detection. In the studies of Alzheimer's disease, different approaches are used to understand the characteristics of the disease, including imaging methods such as MRI scans, biomarkers, and quantitative data from MRI scans. With these methods, the cognitive status of individuals could be determined. It is essential to detect the features affecting Alzheimer’s disease and detect the disease early and automate the detection and prediction by using modeling tools and by examining the datasets.MethodThe symptoms of the disease can be observed over time, making the structure of the study longitudinal. Classical statistical models and machine learning algorithms can be used to analyze these datasets. The OASIS – 2 dataset is employed to find the features affecting dementia status and compare models' performances. The classical mixed models, their extended versions, and hybrid models, Boruta, GEE, GLMM, HGLM, GLMMLasso, GPBoost, GLMMTree, and HRF, are used to analyze the longitudinal dataset with a small sample size and binary outcome. Model performances are measured using accuracy, F1 score, sensitivity, and specificity, and the results are interpreted using the odds ratio.ResultGPBoost learns and classifies the dementia status well but overfits due to the small sample size in the dataset, and tree-based algorithms are efficient in predicting the dementia status when a new subject enters the study.ConclusionThe common significant variables to address the dementia status of individuals are MMSE, nWBV, education, gender, and age. The results also indicate that the increase in MMSE score means the patient is less likely to be demented; the increase in normalized whole brain volume indicates a less probability to be demented; increase in socioeconomic status results in the patient having less probability to be demented; males are more likely to be demented; age has a negative impact on being demented; the increase in education years results in the patient having less probability to be demented.</div
Assessment of Sustainable Water Management Strategies in Porsuk Watershed UsingHydrological WEAP Model
No full textDevelopment of a simulation-integrated mine road quality index (MRQI) for efficient dispatching operations in surface mines
No full textHaulroads are critical to the safety, efficiency, and productivity of surfacemining operations. The deterioration of haul road functionality, whetherpartial or complete, can result in significant challenges such as productiondelays, increased wear on machinery, and induced safety risks, all of which maydisrupt or even halt dispatching activities. In many cases, the construction ofmine roads relies on experience-based practices rather than standardisedmethodologies. The absence of systematic and standardised approaches in haulroad construction contributes to the emergence of various operational problems,as the frequency and severity of uncertainties increase. These issues oftenappear as potholes, slip cracks, and uneven surfaces on the road.Thisstudy aims to address these challenges by developing a standardised framework,the Mine Road Quality Index (MRQI), designed to assess and improve the qualityof mine roads. The methodology integrates fuzzy logic, fault tree analysis, anddiscrete event simulation to analyse and prioritise the uncertainty factorsthat influence road quality and performance. To achieve this, a Fuzzy FaultTree Analysis (FFTA) integrated with expert panel is conducted first toevaluate 17 key uncertainty factors in five main categories: structural designuncertainties, functional design uncertainties, management uncertainties,geometric design uncertainties, and geological uncertainties. These factors areassessed based on their alignment with both planned and actual road conditions,as well as their frequency and severity in contributing to road failures andimpacting production rates.ConstructedFFTA allows the development of a Discrete Event Simulation (DES) model for amore in-depth evaluation of uncertainties affecting road quality and forexploring road improvement strategies by integrating the MRQI into a dispatchalgorithm. This integrated approach offers a systematic approach to simulateand predict road performance across various scenarios, enabling improvedoperational outcomes.</p
THE IMPACT OF ENGINEERING DESIGN INSTRUCTION ON PRE-SERVICE SCIENCE TEACHERS' NATURE OF ENGINEERING VIEWS AND THEIR LESSON PLANNING
No full text“It’s Like You Don’t Have an Identity”: A Narrative Inquiry into a Precarious English Language Teacher’s Professional Identity Negotiations
No full textUnder prevailing neoliberal policies and job market conditions, language teachers often face job precarity, characterized by flexible, insecure, and low-wage conditions. Despite this trend, the research on language teacher identity in relation to job precarity is limited. This study employs critical perspectives to explore the impact of precarious job conditions on the professional identity negotiation of an English language teacher, Pelin, through narrative inquiry conducted in Türkiye. Confronted with the realities of the job market, such as job insecurity and underpayment, Pelin’s negotiation of language teacher identities was negatively influenced. Her identity negotiation process was particularly intertwined with frustration, uncertainty, temporariness, and otherization. While navigating neoliberal work conditions and ethics, Pelin also occasionally found herself aligning with some neoliberal discourses, primarily in pursuit of economic and professional advancement. Based on these findings, we made several suggestions for key stakeholders such as administrators, policymakers, teacher educators, and student teachers
Analysis of the size effect in the nanoscratching process of single crystalline and polycrystalline materials
No full textIn tribology, nanoscratch testing is used to evaluate the mechanical and tribological properties of materials by employing a nanoscale indenter. At this scale, size-dependent plasticity effects, particularly those related to scratch hardness, become significant. Additionally, the anisotropic nature of crystalline materials means that the crystal orientation of the samples plays a crucial role. This study focuses on investigating the effects of indentation, grain size, and crystal orientation by applying the finite element method (FEM) with a lower-order strain gradient crystal plasticity framework. Results for scratch hardness, friction coefficient, and pile-up patterns are analyzed to identify the experimental trends, providing insights into how size and orientation effects influence material behavior in nanoscratch testing