WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection

Permanent URI for this collectionhttps://hdl.handle.net/20.500.11779/256

Browse

Now showing 1 - 20 of 749

Weak Label Supervision for Monaural Source Separation Using Non-Negative Denoising Variational Autoencoders
(IEEE, 2019) Karamatli, Ertug; Kirbiz, Serap; Cemgil, Ali Taylan
Deep learning models are very effective in source separation when there are large amounts of labeled data available. However it is not always possible to have carefully labeled datasets. In this paper, we propose a weak supervision method that only uses class information rather than source signals for learning to separate short utterance mixtures. We associate a variational autoencoder (VAE) with each class within a non-negative model. We demonstrate that deep convolutional VAEs provide a prior model to identify complex signals in a sound mixture without having access to any source signal. We show that the separation results are on par with source signal supervision.
Turkish Version of the Multidimensional Measure of Emotional Abuse: Preliminary Psychometrics in College Students
(Springer Publishing Co, 2018) Sumer, Zeynep Hatipoglu; Murphy, Christopher M.; Demirtas, Ezgi Toplu
The aim of the current study was to investigate the basic psychometrics of the Multidimensional Measure of Emotional Abuse (MMEA; Murphy & Hoover, 1999) in a Turkish sample. Two hundred and fifty-four college students participated and completed the Turkish version of the MMEA (MMEA-TR) along with the Physical Assault of Conflict Tactics Scale-Revised, Experiences in Close Relationships Inventory, Relationship Assessment Scale, and Social Desirability Questionnaire. Confirmatory factor analysis supported the four-factor structure of the MMEA-TR for both victimization and perpetration reports. This factor structure was cross-validated with an independent older sample of 328 dating college students for perpetration reports. Satisfactory criterion validity and internal consistency reliability results were obtained as well. Based on the preliminary investigation, the MMEA-TR appears to be a psychometrically sound measure of psychological dating aggression perpetration and victimization among college students in Turkey. The results, limitations, and recommendations for future studies were discussed.
Time - Cost Relationships for Superstructure Projects in Turkey
(Turkish Chamber Civil Engineers, 2020) Sonmez, Murat; Akbiyikli, Rifat; Dikmen, S. Umit
The concept of time-cost relationship in construction projects was first introduced by Bromilow, using the data of the 328 superstructure projects completed in Australia. The aim of this study is to determine the time-cost relationship of superstructure projects in Turkey. Time and cost data of 460 superstructure projects completed between the years of 1999-2018 was used in the study. Data was grouped primarily on the basis of the intended use of buildings (individual buildings, educational buildings, hospitals, industrial buildings and social housing), and then time-cost relationships were separately determined for each group through statistical analysis. In addition, the effects of the parameters such as exchange rates and the number of non-working days on the time-cost relationship were investigated statistically, as well. As a result of this study, highly meaningful time-cost relationships (R-2 =0,60) are determined for the construction projects in Turkey, excluding disaster relief projects, which is not part of this study. It has also been determined that while the hospital projects in Turkey have the highest coefficient of determination (R-2=0,87), social housing projects have the lowest (R-2=0,54).
The Role of Stakeholder Engagement in E-Health and the Use of Big Data to Predict Health Outcomes
(Elsevier Science Inc, 2017) Cinaroglu, S.; Baser, O.
Recognizing Non-Manual Signs in Turkish Sign Language
(IEEE, 2019) Akarun, Lale; Aktas, Mujde; Gokberk, Berk
Recognition of non-manual components in sign language has been a neglected topic, partly due to the absence of annotated non-manual sign datasets. We have collected a dataset of videos with non-manual signs, displaying facial expressions and head movements and prepared frame-level annotations. In this paper, we present the Turkish Sign Language (TSL) non manual signs dataset and provide a baseline system for non manual sign recognition. A deep learning based recognition system is proposed, in which the pre-trained ResNet Convolutional Neural Network (CNN) is employed to recognize question, negation side to side and negation up-down, affirmation and pain movements and expressions. Our subject independent method achieves 78.49% overall frame-level accuracy on 483 TSL videos performed by six subjects, who are native TSL signers. Prediction results of consecutive frames are filtered for analyzing the qualitative results.
Robust HMM-Based Remaining Useful Life Estimation Using a Ridge-Regularized EM Algorithm
(MDPI, 2026) Kucukdag, Halime Beyza; Kirkil, Gokhan; Hekimoglu, Mustafa
Estimating the remaining useful life (RUL) of engineering systems is crucial for maintenance planning and the reliability of complex mechanical units. Accurate RUL predictions support timely interventions and help to prevent unexpected failures. This study proposes a statistically robust framework that models degradation signals up to the end of life using a hidden Markov model (HMM) with a simple-failure structure and an absorbing terminal state. The proposed method estimates state-dependent linear emission parameters and transition probabilities using a ridge-regularized expectation-maximization (EM) algorithm. The ridge penalty stabilizes slope estimates under limited data, while a robust Huber-based scale estimator reduces sensitivity to outliers in the sensor-derived health indicator. RUL is computed as a weighted expected time to absorption, combining transient-state survival characteristics with smoothed posterior-state probabilities obtained via the forward-backward algorithm. This yields a low-variance state-aware estimator that preserves the probabilistic structure of the HMM. Simulation studies show that the proposed ridge-regularized EM significantly reduces parameter variance and improves predictive accuracy compared with the baseline weighted least squares EM (WLS-EM). A real-data case analysis demonstrates further improvements in RUL estimation accuracy and smoother, more reliable prediction trajectories. Overall, the framework provides a robust and interpretable approach for practical prognostics applications.
Model Selection for Relational Data Factorization
(IEEE, 2019) Cemgil, Taylan; Kirbiz, Serap; Hizli, Caglar
As a fundamental problem in relational data analysis, model selection for relational data factorization is still an open problem. In our work, we propose to estimate model order for mixed membership blockmodels (MMSB) within the generic allocation framework of Bayesian allocation model (BAM). We describe how relational data is represented as Poisson counts of the allocation model, and demonstrate our results both on synthetic and real-world data sets. We believe that the generic allocation perspective promises a generalized model selection solution where we do not only select the model order, but also choose the most appropriate factorization.
Legal and Institutional Foundations of Turkey’s Domestic and Transboundary Water Policy
(Springer International Publishing AG, 2020) Kibaroglu, Aysegul
Turkey's water policy and management is a culmination of various laws and regulations governed by a range of national ministries and executive administrations. Over time, several changes were made in the existing legislation and institutions, which ended up with complex water management system in Turkey. Existing surface and groundwater laws have become insufficient in responding to the increasing water demand and diminishing water supply. On the other hand, neoliberal transformation of Turkish economy in the 1980s and the country's harmonization process with the European Union since the early 2000s have produced new primary and secondary water legislations in the domestic water, irrigation, hydropower and the environment sectors. In this context, this chapter, firstly, describes the principal water legislation in Turkey. Secondly, main water institutions are depicted with specific attention to the reorganization processes of various key ministries due to domestic and regional political changes. Finally, Turkey's transboundary water policy is delineated with its basic principles and prevailing practices.
Integrating Genre-Based Writing and Critical Thinking in Developing Writing Skills of Preservice Language Teachers
(Multilingual Matters Ltd, 2024) Altinmakas, Derya; Aptoula, Nur Yigitoglu
Kernel Density Estimation for Optimal Detection in All-Bit-Line MLC Flash Memories
(IEEE, 2019) Ashrafi, Reza A.; Pusane, Ali E.; Arslan, Suayb S.
NAND flash memories have recently become the main component of large-scale non-volatile storage systems. Recent studies have shown that various error sources degrade the Multi-level cell (MLC) memory performance, including inter-cell interference, retention error, and random telegraph noise. Accurate integration of these error sources into the analytical model to optimally derive the governing probability distributions and consequently the detection thresholds to minimize error rates lie at the heart of MLC research. Utilizing static derivations will not address the detection problem, as aforementioned error sources exhibit a strong non-stationary behavior. In this paper, a novel low-complexity implementation of a non-parametric learning mechanism, kernel density estimation, shall be used to periodically estimate the underlying probability distributions and hence approximate the optimal detection performance for time-varying all-bit-line MLC flash channel.
Examining Health Care Utilization and Costs among Atherosclerosis Patients in the Us Veteran Health Administration Population
(Elsevier Science Inc, 2017) Kariburyo, M. T.; Xu, J.; Baser, O.; Xie, L.; Zhang, Q.
Fault-Tolerant Strassen-Like Matrix Multiplication
(IEEE, 2020) Oblokulov, Muhtasham; Arslan, Suayb S.; Guney, Osman B.
In this study, we propose a simple method for fault-tolerant Strassen-like matrix multiplications. The proposed method is based on using two distinct Strassen-like algorithms instead of replicating a given one. We have realized that using two different algorithms, new check relations arise resulting in more local computations. These local computations are found using computer aided search. To improve performance, special parity (extra) sub-matrix multiplications (PSMMs) are generated (two of them) at the expense of increasing communication/computation cost of the system. Our preliminary results demonstrate that the proposed method outperforms a Strassen-like algorithm with two copies and secures a very close performance to three copy version using only 2 PSMMs, reducing the total number of compute nodes by around 24% i.e., from 21 to 16.
Exploring Environmental Justice: Meaningful Participation and Turkey’s Small-Scale Hydroelectricity Power Plants Practices
(Springer International Publishing AG, 2020) Sayan, R. Caner; Kibaroglu, Aysegul
This chapter explores the emerging concept of meaningful participation within the framework of environmental justice, with specific reference to Turkey's recent experience of building several small-scale hydroelectricity power plants (HEPP). The paper scrutinizes the HEPP process, including its entrenched legal framework, and attempts to come up with suggestions to elaborate further on the concept of meaningful participation.
Exact Construction of BS-Assisted MSCR Codes With Link Constraints
(IEEE-Inst Electrical Electronics Engineers Inc, 2022) Arslan, Suayb S.
It is clear that 5G network resources would be consumed by heavy data traffic owing to increased mobility, slicing, and layered/distributed storage system architecture. The problem is elevated when multiple node failures are repaired to address service quality requirements. Typical approaches include individual or cooperative data regeneration to efficiently utilize the available bandwidth. It is observed that storage systems of 5G and beyond technologies shall have a multi-layer architecture in which base stations (BS) would be present. Moreover, communication with each layer would be subject to various communication costs and link constraints. Under limited BS assistance and cooperation, the trade-off between storage per node and communication bandwidth has been established. In this trade-off, two operating points, namely minimum storage, and bandwidth regeneration are particularly important. In this study, we first identify the optimal number of BS use at the minimum storage regeneration point. An explicit code construction is provided subsequently for the exact minimum storage regeneration whereby each layer may help the repair process subject to a communication link constraint.
Data Repair in BS-Assisted Distributed Data Caching
(IEEE, 2020) Kaya, Erdi; Haytaoglu, Elif; Arstan, Suayb S.
In this paper, centralized and independent repair approaches based on device-to-device communication for the repair of the lost nodes have been investigated in a cellular network where distributed caching is applied whose fault tolerance is provided by erasure codes. The caching mechanisms based on Reed-Solomon codes and minimum bandwidth regenerating codes are adopted. The proposed approaches are analyzed in a simulation environment in terms of base station utilization load during the repair process. Based on the intuitive assumption that the base station is usually more costly than device-to-device communication, the centralized repair approach demonstrates a better performance than the independent repair approaches on the number of symbols retrieved from the base station. On the other hand, the centralized approach has not achieved a dramatic reduction in the number of symbols downloaded from the other devices.
Distributed Matrix Multiplication with MDS Array BP-XOR Codes for Scaling Clusters
(IEEE, 2019) Arslan, Suayb S.
This study presents a novel coded computation technique for distributed matrix-matrix product computation at a massive scale that outperforms well known previous strategies in terms of total execution time. Our method achieves this performance by distributing the encoding operation over the cluster (slave) nodes at the expense of increased master-slave communication. The product computation is performed using MDS array Belief Propagation (BP)-decodable codes based on pure XOR operations. In addition, our scheme is configurable and suited for modern compute node architectures equipped with multiple processing units organized in a hierarchical manner. Assuming the number of backup nodes being sublinear in the size of the product, we shall demonstrate that the proposed scheme achieves order-optimal computation from an end-to-end latency perspective while ensuring acceptable communication requirements that can be addressed by today's high speed network link infrastructures.
Cost of Guessing: Applications to Data Repair
(IEEE, 2020) Arslan, Suayb S.; Haytaoglu, Elif
In this paper, we introduce the notion of cost of guessing and provide an optimal strategy for guessing a random variable taking values on a finite set whereby each choice may be associated with a positive finite cost value. Moreover, we drive asymptotically tight upper and lower bounds on the moments of cost of guessing problem. Similar to previous studies on the standard guesswork, established bounds on moments quantify the accumulated cost of guesses required for correctly identifying the unknown choice and are expressed in terms of the Renyi's entropy. A new random variable is introduced to bridge between cost of guessing and the standard guesswork and establish the guessing cost exponent on the moments of the optimal guessing. Furthermore, these bounds are shown to serve quite useful for finding repair latency cost for distributed data storage in which sparse graph codes may be utilized.
Audio Source Separation Using Variational Autoencoders and Weak Class Supervision
(IEEE-Inst Electrical Electronics Engineers Inc, 2019) Karamatli, Ertug; Kirbiz, Serap; Cemgil, Ali Taylan
In this letter, we propose a source separation method that is trained by observing the mixtures and the class labels of the sources present in the mixture without any access to isolated sources. Since our method does not require source class labels for every time-frequency bin but only a single label for each source constituting the mixture signal, we call this scenario as weak class supervision. We associate a variational autoencoder (VAE) with each source class within a non negative (compositional) model. Each VAE provides a prior model to identify the signal from its associated class in a sound mixture. After training the model on mixtures, we obtain a generative model for each source class and demonstrate our method on one-second mixtures of utterances of digits from 0 to 9. We show that the separation performance obtained by source class supervision is as good as the performance obtained by source signal supervision.
Art Museums and the Middle East: A Contested Territory
(Intellect Ltd, 2020) Yucel, Sebnem
Array BP-XOR Codes for Hierarchically Distributed Matrix Multiplication
(IEEE-Inst Electrical Electronics Engineers Inc, 2022) Arslan, Suayb S.
A novel fault-tolerant computation technique based on array Belief Propagation (BP)-decodable XOR (BP-XOR) codes is proposed for distributed matrix-matrix multiplication. The proposed scheme is shown to be configurable and suited for modern hierarchical compute architectures such as Graphical Processing Units (GPUs) equipped with multiple nodes, whereby each has many small independent processing units with increased core-to-core communications. The proposed scheme is shown to outperform a few of the well-known earlier strategies in terms of total end-to-end execution time while in presence of slow nodes, called stragglers. This performance advantage is due to the careful design of array codes which distributes the encoding operation over the cluster (slave) nodes at the expense of increased master-slave communication. An interesting trade-off between end-to-end latency and total communication cost is precisely described. In addition, to be able to address an identified problem of scaling stragglers, an asymptotic version of array BP-XOR codes based on projection geometry is proposed at the expense of some computation overhead. A thorough latency analysis is conducted for all schemes to demonstrate that the proposed scheme achieves order-optimal computation in both the sublinear as well as the linear regimes in the size of the computed product from an end-to-end delay perspective.

Browse

Recent Submissions